Competences in Organizational E-Learning

i
Com pe t e nc ie s in Orga niza t iona l E-Le a r ning: Conc e pt s a nd Tools

Miguel-Angel Sicilia, University of Alcal, Spain
I nfor m at ion Sc ie nc e Publishing

Hershey London Melbourne Singapore
ii Acquisition Editor: Senior Managing Editor: Managing Editor: Assistant Managing Editor: Development Editor: Copy Editor: Typesetter: Cover Design: Printed at: Kristin Klinger Jennifer Neidig Sara Reed Sharon Berger Kristin Roth Toni Fitzgerald Jamie Snavely Lisa Tosheff Integrated Book Technology
Published in the United States of America by Information Science Publishing (an imprint of Idea Group Inc.) 701 E. Chocolate Avenue Hershey PA 17033 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: cust@idea-group.com Web site: http://www.idea-group.com and in the United Kingdom by Information Science Publishing (an imprint of Idea Group Inc.) 3 Henrietta Street Covent Garden London WC2E 8LU Tel: 44 20 7240 0856 Fax: 44 20 7379 3313 Web site: http://www.eurospan.co.uk Copyright 2007 by Idea Group Inc. All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher. Product or company names used in this book are for identiication purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI of the trademark or registered trademark. Library of Congress Cataloging-in-Publication Data Competencies in organizational e-learning : concepts and tools / Miguel-Angel Sicilia, editor. p. cm. Summary: This book addresses competencies as the key observable workplace behavior that is able to drive learning and knowledge dissemination processes inside organizations, and can be used as a guide for implementing or improving competency-based approaches to e-learning--Provided by publisher. Includes bibliographical references and index. ISBN 1-59904-343-2 (hardcover) -- ISBN 1-59904-344-0 (softcover) -- ISBN 1-59904-345-9 (ebook) 1. Organizational learning. 2. Knowledge management. 3. Communication in organizations. I. Sicilia, Miguel-Angel, 1973HD58.82.C653 2007 658.4012--dc22 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the authors, but not necessarily of the publisher.
iii
Com pe t e nc ie s in Orga niza t iona l E-Le a r ning: Conc e pt s a nd Tools

Ta ble of Cont e nt s
Preface ...........................................................................................................................vi
Section I: General Concepts and Management Issues Chapter I Competencies and Organizational Learning: A Conceptual Framework ............... 1 Miguel-Angel Sicilia, University of Alcal, Spain Ambjrn Naeve, Royal Institute of Technology, Sweden Chapter II Competency Management Systems and Technologies ............................................. 10 Shantha Liyanage, University of Auckland, New Zealand Chapter III A Survey of Competency Management Software Information Systems in the Framework of Human Resources Management ............................................ 41 Alfonso Urquiza, Francisco de Vitoria University, Spain
iv
Chapter IV Competency Management using the Competence Performance Approach: Modeling, Assessment, Validation, and Use .............................................................. 83 Tobias Ley, Know-Center, Austria Dietrich Albert, University of Graz, Austria Stefanie Lindstaedt, Know-Center, Austria Chapter V Competencies Management: Applying Technologies for Strategic Management .............................................................................................................. 120 Miltiadis Lytras, University of Patras, Greece Maria Mantziou, Open University of Greece, Greece Maria Pontikaki, Intralot S.A., Greece
Section II: Competencies from the Viewpoint of E-Learning and Organizational Memory Chapter VI An Organizational Memory Tool for E-Learning .................................................. 146 Marie-Hlne Abel, Laboratory Heudiasyc UMR CNRS 6599, University of Compigne, France Chapter VII An Ontological Representation of Competencies as Codiied Knowledge........... 169 Salvador Snchez-Alonso, Unversity of Alcal, Spain Dirk Frosch-Wilke, University of Applied Sciences, Germany Chapter Chapter VIII Ontology-Based Approach to Formalization of Competencies ............................. 185 April Ng, Simon Fraser University Surrey, Canada Marek Hatala, Simon Fraser University Surrey, Canada Chapter IX Linking E-Government with Organizational Memory through Individual Competencies ............................................................................................................. 207 Juan G. Cegarra-Navarro, Universidad Politcnica de Cartagena, Spain Chapter X Designing and Implementing Views on Competencies .......................................... 221 Marcel van der Klink, Open University of The Netherlands, The Netherlands Kathleen Schlusmans, Open University of The Netherlands, The Netherlands Jo Boon, Open University of The Netherlands, The Netherlands Chapter XI Ten-Competence: Life-Long Competence Development and Learning............... 234 Rob Koper, Open University of The Netherlands, The Netherlands Marcus Specht, Open University of The Netherlands, The Netherlands
Section III: Competencies and the Semantic Web Chapter XII An Ontology-Based Competence Management Model to Support Collaborative Working and Organisational Learning .......................................... 253 Jos Braga de Vasconcelos, Universidade Fernando Pessoa, Portugal Chris Kimble, University of York, UK Chapter XIII Integrated Semantic-Based Composition of Skills and Learning Needs in Knowledge-Intensive Organizations ....................................................................... 266 Simona Colucci, SisInfLab, Politecnico di Bari, Italy Tommaso Di Noia, SisInfLab, Politecnico di Bari, Italy Eugenio Di Sciascio, SisInfLab, Politecnico di Bari, Italy Francesco Maria Donini, Universit della Tuscia, Italy Azzurra Ragone, SisInfLab, Politecnico di Bari, Italy Chapter XIV Semantic Web Technologies in the Recruitment Domain...................................... 299 Ralf Heese, Humboldt-Universitt zu Berlin, Germany Malgorzata Mochol, Freie Universitt Berlin, Germany Radoslaw Oldakowski, Freie Universitt Berlin, Germany Chapter XV Ontology-Based Competency Management for Corporate E-Learning .............. 319 Fotis Draganidis, University of Athens, Greece Gregoris Mentzas, University of Athens, Greece Chapter XVI Applying Semantic Web in Competence Management ......................................... 333 Mikko Laukkanen, TeliaSonera, Finland Heikki Helin, TeliaSonera, Finland About the Authors ....................................................................................................... 358 Index .......................................................................................................................... 367
vi
Pre fa c e
Companies should decide what processes and competencies they must excel at and specify measures for each. ~ David P. Norton The term competency (plural competencies) has become a widely used term in discussions and technical-scientiic literature that deals with organizational or workplace learning in the last years. This has raised the need for many professionals and scholars to have a full understanding of the meaning and implications of the concept. However, there is not a reference guide for people who want to have such overall understanding, and the monographs or books dealing with the subject do it from a single, speciic perspective, which in my view restricts the landscape to a single aspect. This book attempts to ill this gap in the subject of competencies by bringing together different author proiles and different interests and perspectives with regard to competencies. In a nutshell, competencies can be deined as observable capabilities of individuals for speciic work situations. Since they are observable, they are subject to measurement in some way. This provides a convenient approach to the measure of organizational capability in regards to human resources. However, there is not a single account of how competencies can be measured, since their deinition is complex in several aspects. These aspects include their deinition, relationships, similarities or complementarities, and the issues of how to derive aggregated competency scores or igures. This last aspect is fundamental in measuring the competencies for groups or teams, and also to measure the aggregated competency of a department or an organization. Aggregates are also required for the important issue of assessing the competency gap, in which the objective is to measure the actual capability of the organization in contrast to some organizational objectives or needs derived for some kind of statement, be it from management, market forecasts, or individual desires. These are the core concepts approached from different perspectives in this book. Each chapter in this book takes a different perspective in the above-mentioned and related issues. I hope to have collected a selection that covers the main issues and fosters further efforts in research and in the elaboration of reference material in the topics.
vii
Motivation
The concept of competency can be used as a key element in the organizational learning and the design of information systems that are driven by representations of competencies making a more eficient organizational learning processes. This statement is the underlying motivation for this book. However, for such a claim to become a reality, organizations require a clear understanding and strategy regarding competencies. Competencies as patterns of observable, measurable human behaviour in workplace situations can be used as a standard for measuring the capacity of employees in activities that create value of some kind. Thus, a irst important issue is clarifying how to deine and represent competencies and their constituents. Competency schema deinition is a legitimate research subject in itself. It comprises the representation in formal languages of competency descriptions and actual competencies, and the inquiry on the relationships possible between competencies. Knowledge, skills, and abilities are often considered ingredients required for competency in concrete work situations. A second, related issue is assessing, measuring, and computing with competencies. This includes both the procedures and methods for knowing the competency levels of employees and also the computing of aggregated competency levels for groups or organizational units. This is a topic that requires a lot of attention due to the diversity of computing approaches that can be devised to solve it. In addition, the assessing of competency gaps connects with the cycle of needs and behaviour of the organization. It formalizes in some way the learning needs of an organization or unit. Thus, learning activities can be planned to ill the competency gap, selecting for that the best target human resources, considering their previous competency record and the constraints of their calendar. The organizational information systems, if linked to a learning management system, could be arranged and scheduled learning processes and their results tracked, selecting the best-suited learning resources. Considering further the organizational issues, there is also a need to clarify the relationships of competency-driven learning with knowledge management (KM) activities, and particularly with organizational memories. A straightforward connection might be that of considering learning resources as particular knowledge assets, and the processes that manage them as comprising also learning activities. This allows for a competency approach in KM also. All the issues mentioned above are dealt with in this book, but the material provided does not exhaust the range of approaches possible to address each of them. However, the material included provides a solid ground for researchers and practitioners who in some way or another need to deal with competencies, learning technology, and/or knowledge management systems. If those needs are fulilled to some extent, the objectives of the book will have been achieved. The rest of this preface provides information on the process of editing this book and an overview of its organization and contents.
viii
The Process
The process of selection of papers combined an open call for chapters in popular newsgroups with targeted invitations to submit to speciic people who had experience or previous publications on the topic. All the chapters had undergone a peer review process to ensure both technical and scientiic quality and also to check the appropriateness for the objectives of the book. The review process proceeded in two phases: a irst one in which chapter proposals were requested and a second one in which full papers were required for those chapters that passed the irst ilter. The review of the proposals was aimed at having the right contents in the book, excluding chapters that were clearly out of the scope and objectives of the call for chapters. It included some brief comments on appropriateness when required, provided as a convenience for authors of full chapters to align the writing with the overall objectives of the book. The review of the full chapters was aimed at ensuring technical and scientiic quality, and it attempted to provide constructive suggestions for chapters that were clearly acceptable but could be improved to provide further value to the reader. After the peer review process, minor but important matters on formats and additional material were addressed. Now that the book is available, it is time for the last part of the process, that is, gathering feedback from readers. Receiving feedback would be for me very rewarding, even in the cases of critical comments, since it would allow me to redirect my future planned efforts in the area of competencies and e-learning. This is a sincere invitation for any reader interested in the topics of the books to send me any kind of suggestion or comment.
Overall View of the Book

The book has been organized in three parts, in an attempt to give some structure to the different viewpoints provided in each chapter. However, this organization is not intended as a sharp classiication but rather as an aid for guiding the reader to the parts that could be of most interest to her or him. It is my hope that chapter authors feel that their chapters have been put in the appropriate part, but many of them could have been classiied under more than one label, so that there is a degree of subjectivity in the inal choice that could perhaps be understood by some readers as a misclassiication. In any case, I suggest a quick reading of the abstracts of the chapters before starting with their contents, just to have a quick idea of the diversity of topics and orientations that will be found. The rest of this section provides a brief reference to the contents and main points of each chapter, intended as a roadmap for the reader. Section I. General Concepts and Management Issues Section I groups together papers that provide comprehensive accounts of the main concepts in the topics or the book, or that touch general issues. In both cases, these chapters provide an overall view of the many surrounding topics of the paradigm of competencies. This irst section of the book starts with a short contribution Ambjrn Naeve and myself. This contribution is intended only as a brief essay for roadmapping future efforts in the area. It built directly from the experience of the authors in modeling competencies for e-learning,
ix
and writing it was an useful exercise in the inception of our joint European project LUISA (Learning Content Management System Using Innovative Semantic Web Services Architecture), code FP62004IST4 027149, and other research initiatives. In the second chapter, Dr. Shantha Liyanage provides a comprehensive account of the competency approach and its relation to organizational learning, Web systems, and knowledge transfer. The chapter addresses why learning and the competency approach are important and how learning is a part of the organizational functioning. The chapter is especially interesting in that it explains clearly the main elements that are dealt with in the rest of the book, following a seamless low. I strongly recommend readers start with this chapter, since it provides a good start for thinking in competencies and learning from a holistic perspective. The third chapter turns attention to the software market of competency systems and also focuses on the view of competencies that is usual in the human Resource management perspective. Alfonso Urquiza provides a clear view on the state of the market of software tools that includes competency management, using a detailed functional description. This chapter can be read independently of the others and is especially targeted at practitioners and decision makers who are in the process of choosing a software platform for Human Resource Management that includes the competency approach in some way. Tobias Ley, Dietrich Albert, and Stefanie Lindstaedt provide in the fourth chapter a different contribution, which reports a substantial advancement in the practice of competency management in organizations. They start reviewing current competency approaches and the models available, along with the main future challenges for the ield. From that initial analysis, the authors devise a new approach with a strong theoretical background, and then provide relevant evaluation data. The resulting model is a relevant contribution to competencies as a management discipline, and the chapter illustrates a research methodology that can be useful for readers who are seeking ways of testing or creating tailored competency management models for their organizations or institutions. Chapter V takes a completely different approach in introducing concepts related to competencies. Miltiadis Lytras, Maria antziou, and Maria Pontikaki approach competencies through real-world scenarios, providing a fresh look at several managerial topics. The chapter provides a number of must haves for competency solutions from a management perspective. For readers who have never been involved in competencies, this chapter might be the best option to start, since it provides informal, real-life examples of the implications of competencies in management. Section II. Competencies from the Viewpoint of E-Learning and Organizational Memory The second section of the book includes chapters with an emphasis on e-learning or knowledge management elements as organizational memories. In Chapter VI, Marie-Hlne Abel describes the E-MEMORAe tool, which supports an organizational goal-driven approach based on the concept of learning organizational memory. It takes knowledge, know-how, and behaviours as the main elements of competencies, and connects them with the concept of organizational memories. The tool described takes a Semantic Web approach, since it uses ontologies and semantic annotation, and it also builds on the concept of learning object. This results in a state-of-the-art example on the combination of learning objects and Semantic Web for the purpose of organizational management of knowledge.
Chapter VII starts with a discussion of knowledge lifecycle models as the Knowledge Management Consortium International (KMCI). Then, Salvador Snchez-Alonso and Dirk Frosch-Wilke proceed to map the main concepts of competency management with the elements of the KMCI. This provides an understanding of the degree overlapping competency management has with the dynamics of knowledge management. Finally, the ontological representations that underlie these mappings are sketched. Chapter VIII by April Ng and Marek Hatala enters into a very speciic but critical aspect of the integration of learning technology with competency management: the linking of learning objects with competency representations. Ontologies of competencies are included in standardized learning object metadata records, thus enabling a uniform, interoperable way to retrieve learning objects based on the competency gap. This chapter is especially relevant to researchers and practitioners in standardized learning technology. In Chapter IX, Juan G. Cegarra-Navarro turns attention to the role of organizational memory in the adoption and use of e-government systems. He focuses on the competences that are required to manage and exploit knowledge at the organizational level. The chapter ends with an assessment of some important aspects that require more attention in the ield of e-government. Chapter X, by Marcel van der Klink, Kathleen Schlusmans, and Jo Boon, provides an insightful discussion on the concept of competency and proposes ive fundamental dimensions: speciicity, coherency, durability, activity, and trainability. These dimensions are of a great importance in clarifying the approach taken on competency in research projects and organizational approaches. Then, the chapter continues with two differentiated competency models: function-based and employee-based. The former focuses on capabilities for speciic functions, while the latter focus on the combined competencies of the individual. These different models lead to different competency management strategy, which makes this chapter especially interesting for devising organizational models for competencies. Chapter XI, by Rob Koper and Marcus Specht, describes the main issues being dealt with in the European Unions TENCompetence project. It starts by describing a number of major problems associated with achieving a platform that supports individuals, groups, and organizations in lifelong competence development using open-source, standards-based, sustainable, and innovative technology. Then it moves onto a description on how the TENCompetence approach provides the main directions that are required to tackle with those problems. Some of these chief issues include dealing with different levels of learning support and providing advanced mechanisms for learning networks. Section III. Competencies and the Semantic Web This third section of the book focuses on an advanced information technology topic. The Semantic Web vision described by Berners-Lee, Hendler, and Lassila (2001) represents an approach to enhancing the current World Wide Web with machine-understandable semantics. The essential idea is that Web resources as identiied by URIs can be described by metadata with the purpose of enabling automated processing. The precision of logics-based description would allow for the creation of tools that do not rely on natural language processing as current search tools, thus ideally removing the problems associated with ambiguity and implicitness in natural language. Metadata requires shared semantics in the context of an open Web, so that the proliferation of descriptions and vocabularies do not end up in a Tower of Babel. Ontologies provide the support for the shared representation of such semantics,
xi
and ontology mapping procedures could resolve the issues of semantic interoperability in case of disparate representations over the same domain. Ontologies act as descriptions of semantic domains that are used to express metadata records according to previously agreed deinitions. Ontologies provide an explicit, shared representation of a domain providing unambiguous deinitions for the main concepts and relations describing the phenomena under consideration (Gruber, 1995). Description logics (Baader et al., 2003) are well known and thoroughly studied knowledge representations that can be used to develop formal ontologies and systems that use them, enabling reasoning services of a diverse kind. Formal ontologies thus provide a foundation for representing the diverse aspects of virtual communities and the artefacts they deal with in terms of ontologies, which would eventually result in advanced tools that are aware of the structural and cultural issues that are part of a community. Ontologies can be expressed in Web markup languages for ease of processing and interoperability. Notably, the OWL language has reached the status of W3C recommendation, and several tools are yet available to edit or process OWL representations. Ontologies are a natural target as a representation language for competencies, since they provide advanced support for intelligent agents and are able to reuse existing knowledge representations in the form of domain ontologies. The book provides ive chapters that deal with Semantic Web issues from different perspectives. Chapter XII focuses on management issues and uses ontologies as the key structuring principle for the competency approach. More concretely, Jos Braga de Vasconcelos and Chris Kimble describe the functional architecture of a system based on ontologies that use competencies and expert annotations to drive collaborative work and learning. It is of special interest that the chapter describes a multi-level ontology architecture connecting with the main components of an organizational memory. Chapter XIII, by Simona Colucci, Tommaso Di Noia, Eugenio Di Sciascio, Francesco Maria Donini, and Azzurra Ragone, has a much different focus than the previous chapter, in that it describes in detail a computational model for searching among available skills and facilitating the acquisition of missing ones. Concretely, the proposed approach uses a description logic and reasoning services, providing the main formal deinitions. This chapter provides a concrete, fully speciied approach to competency gap analysis and targeting, and as such, it constitutes a novel and signiicant research result that can be used for contrasting other approaches or extending the one presented in some concrete ways. One speciic and important domain of application for competency approaches is that of recruitment. Chapter XIV by Ralf Heese, Malgorzata Mochol, and Radoslaw Oldakowski addresses precisely that area from a Semantic Web perspective. The application of Semantic Web technology is put in context by pointing out the requirement for semantic metadata to cope with the large and heterogeneous volume of job offerings in popular job portals and systems. A prototype implementation is described, providing a really interesting case of matching offerings and needs through competencies for competency as enabled by semantics. Chapter XV by Fotis Draganidis and Gregoris Mentzas describes the architecture, design, and deployment of a system that integrates ontologies with competency management and e-learning. It provides the architecture of such a system, describes the approach to specifying competencies, and inally proposes a method for adopting competencies under such a technological view. The chapter ends with a summary of lessons learned of special relevance to managers and decision makers.
xii
Finally, Chapter XVI, by Mikko Laukkanen and Heikki Helin, describes another approach to Semantic Web and competency management, in this case providing both the description of a concrete tool and the details of the algorithms used for matching competencies and needs. A description on the use of the tool illustrates the advantages of the ontology approach to competencies. A very important aspect of this last chapter is that it provides both the technical material and the organizational context in which the requirements were originated, thus becoming a complete technology development case.
Outlook
As evidenced in the brief summary of the chapters, there are a number of recurrent problems in the application of competency management to learning in organizations, but the solutions, perspectives, and areas of research range from pure management issues to technical knowledge representation using logic languages. The chapters in this book have pointed out the main questions opened in competency-based approaches as a research (sub-)domain. The effort in editing this book is a irst step toward a clariication of the different viewpoints on competencies as applied in organizational settings. I irmly believe that competencies will be the focus of a great deal of research attention in the following years, especially in the domains of learning technology and knowledge management. The natural follower of this book would be a monograph providing an in-depth and comprehensive account of competencies deinition, assessment, and management that combines both a managerial and a technical view. That would serve as a blueprint for the adoption of a rich competency management approach in organizations, built on best practice and powerful enough to adapt to the different requirements of various industrial sectors and types of enterprises. Dr. Miguel-Angel Sicilia Alcal de Henares July 2006
References
Baader, F., et al (Eds.). (2003). The description logic handbook. Theory, implementation and applications. Cambridge. Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The Semantic Web. Scientiic American, 284(5), 34-43. Gruber, T. R. (1995). Toward principles for the design of ontologies used for knowledge sharing. International Journal of Human and Computer Studies, 43(5/6), 907.
xiii
Ac k now le dgm e nt s
This edition of this book was, in fact, the original idea of Dr. Miltiadis Lytras, and it came to light during my stay in Athens with him in 2004. My interest in research on competencybased techniques for the Semantic Web and e-learning resulted in a chapter in Dr. Lytras and Dr. Ambjrn Naeves excellent book Intelligent Learning Infrastructure for Knowledge Intensive Organizations: A Semantic Web Perspective, published by IGP. That chapter was the seminal concept for this book. Later, I received feedback from the ideas on that paper from many colleagues worldwide. Those comments helped me in setting the scene for this book, which attempts to provide a comprehensive account of the use of competencies in organizational learning. In other aspects, this book is also a result of the patience and understanding of my wife Elena and my two-year-old son, Miguel-Angel. They are the real energy behind all my undertakings. I have also to mention the help, support, and patience of Kristin Roth, Idea Group development editor, and the rest of the IGI staff. Dr. Miguel-Angel Sicilia Alcal de Henares July 2006
xiv
Section I: General Concepts and Management Issues
Competencies and Organizational Learning 1
Chapter I
Competencies and Organizational Learning:

A Conceptual Framework
Miguel-Angel Sicilia, University of Alcal, Spain Ambjrn Naeve, Royal Institute of Technology, Sweden
Abstract
Organizational learning can be considered as systemic behaviour oriented to acquire capacities for dealing with the needs and challenges of organizations in competitive environments. This entails that there must be some measurable and lexible concept that drives the acquisition of those capacities. Competencies understood as the workplace capabilities of individuals or groups can be used as one of the approaches for managing such capacityacquisition behaviour. Even though competency is not a new concept, the management of competencies through information technology (IT) for improved effectiveness and eficiency poses a number of signiicant challenges that still require much research to come up with general-purpose and reliable solutions in the information systems discipline. This chapter delineates the main concepts for a competency-based framework in the context of organizations and enumerates some requirements for which deinitive and commonly accepted solutions are still not available.
Copyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
2 Sicilia and Naeve
Introduction and Background

Competencies as related to learning and work performance have received a great deal of attention in the last years. This has resulted in a plethora of papers and reports on how competencies should be deined or assessed, including some papers that deal with their representation in a computational form. But the literature on the topic is fragmentary in what concerns the concept of competency, since deinitions are in many cases taken for granted, and, in other cases, the studies focus on a single aspect of competencies. This makes it dificult for newcomers to have a clear understanding on the main issues that need to be dealt with when approaching competency-based systems. That lack of a systematic way of dealing with competencies calls for a clear statement of the requirements that are signiicant for studies and future research on the topic. This chapter attempts to provide a checklist, or roadmap, for such requirements. To do so, we will start with a concrete deinition for the term competency, and then we will proceed to describe the main requirements in a synthetic way, hoping that the resulting collection of aspects is valuable as a skeleton for the clariication and comparison of the concrete approach each author or system is providing in what regards to competencies. The notion of competency is linked to the concept of human performance, which according to the model of Rummel (Rothwell & Kazanas, 1992) encompasses several elements: (a) the work situation is the origin of the requirement for action that puts the competency into play, (b) the individuals required attributes (knowledge, skills, attitudes) in order to be able to act in the work situation, (c) the response, which is the action itself, and (d) the consequences or outcomes, which are the results of the action and which determine if the standard performance has been met. Finally, individuals usually receive some kind of feedback depending on the success or failure of their action. Since competency is related to human capacity, it is clear that learning is a process that eventually results in improving competency. In straightforward terms, then, learning activities can be arranged, scheduled, carried out, and evaluated with the aim of helping employees to acquire some required competencies for concrete work situations. This link is a irst important assumption in competency approaches to e-learning in organizational settings, which in turn leads to a irst requirement for them. Requirement #1. Competency-based approaches to organizational learning require an explicit model for linking competency deinitions to learning activity objectives. Even though this irst requirement may seem obvious, it entails a number of related requirements that are described in what follows and that do not have a straightforward solution. The irst one is the formal deinition of competencies, that is, the aspects of competencies that should be considered. Knowledge, skills, and attitudes are often mentioned as examples of such aspects or contributing elements. However, there is not a single deinition accepted, and the approaches will be different depending on the granularity and level of detail considered. This leads to a second requirement.
Requirement #2. The components of competencies must be clearly described. An ontological schema for the Rummel model cited above can be found in the work of Sicilia (2005). Further, relationships between competencies are an important element to be considered. A discussion of techniques for dealing with some typical competency relationships can be found in Sicilia, Garca, and Alcalde (2005). In addition to describing competency components, a deinition of the measurement instruments and scales is required. An option is an all or nothing approach, in which an individual has or not a given competency. However, it is more common to have some measurement scale, numerical in a range or qualitative, based on linguistic labels, as high or low. Requirement #3. The scales or constructs for measuring competencies must be deined. Following with measurement issues, competencies in organizational approaches will likely be aggregated in constructs as competencies of a team or competency level for the organization. These aggregations entail a concrete kind of algebra of competencies. For example, existing approaches differ in these aggregations to be compensatory or not (Sure, Maedche, & Staab, 2000). Requirement #4. The aggregation techniques considered valid must be deined to the level of allowing a form of computing aggregated competency levels. Competencies are, of course, the actual capabilities of employees, but the same concept can be used to express desires or needs, that is, competency requirements. When understood as a lack that should be ixed, one can use the term competency gap. This raises an additional requirement. Requirement #5. The expression of needs and objectives in terms of competencies must be speciied in terms of the same model used to express the capabilities of individuals or groups. In addition to the representational problems described so far, the procedures or techniques that are considered proper for the assessment of the presence of competencies in individuals require consideration. Such techniques may involve observation, tests, inter-subjective assessments, or any other technique that is considered reliable. Requirement #6. The procedures for the assessment of competencies considered valid must be stated.
4 Sicilia and Naeve
With the above requirements, the representation is equipped with the fundamental aspects. But more is required for an effective selection and targeting of learning activities to cover competency gaps. The irst of these additional aspects is that of how to combine pedagogies with the competency-based selection of activities or learning resources. Different kinds of pedagogical accounts could be considered (Sicilia & Lytras, 2005), and these could better it different settings. Another important element is how to combine the agendas and time constraints of employees with the targeting of learning activities. It is possible that the employees that are better prospects for the immediate acquisition of some competencies have, in fact, a workload that impedes it. These aspects illustrate the fact that algorithms that target learning activities to ill competency gaps in organizations are complex and require the consideration of diverse dimensions. The following section briely sketches the main elements of an information system approach to competency-based learning. Later, some speciicities of such an approach for the current state of technology will be mentioned, with the aim of serving as a roadmap for research and practitioner work in the ield.
Main Components of an IT-Based Approach to Competency Management

The above discussion provides a blueprint for competency management information systems. Since competency deinitions are potentially complex, and the amount of information on
Figure 1. Main generic components for competency-based approaches

Organizational planning database Organizational needs registry
Competency gap analysis
competency gap Learning objects/ learning activity repository
Competency assessment
Learning activity targeter
Organizational competency database
Organizational planning database
Organizational learning schedule
Competency defintion schema Human resources information database
employees, teams, and business needs is large, an information technology-intensive solution becomes a critical issue. Figure 1 depicts the main functional, generic components of a competency-based approach to organizational learning. Figure 1 includes functional components that store information, represented as rounded rectangles, and also a number of processes, namely competency gap analysis, competency assessment, and learning activity targeter. The following is a concise description of the elements in the igure: Thecompetency deinition schema is the realization of requirements #1 and #2 above, and the organizational competency database is the registry of all the competencies that are believed to be possessed by the employees. This last component is basically a part of a human resources database in which the actual competencies of each employee are described in detail. The competency database is updated by the activity ofcompetency assessment, which should at least be triggered after each scheduled learning event. According to requirements #3 and #6, this assessment requires pre-established, documented procedures to provide a degree of homogeneity that allows the aggregation of competencies. The organizational planning database contains the information of the calendar and activities of the whole organization, and the project plans with the information of the employees assigned to each of them. Theorganizational learning schedule (which could be considered a part of the planning database) contains the information of the learning activities scheduled, be they individual or team-based. It is supposed that some kind oflearning resources and/or learning activity repository exists, storing contents, learning plans, and other resources for reuse in future activities. In some cases, these resources might be outsourced or required on demand. Theorganizational needs registry aims to store the needs or desired capabilities of the organization. These may be derived from the needs in the projects or customer demands, or could be strategic, medium, or long-term directions. To comply with requirement #5, these needs must be expressed in terms of desired competencies.
The process of competency gap analysis takes as inputs the organizational needs and the organizational competency database to assess the mismatch between the two. For this to be effectively accomplished, clear algorithms and computation techniques (requirement #4) must have been agreed upon. Without that previous agreement, measurement of organizational progress in terms of competencies will be seriously lawed. The competency gap then becomes the key measurement element that is used for action in the organization. Following it, the decision process of targeting learning activities to the right employees at the right time requires a consideration of the competency gap, the available competencies, and the agendas of the people. The result of such a targeting process is that some learning activities with some concrete objectives will be scheduled and initiated through the e-learning platform available company wide.
6 Sicilia and Naeve
The succinct description above serves as an abstract framework of the key organizational assets that are required for a competency approach to learning. It can be used as a reference framework for the comparison of concrete studies, tools, or techniques, and, of course, it is simply an abstraction of the complexities of the competency paradigm that are relected in the requirements stated in the previous section. In what follows, some e-learning technologies providing important services for that framework are referenced.
Bridging Competencies and Modern E-Learning

Modern e-learning technology in the last years has been inluenced by the paradigm of learning objects. The concept of learning object is at the centre of the new paradigm for instructional design of Web-based learning that emphasizes reuse as a quality characteristic of learning contents and activities. For example, Polsani (2003) includes reuse in his deinition of learning object as an independent and self-standing unit of learning content that is predisposed to reuse in multiple instructional contexts, and Wiley (2001) also mentions the term in his learning object deinition any digital resource that can be reused to support learning. A number of speciications and standards that describe or make use of the learning object concept have evolved in the last years. The basic metadata elements associated with learning objects have been described in the IEEE learning object metadata (LOM) standard (IEEE, 2002), which organizes its conceptual metadata schema in nine categories: general, lifecycle, meta-metadata, technical, educational, rights, relation, annotation and classiication. Learning objects are considered as reusable elements that can be used as part of learning designs. In the experience and view of the authors, there are two important directions in learning technologies that are especially relevant for the competency approach described above. These are activity-orientation and the use of Semantic Web technologies. These are sketched in what follows.
Activity Orientation in E-Learning

The IMS learning design (IMS LD)1 provides a powerful language for the expression of learning designs targeted at the realization of activities. An activity is considered as a piece of interaction among a number of speciied roles played by persons who produce a tangible outcome by using a concrete environment made up of learning objects and services (facilities available at runtime). Activities can be further decomposed in sub-activities, and they are aggregated into methods that specify the conditions for application of the learning design, along with the planned objectives that will eventually match the outcomes of the activities. Methods can be structured around concurrent plays, and these in turn can be structured in sequential acts, the latter allowing the speciication of execution conditions. This schematic description of LD gives an idea of the lexibility the speciication provides in describing activity-based learning programs.
These activities can be framed in the context of process-orientation. The processes in an organization are related to different goal, obstacles, actions, and prerequisites (GOAP). We will now describe the main elements of the GOAP approach to process modeling (Eriksson & Penker, 2000). To start with, relationships between goals as dependencies and associations are introduced. The dependency should be interpreted as stating that the fulilment of the smaller (partial) goal contributes toward the fulilment of the larger (dependent) goal. A goal that has been completely broken down into partial goals indicates that the goal will automatically be fulilled if all of the partial goals are met. In connection with describing the goals we also describe the obstacles that stand in their way. An obstacle is a problem that hinders the achievement of a goal. By analyzing the problem, new goals or partial goals are discovered that attempt to eliminate the problem. An obstacle is therefore always linked to a goal. Similar to a goal, an obstacle can also be broken down into partial obstacles. Obstacles are eliminated (overcome) by actions. An action plan can be formulated from the goal/obstacle model, where temporary obstacles are resolved as soon as possible, and the goals linked to the continuous obstacles are allocated to processes in the business. Finally, for each process, prerequisites take the form of input resources or supporting resources. The outcomes of the process module are relevant to different stakeholders in the organization, and the connection of the outcomes of concrete activities with the input and support of others provides a way to explain the transition from the individual to the organizational behaviour. Competencies in the GOAP framework can be used as a speciic kind of goal with some added descriptions. In this view, relationships between competencies and their constituents can be considered prerequisites; for example, it is required to have competency X prior to be able to have competency Y, or competency Z requires some attitude A. Further, learning activities as expressed in IMS LD can be considered speciic actions. Thus, process models as GOAP combined with more learning-speciic languages as IMS LD provide a seamless platform for competency approaches, with the added beneit of having available free technology implementing the worklow and automation aspects of IMS LD units.
Semantic Web Technologies and Competencies

The Semantic Web vision described by Berners-Lee, Hendler, and Lassila (2001) represents an approach to enhancing the current World Wide Web with machine-understandable Semantics. The essential idea is that Web resources as identiied by URIs can be described by metadata with the purpose of enabling automated processing. The precision of logics-based descriptions (Baader, Calvanese, Mcguinness, Nardi, & Patel-Schneider, 2003) when used for creating metadata would allow for the creation of tools that do not solely rely on natural language processing as current search tools, thus ideally removing the problems associated with ambiguity and implicitness in natural language. There are two fundamental kinds of beneits of using ontology description languages to express learning object metadata (Sicilia & Garca, 2005). On the one hand, those languages provide richer knowledge representation formalisms for metadata descriptions than using plain text, extensible markup language (XML) bindings, or even resource description frameCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
8 Sicilia and Naeve
work (RDF). Here the beneits are the result of using description logics, instead of simply using structured data in XML format or using RDF, which is a less-expressive language than DAML+OIL or ontology Web language (OWL). On the other hand, the use of ontologies may eventually produce synergies with the technological advances that are taking place under the overall label of Semantic Web. The most prominent of such synergies may come from the availability of shared, consensual ontologies on many domains along with tools to develop systems that exploit them for diverse intelligent behaviors. Further, the activity-oriented approach can be used in conjunction with ontologies as described by Sicilia (2006). Further, competencies can be speciied also through ontologies as described in (Sicilia, 2005). Thus, ontologies of competency descriptions and learning object ontologies can be used together as a framework for the competency-based approach.
Conclusion
Competencies as a representation and measurement paradigm can be used as a structuring principle for the selection and targeting of learning activities. However, such an approach requires the consideration of a number of non-trivial requirements, including how competencies are described and assessed, and how they are aggregated and combined. The competency approach for driving organizational learning can be expressed in a number of functional elements and processes that provide a framework for the analysis and comparison of the variety of techniques, representation schemas, and algorithms that can be devised. Such framework has been sketched in the paper. Finally, the deployment of IT-intensive competency systems requires advanced support for the representation of competencies and learning activities. IMS LD and ontology languages as those used in the Semantic Web have been described as two key enabling elements in that direction.
References
Baader, F., Calvanese, D., Mcguinness, D., Nardi, D., & Patel-Schneider, P. (Eds.). (2003). The description logic handbook: Theory, implementation and applications. Cambridge: Cambridge University Press. Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The Semantic Web. Scientiic American, 284(5), 34-43. Eriksson, H. E., & Penker, M. (2000). Business modeling with UML: Business patterns at work. Wiley Computer Publishing. IEEE Learning Technology Standards Committee (2002). Learning object metadata (LOM). Final draft standard, IEEE 1484.12.1-2002.
Polsani, P. R. (2003). Use and abuse of reusable learning objects. Journal of Digital Information, 3(4). Rothwell, W., & Kazanas, H. (1992). Mastering the instructional design process. San Francisco: Jossey-Bass. Sicilia, M. A. (2005). Ontology-based competency management: Infrastructures for the Knowledge Intensive Learning Organization. In M.D. Lytras & A. Naeve (Eds.), Intelligent learning infrastructure for knowledge intensive organizations: A Semantic Web perspective (pp. 302-324). Hershey, PA: Information Science Publishing. Sicilia, M. A. (2006). Semantic learning designs: recording assumptions and guidelines [Special issue]. British Journal of Educational Technology, 37(3), 331-350. Sicilia, M. A., & Garca, E. (2005). On the convergence of formal ontologies and standardized e-learning. Journal of Distance Education Technologies, 3(2), 12-28. Sicilia, M. A., Garca, E., & Alcalde, R. (2005). Fuzzy specializations and aggregation operator design in competence-based human resource selection. In F. Hoffman (Ed.), Advances in soft computing (pp. 219-230). Springer. Sicilia, M. A., & Lytras, M. (2005). On the representation of change according to different ontologies of learning. International Journal of Learning and Change, 1(1), 66-79. Sure, Y., Maedche, A., & Staab, S. (2000). Leveraging corporate skill knowledgFrom ProPer to OntoProPer. In D. Mahling & U. Reimer (Eds.), In Proceedings of the Third International Conference on Practical Aspects of Knowledge Management (PAKM 2000) (pp. 30-31). Wiley, D. A. (2001). The instructional use of learning objects. Association for Educational Communications and Technology, Bloomington.
Endnote
1
http://www.imsglobal.org/learningdesign/
10 Liyanage
Chapter II
Competency Management Systems and Technologies

Shantha Liyanage, University of Auckland, New Zealand
Abstract
Learning processes are essential in building individual and organizational competencies and transfer of learning and knowledge among individual and organizational learning space. Several types of learning technologies are now available for systematic building of individual and organizational competencies through formal and non-formal learning processes. This chapter examines the need for competency-based management systems that provide organizations with opportunities to implement such learning processes through Web-based e-learning technologies that are conducive for creating and building systematic knowledge transfers across individuals and groups. It is argued that the transfer of knowledge, learning, and learning interactions through such technologies is a seamless process that reinforces the continual renewal of knowledge and learning through appropriate management systems. Such systems are created by deliberate interventions for individual and organizational learning and should be context- and content-speciic.
Competency Management Systems and Technologies 11
Introduction
Why is Learning Important?
Learning is the initial and subsequent acquisition of capability that allows individuals and organizations to plan, organize and perform particular tasks and actions. Human beings, by their nature, are inquisitive about the unknown and are eager to upgrade their skills in performing various tasks with conidence. They are adaptive creatures who are capable of learning and imitating (Carroll et al., 2002). Different individuals learn using different methods. Some learn by reading, conceptualizing, and observing, whereas others learn by trial and experimentation. Learning is essentially a tool to acquire and build competencies. With learning, individuals respond to the changing conditions and deal with new challenges and fulil aspirations. Learning is also a fundamental process to quench the thirst for knowledge. Learning takes place at all times throughout human life. Through learning, individuals build the character and the conidence to acquire skills that are useful and marketable. Certain things can be learned through compulsion, such as having to work in an unfamiliar organization. Several behavioral, cognitive, and constructivist theories have been proposed to explain why and how individuals learn. Individual learning contributes to collective organizational learning and to building organizational competencies and facilitating growth. Learning is also seen as a reliable process by which organizations can identify, deploy, and build on existing capabilities, as well as planning to build future competencies. Without learning, individuals and irms are like insects without antennae. Learning needs to take place at all levels in an organization. Even chief executive oficers of organizations have to undergo the process of learning. As long as an organization continues to evolve and undergo changes, learning has to take place. Winter (2000) argues that the process of learning is a series of online trials interspersed or alternated with variable periods of off-line deliberations and analysis. In this way, a irm will draw on multiple sources of knowledge interspersed with various individuals and organizations. Individuals in a irm are expected to churn this knowledge into discrete experiences through relection and action. How individuals learn is an important topic that has been explored over centuries. Aristotle (384 - 322 BC) emphasised a balanced individual development that requires nourishing body, mind and soul. Learning is focused on individuals who are capable of learning, and consequently the process requires investment from individuals as well as organizations. Firms are compelled to invest in both inancial and human resources to acquire necessary capabilities when they do not already possess those capabilities. Some countries, for example, Australia and Germany, have adopted legislation to impose compulsory training levies from organizations in order to train employees and thereby provide more places for apprenticeships for young people entering the labour market. In Australia, the building and construction industry training levy is payable before a building proposal is granted. Some countries, however, resort to voluntary arrangements of training between employer and employee. British policy, for example, relies largely on individual initiatives for train-
12 Liyanage
ing investment by both employers and workers (ECITB, 2006). Some of these voluntary contributions are managed by organizations such as Skills Investment Funds (SIF). New Zealand has approached the skills and training development through the establishment of Centres of Excellence for Training in different industry sectors. Such efforts are supported by major government initiatives, such as the Growth and Innovation Fund (GIF), which provides skills and training development funds. Learning has many outcomes: new ways of doing things, sharpening existing skills, building conidence, character and capability of individuals and organizations, creating collective interests, and closing the competency gaps in organizations. The decision to invest in learning is an implicit recognition of the need to develop individuals and renew organizational competencies in order to perform and deliver work and production in accordance with the acceptable industry standards. Learning investement is based on the managerial perception of the value of learning, the appropriateness of learning methods and costs involved. The organization must have strategies to incorporate an individuals learning into its business routines. Incorporation of individual and organizational learning will depend on managers who recognise the training and learning needs of individuals and who have committed themselves to attaining a desired level of human capability development. Once this level of competency is achieved, managers may be inclined to cut back on training efforts. In most organizations, learning is aimed at acquiring essential skills and competencies, changing attitudes and cultures, and developing deep skills and knowledge bases that differentiate one organization from another. As a result, the competencies in learning spread across institutional boundaries and power structures. Decisions to adopt the most appropriate and effective methods of learning are based on the subjective judgment of the managers of an organization. Such decisions often come with a mandate from the chief executive oficers (CEOs). Some irms have embarked on learning strategies to deal with managing change. For instance, a leading Australian construction company in Brisbane has implemented institution-wide learning partnership programs to prepare both the staff and the institution to deal with future challenges brought about by sector technological changes. The CEOs future vision and strategic directions are instrumental in turning this construction company into an innovative and knowledge-based learning organization. As evidenced in this organization, different management approaches have shaped individual and organizational learning and knowledge management processes.
Why are Competency Management Systems and Technologies Important in a Modern Economy?
Today, many organizations are confronted with new forms of work and production. Employees are highly mobile, and the traditional system of life-long working, secure employment and employee loyalty is rapidly changing. Industries are relocating to countries with comparative advantages, giving rise to job insecurity, redundancy and the necessary multi-skilling of workers. These pressures, together with collocation of industry to different parts of the world and outsourcing non-core activities, have introduced new models of learning, especially adaptive learning among employees.
This chapter deals with the competency management systems and technologies available for individual and organizational learning. It is necessary to understand the process of learning before actual learning technologies can be adopted and used in organizations. Different individuals display different styles of learning and action. The connection between individual and organizational learning is explored to address these questions: Why do individuals and organizations differ in learning? What management structures and processes are needed for effective implementation of learning processes? How does learning improve the level of competencies among individuals and organizations? Using research studies conducted on e-learning academic programs, the merits and demerits of e-learning technologies are discussed. The lessons learned and the challenges faced in managing competencies are discussed to address the complexity of adopting individual and organizational learning strategies in organizations. Although various technological options are available to improve learning processeses and upgrade skills and competencies, knowledge-base organizations require careful learning strategies and organizational structures.
Managerial Attitudes toward Learning and Change

Not all managers favour investment in learning. The expectations of some senior managers are that people with skills should be able to it in directly with organizational needs. As a result, these organizations tend to regard training and retraining as a necessary but costly investment that they would rather do without in an organization. Change is a dificult process to manage, and all changes require learning and unlearning. As a result, most organizations are reluctant to accept changes and to prepare for learning. Traditionally, managers have been reluctant to recognise learning as an important component of capability building and augmenting organizational competencies. This view, although prevalent in some organizations, has gradually changed with the increase in competition and globalisation. Increasingly, most industries have allocated a training budget and now encourage continuous learning for their staff. Continuous and rapid changes to science and technology knowledge are another reason to encourage training. Individuals and organizations now have to deal with technical complexities of operations that require a new set of skills. Although most routine work is now mechanised, human intervention is necessary to conduct critical functions and the advancement of the industry. Rapid changes to industry structures have also contributed to the undertaking of continuous learning in organizations. For example, in the computer industry, even as software is continuously produced, vendors also need to create and recreate their own software production, as well as sales, in order to be competitive. In some cases, it is dificult for the industry to keep track of all changes, and it can often fall behind when there are no systems in place for learning and change management. The changes to labour laws, together with the expectations of employees to advance their career prospects, have also played an important role in training and learning. From a career development perspective, learning needs to be seen as a mechanism to enhance an individuals current and future potential to perform at his or her best. Consequently, learning is now recognised as an important part of organizational survival and renewal.
14 Liyanage
The Process of Learning

What is the Exact Process of Learning and How are Individuals and Organizations Involved in Learning?
Typically, the assessment of training needs is conducted at the same time as performance reviews of employees. The managers discuss an individuals performance with the respective individual to ensure organizational expectations are met. A checklist of training needs is drawn up to ensure individuals are given the opportunity to improve their performance in order to contribute to the future growth and development of the organization. Such discussions are conducted by the peers of individuals and can be based on subjective interpretation of performance, which is often dificult to measure accurately. There can also be complete or partial comprehension of an individuals and the organizations aspirations. The training programs for individuals are provided taking into consideration the future plans of the organization and the availability of resources, time and appropriateness of such training. Given such individual and organizational needs, the process of learning takes place with individual initiatives followed by collective learning through sharing experience and facilitating interactions that are both formal and informal learning mechanisms among individuals and groups.
Instilling Learning as an Agent of Change

What are the Dynamics of Providing Training for Staff?
In high-performing irms, learning is instituted in organizational procedures. The City Bank, a leading inancial organization, for example, has adopted a learning procedure in which staff are encouraged to identify their needs and then allowed to attend training of their choice. This training philosophy caters to both informal and formal training according to each individuals choice and needs. This has become part of senior managers' strategy to ensure that staff are provided with adequate incentives for motivation and commitment to the organization. Training can be stratiied according to management functions. For example, at Beca Carter Ltd in New Zealand, training is well integrated into management functions though formal training providers. At those organizations in which training is not a part of organizational culture, managers expect individuals to take their own initiatives to upgrade skills in order to perform daily business routines. It is customary that employees will learn from each other, and senior employees are expected to coach and mentor new employees. Several knowledge management tools are available to train new staff. The training managers, as in the case of PriceWaterhouseCoopers (PWC), New Zealand, are able to integrate knowledge management
tools for training and the diffusion of knowledge. The training manager of PWC, has the dual role of training and managing knowledge in the company. Some other irms deal with training as a separate function from knowledge management activities. Various knowledge management tools, such as Web-based learning, are used to gather classiied and iltered information that is essential to conducting business tasks. The attitude toward training depends largely on an individual managers perception of the value of such training. Some are reluctant to impart their skills, due to fear of competition and losing ones control over certain tasks. The training manager of a leading company in Hong Kong argued that, We are reluctant to train the staff due to the fear of losing employees. However, there is always a risk of losing a well-trained person, who may potentially be hired by competitors. A well-trained person has a high probability of inding a new employer, and employees look for organizations where their services are appreciated and their careers and self-esteem are enhanced. Salaries are not always a dominant factor in deciding who to work for. Highly marketable individuals look for organizations they feel have the right environment to work in. This is particularly common in information technology (IT) and communication companies, where skilled individual employees tend to change jobs frequently. This has led to managerial reluctance to provide training for their staff and a general preference to hire well-trained people who will perform at no extra cost to the irm for training or retraining. Such fears are often a hindrance to maintaining learning and training efforts in organizations. On the other hand, some employees maintain that they are capable of performing all tasks, hence future training is seen as a waste of time. The managements insistence on training is sometimes viewed as a lack of recognition of their capabilities. Small- to medium-sized irms approach learning by allowing greater interaction among staff. In a small irm, the level of interaction among staff is considerably high. In such an environment, learning can take place through consultation, face-to-face interactions, discussions, and doing things together.
Learning as a Continuous Process

Learning requires conscious and unconscious processes that assimilate synthesis and relect on learning experiences. These learning experiences can be direct, wherein individuals encounter certain experiences through his or her senses. The other form of learning is through knowing and relecting on others experiences. Discussions of such experiences, and the sharing of insights by teachers and others, allows the learning to take place within individuals. A variety of learning processes are adopted in formal and informal learning. These processes and methods of learning have been discussed extensively. Kolb (1984) argued that learning is an interactive process that consists of action and relection. Kolbs learning cycle explains how individuals learn through a four-stage model. According to this model (Figure 1), individuals learn through concrete experiences followed by relection. These experiences and relections can trigger the understanding of abstract concepts leading to doing, connecting and testing. In this way, individuals and organizations acquire skills, develop technological expertise, and investigate the how and why of processes. Relection is an important process of internalising information and knowledge. It consists of thinking about and making sense of experience, options and possibilities. Such relective
16 Liyanage
activities involve planning, evaluation, execution, control and management. Conceptualization can also be a starting point, where some will rely on abstract thinking and then doing, rather than the other way around. Leading institutions and industries bring forth advances in technology and production processes, thereby attracting competent individuals into the organization. Such organizations look for individuals who have a broad range of skills aside from technological competencies. They are also looking for loyal and committed individuals who can it into the organizational culture and people. Individuals need to be happy in their work place and should have career development prospects. Several techniques are available to measure behaviors and personality traits of individuals in order to determine whether their aptitude and capabilities it the work environment. These tests, such as psychometric tests, have been developed and used by various irms. For example, leading information technology companies, such as IBM, have developed the team talented project to attract the highest-skilled graduate students as part of hiring and mid- and long-term business generation. The ability to conceptualize a problem derives from the conscious processes of learning new theories and developing cognitive capacities to understand and to reason out certain activities and tasks. Planning activities involves testing new situations and setting the learning processes on the correct trajectory. Institutions have developed various learning processes in order to satisfy the learning needs of individuals. In the process of learning strategy development, it is important to focus learning on particular objectives in organizational development and to assess the current levels of skills among individuals as well as future skills development needs. This allows the organization to know what individuals can do now, as well as what they can do in the future. A critical success factor for a knowledge-based organization is inding the right person to do the right joba highly qualiied person can bring only partial skills to the organization, and the remainder needs to come from learning through interaction with others. There is a sharp distinction between individual and organizational learning. This distinction is evident in the manner in which information is assembled and digested between individuals and organizations. Individuals tend to focus on information and knowledge that relates to their competency levels and that satisies their intellectual needs, whereas organizations are keen to satisfy information and knowledge needs that relect on external conditions, such as markets or collective interests held by shareholders or a board of directors. One persons learning may positively affect the behavior and attitudes of others, depending on the observed outcomes. If there is much to gain from observed differences, others will learn from that outcome by trying to imitate it. Individuals commence the learning process by understanding laws, principles and other determinants that lead to rules and procedures of engagement. These rules and procedures determine the behavior and actions of individuals, and the learning strategies that will lead to speciic results and outcomes. Learning is individually driven, and once individuals have learned some skills, the next question is how the organization will incorporate that learning into the organizations business routines, procedures and assets. In other words, how do we incorporate individual learning into organizational learning? The concept of organizational learning (sometimes referred
Figure 1. Modiied Kolb learning cycle for organizational learning
Conceptualization
Planning
Reflection
Experience
to as a learning organization) has been discussed to deal with the organizational aspect of learning and to ensure that organizational structures facilitate learning and actively create and diffuse knowledge. The deinition of a learning organization can therefore be termed as an organization that is skilled at creating, acquiring and transferring knowledge, and at modifying its behaviour to relect new knowledge and insights (Garvin, 1993). This deinition focuses on the style of learning rather than the content. The content and product of learning are also important, however, as some organizations have multi products and services that are usually provided by a chain of stores and outlets. Organizational learning takes places when the organization concerned addresses a particular problem or cluster of problems confronting the organization. It then solves those problems through building a collection of lessons learnt and assimilating competencies that represent the collective learning of present, past, and future employees. Thus organizational learning is an ongoing interactive process that combines the skills and knowledge of employees.
18 Liyanage
In a cycle of experiential learning, Argyris and Schoen (1974) proposed that people have mental maps determining how they proceed to act in a particular situation. This results in planning, implementing and reviewing their actions. These three elements function as distinct loops. Learning proceeds with the engagement and outcomes of learning, and, thus, the process of learning can be considered as an integrated loop with feedback functions. Two types of learning loops have been identiied. Single-loop learning starts by following the rules of engagement that determine behavior and action, leading to outcomes (Argyris & Schoen, 1978). In single-loop learning, individuals attempt to ask how questions, which lead to accomplishing the task at hand. A majority of workers follow such single-loop learning by knowing the exact rules and doing them accordingly to achieve the desired outcomes. Single-loop learning emphasizes the detection and correction of errors within a given set of governing variables. Double-loop learning, on the other hand, questions and challenges the principles and determinants that lead to rules. Double-loop learning is about iterative interactions with the governing variables themselves. It involves radical changes and revisions to the organization and its learning systems. Double-loop learning asks why questions and challenges the organization and its relationship to outcomes (Figure 2). Experiential learning by trial and error is about repositioning capacity to know how reliable our actions and responses are and how to evaluate them accordingly. What we learn provides us with cumulative capabilities to act in anticipation or respond to a particular set of problems. Even among advanced learning centers, such as universities, learning cultures differ vastly and have adopted different learning, experimenting and researching techniques.
The Role of Higher Education Institutes

Historically, formal learning is organized by higher learning institutions that are empowered to formalize skills and training development work. Universities are as old as human civilization. Figure 2. Learning processes in knowledge-based irms
Single Loop Learning
Governing Variables And Concepts
Action and Strategy
Consequences And Outcomes
Double loop Learning
The Royal Charters were set out for constitution and governance structures outlining how the university should function. Most contemporary universities operate under the legislation of governments and under authorities responsible for higher education. These institutions are primarily responsible for turning out future skilled workers needed for industry and government. In general, higher education institutions develop teacher-centred learning processes where the culture of teaching is the central focus. Innovative learning organizations have taken a different approach with the scope of learning extended to deal with speciic areas of operation. For example, miners working in a coal pit will have to perform certain routines and will have a limited scope of learning and understanding. This scope varies with the products and techniques that are used in production. These variances need to be taken into account when developing learning strategies and approaches to learning.
What is the Role and Function of Higher Education Institutions in Learning?

The main objectives of higher education institutions are to instil necessary higher learning skills and capabilities among individuals by way of providing new state-of-the-art knowledge. Learning is partly an individual decision-making process whereby each individual makes a conscious decision to acquire a set of skills. At higher learning institutions, individuals are expected to learn not only a particular set of skills but also the total transformation of character and behavioral attitudes that are necessary to deal with various decision-making situations. These individual decisions are based on particular needs, as well as an organizations future development strategies. Some of these decisions are based on rational thinking that steers the organization toward a certain direction. However, decisions to undertake training are not always based on rational criteria. Often such decisions are based on the satisicing principle, a theory of choice focused on the process by which alternatives are examined and assessed (Winter, 2000). Not all learning is feasible in ones lifetime, and, consequently, decisions can be arrived at according to what is practicable under speciic conditions. As Simon (1987) puts it, what is desirable is to look for the bestlooking for the sharpest needle in the haystack (optimizing)while acquiring the best possible skills in a given timelooking for a needle sharp enough to sew with (satisicing). In this way, learning institutions help prepare individuals for decision-making tasks by providing them with different skills.
Managerial Inluence in Learning

Learning decisions are also inluenced by differing managerial styles where the extraction of the best out of people versus providing optimal conditions to perform at their best become the guiding principles. The availability of funds, the management perception of future training, and learning needs all inluence the decision-making process. In a changing climate, it is not easy to identify precisely the level of organizational needs and competencies. Increasingly, most organizations have realized the important role of learning in order to
20 Liyanage
enhance the quality of services and improve the organizational eficiency and competencies. In the process of empowering varied individuals, organizations have adopted multi-prone strategies to provide learning opportunities. These include secondments and rotation of individuals in their jobs and workplaces, conducting in-house training, encouraging personal development, relocation of staff to different overseas ofices, adopting a learning culture in the organization and developing explicit learning strategies, and conducting formal learning for employees in afiliation with recognized learning providers. With the expansion of information and computing technologies, organizations are now in a position to develop knowledge management systems, including extranet, intranet and e-learning tools. The methods of linking knowledge, learning and innovation are not always obvious. Individuals view learning as a method of fast-tracking career advancement and enhancing individual competencies. As a result, learning is highly dependent upon peoples motivations, commitment and attitudes toward learning. From an organizational viewpoint, the process of learning and knowledge creation aims to develop intellectual capital that contributes to organizational competencies. These competencies are distributed unevenly across the organization and are spread variably among individuals. In order to improve competencies and bring them to certain comparable levels, institutions have to adopt training strategies for their staff. Such training and learning enables the organization to maintain distinct competencies. Competencies can be regarded as composite layers of skills and knowledge bases residing in the organization. These competencies are dynamic and change as a irm matures and grows, together with the development of a network of services and external relations. In response to constant changes induced by scientiic and technological advances, a continuous renewal of competencies is always required in order to gain new knowledge and skills. In this way, an organization is able to build critical competencies that are unique and hard to imitate. Learning can also be viewed as a part of strategies to acquire knowledge and skills from other organizations and individuals. Rather than looking for internal sources, this kind of learning allows organizations and individuals to transfer and adopt skills and knowledge from others within a short period of time. Individuals are able to observe how others work and imitate those activities.For an organization to operate at a level at which it can competently deliver a particular product, process, or service that meets an industry standard, it has to learn to acquire a composite set of skills and competencies. Consider, for example, how a supermarket selling goods will need employees who are good at organizing, rearranging, receiving and dispatching goods, providing client services, and managing logistics. Although some of these skills can be easily learned, some essential skills differ according to educational level and competencies of employees. What is learned needs to be put into practice. The way different individuals perform the same task can vary according to their learning capacities and the ability to translate such knowledge into action. For example, dealing with customers and handling complaints usually depends on individuals temperaments and their behavioral attitudes. Composite skills, knowledge, behavioural, and attitudinal traits prepare an individual to perform at an optimum level. What has been learned therefore forms a veil of competencies that assists individuals to respond and react to particular circumstances. Given such variations as well as connectedness of learning with competencies, how can learning, in particular e-learning technologies, assist in building management competencies?
The key capabilities for effective learning and development are: a. b. c. d. Knowledge synthesis capability Knowledge absorption capability Knowledge sharing capability Knowledge exploiting or utilizing capability
These four elements form the backbone of learning processes of all organizations. Knowledge-sharing activities in an organization are closely linked with learning processes. If the organization and individual engage in learning within the organization, knowledge sharing takes place irrespective of dedicated technology and procedures available. Different individuals and organizations approach the learning processes remarkably differently. For example, within academic institutions, the process of knowledge sharing is institutionalized by research and teaching. For non-academic organizations, the better approach may be to develop learning partnerships. In other words, knowledge sharing is intrinsically connected with organizational culture.
Individual and Organizational Learning

Individual learning is about learning through formal systems where each individual makes his or her choice of learning to gain a required set of skills in order to earn a living. Organizational learning, on the other hand, is a concept that relates to how organizations position themselves to acquire a range of skills and capacities. Organizational learning is described as the way organizations build, supplement, and organize knowledge and routines around their business activities and business cultures, as well as the way they adapt and develop organizational eficiency by improving the use of the broad skills of their workforces (Dodgson, 1993). Literature on individual and organizational learning takes several competing perspectives. Easterby-Smith et al. (1998), March and Olsen (1975), and Simon (1991) support the cognitive approach to learning, in which individuals are considered capable of making rational decisions and processing information that is necessary to respond and react to particular issues. In doing so, individuals need to operate within both the organizational rules and procedures, as well as pragmatic needs. Cognitive needs are based on logical and rational processes of reasoning and understanding. Learning is therefore framed according to an individuals cognitive and absorptive capacities (Simon, 1991). Because of the variation between individuals cognitive capacities, different individuals are capable of processing information to varying degrees, resulting in differences among individuals in their learning, absorbing and retention capacities. The social perspective of learning is another approach in which individuals and organizations learn through interaction and experience, speciically the experience of learning to do something and actually doing it.
22 Liyanage
The formation of communities, networks and groups enables knowledge exchanges. Such social approaches are closely linked with the social construction of knowledge through communities of practices (Brown & Duguid, 1991; Lave & Wenger, 1991). Learning thus takes place as individuals attempt to resolve problems and ind solutions (Gherardi, 1999). Such approaches are particularly useful when performing tasks that have not been attempted before. For example, the construction of particle accelerators at European Organization for Nuclear Research Organization (CERN), Geneva, is a case wherein theoretical understanding and previous experiences are the only guiding lights for performing complex tasks. Individual learning is a process of self-actualisation where contributions and recognitions form a central part of achieving organizational competencies. Most individuals are driven by dreams of achieving and accomplishing things that they believe are central to realizing their intellectual and physical needs. To a great degree, imagination and future dreams are the fodder for learning. Kant (1933) describes imagination as the unconscious and blind form that gives unity to the representations of both judgments and intuitions. Learning sparks creativity and imagination, and uplifts the human spirit. This human spirit and commitment determines the ability to know ex-ante and to form judgments and decisions that envision the possible futures and individual dreams. Organizational learning is largely problem- or situational-based. For example, a company dealing with waste management solutions using a particular process may have to resolve problems that are related to consumer behavior and attitudes. These attitudes present varying types of technological solutions that are appropriate to different waste management processes. In dealing with such problem-based situations, individuals are confronted with illconceived and ill-contextualised, vaguely structured situations. Individuals have to learn by interactions and have to provide meaningful and workable solutions to the problem. Solutions to such problems may be resolved by combining different knowledge bases and interactions with others to ind a common solution. This sort of learning enhances critical thinking and problem-solving capabilities of individuals while generating individual creativity and self-conidence.
Knowledge as a Distinct Capability

The process of learning is also about creating opportunities and capacity to explore knowledge resources. Any system that is capable of installing and renewing management has to renew its knowledge base. The phenomenal impacts that knowledge resources will have on organizations and individuals have been studied from a variety of perspectives, including: new product development processes (Daft & Lewin, 1993); organizing business information (Sanchez & Heene, 1996); strategic management of resources (Lane & Lubatkin, 1998); managing the value chains (Cockburn et al., 2000); organizing innovative competencies (Christensen et al., 1998); and learning capabilities (Winter & Szulanski, 2000). Knowledge competencies are essential for the dynamic growth of an organization. Knowledge is a distinct capability that contributes to the improvement of a irms eficiency. A irm organizes its knowledge through a variety of individuals who bring together distinct competencies. Inkpen (1998) and Inkpen and Dinur (1998) recognise that knowledge provides the capacity for organizational action and new knowledge provides the capacity for organizational renewal.
Organizational competencies should essentially complement individual competencies. A successful organization is one that attracts and retains individuals who can contribute to it. Individuals, on the other hand, continuously source knowledge and information through direct interactions and external connections. Some of these knowledge processes are systematic, while a majority fall under the umbrella of random processes. Such random processes are circumstantial; for example, a group of construction workers building a house may encounter varying and unfamiliar circumstances that need to be dealt with instantaneously using intuition, prior knowledge and smart thinking. Learning under such situations is context-speciic and random, and the previous experiences and knowledge become a distinct capability.
Forming Effective Learning Strategies

Learning is part of human relationships. It is embedded in the personal gratiication obtained through possessing knowledge and imparting that knowledge to others. Learning is an old tradition, and the methods of learning differ according to various philosophies of learning and knowledge diffusion. At least four types of philosophies relate to knowledge: a. b. c. d. Inquiry and debate (Aristotelian and Platonic common-sense way of thinking and reasoning) Experience and relection (Cartesian and Hippocratic styles of learning) Consultation and conferring (Witgenstiens logical space) Theory of practice (Bourdieus social practice-habitus)
A systematic development of a irms capability requires the continuous synthesis of knowledge and learning. Such knowledge and learning takes place as individuals engage in inquiry and debate about new knowledge. The generation and acquisition of knowledge are interconnected processes that are now regarded as essential to facilitating the formation of knowledge and intellectual capital. Traditional discursive structures and forms of learning are based on language and the exchange of meanings. These meanings are usually interpreted variedly by different individuals because of the variation in cultures and richness of languages. Learning is particularly effective when language, meaning and metaphors form a rich interpretation of the logical argumentation and reality of the inquiry and debate. Learning also happens as an individual observes certain phenomena and learns from it. This form of learning takes place in work situations in everyday life. Imitation and experimentation will then lead to relection on how things actually work on the ground. Consultation and conferring refer to learning by doing and discussion. Very often, internships of doctors, nurses, and other trades people have a phase of learning through consultation and conferring, during which an individuals ability to diagnose, analyse and provide effective solutions is assessed. Learning also occurs through ones surroundings and social practices, which is what Bourdieu (2004) described as habitus. According to Bourdieu, the habitus is embodied in individuals and is a collective and homogenous phenomenon that is mutually adopted for a social group or class. The habitus consists of certain elements of perceptions and appreciation in the ield it is created. For example, in agriculture, the habitus will constitute a
24 Liyanage
constellation of perception and habitual conditions of farmers, agricultural extension oficers and researchers who will conduct a meaningful discourse of learning for their existence and operations. Learning is, therefore, a process of the accumulation of knowledge, experience and interpretations. The knowledge process in a biotechnology irm, for example, is accumulated over several years of searching before biotechnological innovations can be marketed. During this process, experience and knowledge form varied combinations, and some become enriched while others erode over time. Similarly, a irm intending to globalize its business will have to extend its knowledge base and learn about the special conditions entailed by a particular technological context. In both cases, knowledge and learning are partly situated contextually and are partly driven by strategic intent. The act of conceptualization, together with the social synthesis of tasks, presents individuals with numerous opportunities to engage in learning.
Learning Strategies
Are Learning Strategies Intended in Learning Organizations?
Learning strategy can be described as various methodologies and approaches to engage the learner in a particular training course or programme. Learning strategies are formed by both the learner and the teacher in order to meet certain outcomes. These strategies often refer to what the teacher or learner may pursue in order to achieve speciic learning outcomes. These strategies will relate to either learner-centered learning strategies or teacher-centered instructional strategies. A irm embarking on a learning strategy has to carefully select the type of competencies and the level required to achieve such competencies. Without such understanding, learning will become a burden to the institution and will not translate to real competencies and capabilities. A irm can ill-afford to invest in developing all conceivable skills and knowledge requirements. As a result, most irms do not operate with the full understanding of every aspect of knowledge needs. Knowledge interchange with other irms is therefore necessary and is a source of a irms externally situated capabilities. This knowledge, however, is context- or problem-speciic and will only be called upon as and when a irm needs this knowledge. Some skills, such as searching for new knowledge, will require multi-tasking abilities, therefore learning out of traditional modes of training. In such cases, most irms opt to acquire knowledge through observing others and imitating. No two individuals perform the same task in exactly the same manner. Those who perform tasks that demand high levels of skills have to focus and exercise patience, as well as be able to rely on their prior knowledge and experience to perform those tasks. Very often, theory and practice help individuals form judgments that are critical to accomplishing the task. A irms ability to internalize such knowledge determines its capacity to generate future knowledge. A irm may choose speciic strategies to renew its internal knowledge capacity or decide to tap into externally situated knowledge bases. In both circumstances, learning is
the key process that allows a irms knowledge to be developed and extended to its business routines (Zollo & Winter, 2002). Different irms display varying levels of capabilities in assimilating experience and knowledge competencies. Such capabilities are generally classiied as a irms absorptive capacity; that is, a irms ability to value, assimilate and apply new knowledge. Individuals display different degrees of capabilities to acquire and process information based on their cognitive capacity. These variations are dependent upon individuals exposure to various knowledge bases, their cognitive capacities and their ability to relate, recall and process information. Organizations also display varying degrees of capability to absorb information, and a irms absorptive capacity also depends on its prior knowledge (Cohen & Levinthal, 1990). The learning process assists the internalization of both new and externally situated knowledge competencies and improves the thought and behavioral quality of individuals and organizations. A irms knowledge base would therefore require a learning process that depends on direction rather than passive experiences. In this way, organizational action can be improved through speciic knowledge applications and deployments.
Individual vs. Group Learning

Individuals have the freedom to learn by choosing various formal and informal paths of learning and interactions. They have to take the initiative to learn and take responsibility for increasing their learning capacity, as well as teaching others to learn (Cowan, 1995). Individual learning in a leading computer systems company, for example, is organized in the form of individual contracts. Individuals are expected to perform a task at a particular level, with or without formal training. It is therefore expected that individuals will teach themselves to perform the task and that a superior will help guide employees by giving instructions and specifying procedures. Alternately, a superior may ask for a sub task to be performed by individuals who do not possess the total knowledge required to perform the entire task. Those who are quick to grasp all the components required to perform the task successfully will become the key people to take control and responsibility of the task. Group discussion is regarded as the best method of learning among mature adults (Knowles, 1990). Many social fora, associations and social clubs provide an avenue for the exchange of ideas. In most countries, the village shops or tea and coffee boutiques have been a meeting place where the exchange of ideas and information takes place. Many countries have adopted common meeting squares: for example, Stora Torget in many Scandinavian countries and Sunday markets in European and Asian countries are unique places for the exchange of goods, services, ideas and social interactions. Knowledge and learning are intrinsically connected, and knowledge grows when there is an effective action (Burton-Jones, 1999; Liyanage & Jones, 2002). Organizational learning is therefore considered a process of improving organizational action through better knowledge and business understanding (Garvin, 2000). Like knowledge, learning can also be treated as part of a irms absorptive capacity; learning provides a irm with a broad set of skills that enables it to deal with explicit and tacit components of internally and externally situated
26 Liyanage
knowledge. The connection between learning and knowledge is important, as they unite to provide a composite organizational competency.
Life-Long and Situated Learning

As long as a irm is confronted with competition and change, learning never stops. Cessation of learning means the denial of the advancement of scientiic and technological knowledge. Some traditional societies have opted to resist change and continue to maintain their current cultural identity. The ability to internalize knowledge and learning provides a irm with varying capacities to respond to impending change processes. The role of learning in the knowledge process can be analyzed from two perspectives: a situated organizational learning and a social learning perspective. This situated organizational learning perspective views knowledge as embedded in individuals, formed by connections between individuals, social groups and artifacts (Edmondson, 2002; Winter 2000). Knowledge can also be situated within certain social and organizational contexts and embedded in certain social practices (Lave & Wenger, 1991). Cumulative experiences lead to a collective individual knowledge, which is primarily based on an individuals ability to deconstruct knowledge and relate it to various situations in life. Often such learning and knowledge bases are inert. Learning untangles the tacit knowledge that resides in individuals and societies in order to enrich the common knowledge base. However, the value of such knowledge depends on each irms absorptive capacity, which will determine the similarities between the routines and knowledge bases of irms (Lane & Lubatkin, 1998). Remote societies in India and southeast Asian countries provide ample evidence of the embedded structures of learning where traditions, cultural norms and religious rituals dominate what is possible to learn and practice in society. Learning from master craftsmen is usually structured according to such cultural habits. Thus, learning is a conduit that transmits knowledge from internal and external resources. The situated organizational learning perspective argues that learning is an ongoing activity carried out by individuals. Knowledge created by this learning process is embedded in both the minds of individuals and the individuals environment, which becomes structured as a result of this activity (Nidumolu et al., 2001). In the learning process, an individual may opt to discuss, relate, inquire and interpret information according to his or her own understanding and experiences. This process may, therefore, involve persuasion, building trust and faith, antagonism and a rejection of ideas and facts. In this way, individuals make sense of ambiguous events (March & Olsen, 1975). Different experiences and learning opportunities allow and restrict the evaluation and adoption of different types of knowledge. For example, those who use one e-mail system may be restricted by the organizational rules and procedures of using that particular system, which will negate the use of another system, although it may be far superior to the existing one. Different kinds of knowledge are available through various sources and interactions that could be communally based. These communal processes often stem from social sense making (Wieck, 1977) and collaborative inquires (Argyris & Schoen, 1978). In all these, the learner is the central focus. Different events, procedures, techniques and ways of doing things are
embedded in a given setting that provides the learner with both speciic instructions and details about the nature of the problem and the solutions that aid in resolving the problem. In this sense, the learner is situated within a particular event or action where learning actually takes place. The learner is bound or constrained by the external environment in which he or she is located and which determines what he or she can learn, how much, and in what ways. This situated organizational learning perspective lends support to the continuous build-up and transformation of knowledge. From a knowledge absorption viewpoint, the other important aspects of learning are the negotiations, interactions and collaborations that take place. These are facilitated by knowledge interchanges with external environments (Millar et al., 1997). Such interchanges allow the mixing and aligning of contextually situated knowledge with a irms internal structures of knowledge. Knowledge interchanges can be achieved through appropriate interactions that are often supported by collaborative diagnoses of information and knowledge (Cicourel, 1990). This, however, needs to be interpreted cautiously. Glynn et al. (1994) argue that organizational learning is neither strictly micro nor macro in nature; instead, it involves a complex interplay between individuals, work units and the overall business processes of an organization. Learning processes, like knowledge, require management and guided development. They also involve a process of knowledge absorption within individuals and organizations. Such an absorptive capacity determines a irms level of knowledge and learning integration. Excessive learning will not necessarily result in a corresponding increase in knowledge. The connection between knowledge and learning therefore requires deeper explorations. In particular, adaptive learning takes place with constant review and evaluation. For example, in response to a negative feedback or remark, an individual may respond positively or negatively and either change their behavior or attempt to overcome the negative remarks. The spirit in which such adaptive learning takes place will depend on the way and manner in which such feedback is conveyed to individuals. For adaptive learning, several intervening processes between learning, knowledge creation and absorption have been identiied (Boden, 1990; Ruggles, 1998). They include: the identiication of the stock and structure of knowledge available to the organization so knowledge gaps can be made known and remedied; the decontextualization of knowledge into explicit forms so as to enable its communication and transmission; the synthesis and assimilation of knowledge interchanges by a multitude of users; the facilitation of knowledge generation and production; and the synthesis, adaptation and transformation of knowledge to generate novel and creative uses. Some of these processes support continuous learning whereas others involve incremental learning through the accumulation of various knowledge and skills. The changes to the rate and direction of knowledge assimilation are therefore inluenced by such learning processes. Learning is induced by changing the various components of knowledge. In this way, knowledge typologies can be varied according to information and data (explicit forms), allowing a variety of learning possibilities. Learning also allows the translation of tacit knowledge or experiential knowledge into related business routines. For example, an apprentice working with a master craftsman learns by observing (with or without interaction) and behavior through trial and error. Learning can therefore be seen as the progressive unlocking of the tacit components of knowledge and the internalization of such knowledge. Firms may also learn through deliberate mechanisms of collaboration, partnerships and alliances. These
28 Liyanage
processes involve continuous interactions. Lundvall (1992) classiied learning as a predominantly interactive process. He argued that learning is socially embedded and could not be understood without accounting for its institutional and cultural contexts. Certainly, all learning processes have a particular cultural context in which the learner is situated.
Industry Response to Learning

Our survey of 140 irms in Australia and New Zealand and selected case studies of industry has revealed a wide variety of learning practices that are closely integrated with the knowledge management practices of the organizations examined. The survey was conducted as a part of knowledge management practices of irms and the innovation practices adopted by leading industry groups. The overall responsiveness to learning is poor: learning is generally considered an individual need rather than an organizational knowledge requirement. Individuals were hired for particular tasks, and they were expected to perform the task at a given knowledge level. In our analysis, we examined the following aspects: learning strategies; knowledge strategies that incorporate learning; management attitudes toward learning; organizational support for learning; and processes of knowledge articulation. Only a handful of irms studied (12%) had well-articulated learning strategies and gave complete freedom for individuals to learn as they saw it. Usually these employers accorded leave and some inancial assistance to staff who initiated learning activities. Some organizations adopted more formal structures and strategies for individual and organizational learning, and created learning partnerships with other organizations. Most irms consider learning and knowledge as two distinctly separate activities. A majority of the managers interviewed held the view that learning and knowledge were related, although distinctly different functions for the organization. Learning and training activities were conducted as part of human resource development policies and were carried out more frequently due to government regulations and requirements, whereas knowledge strategies were conducted in response to organizational crises. The management attitude toward conducting learning varies enormously with some viewing training as a cost and others considering it to be a necessary investment. There is, however, recognition that it is corporate responsibility and senior managers duty to ensure subordinates are adequately coached and trained to undertake certain tasks. When engaged in learning, irms tend to actively observe and imitate the business routines of other irms. In this way, a irms products, processes and services are benchmarked. We identiied the following patterns of learning in most organizations: a. Learning in response to compulsory skills and knowledge gains at no cost to the organization [learning by experience and engagement]: A majority of irms surveyed (42%) developed competencies management systems based on voluntary learning efforts where employees either have to acquire skills at their own cost or lose the prospect of employment.
b.
Learning as a result of a particular strategy- for example, a knowledge or innovation strategy, which comes at a cost to an organization [directed learning]: A small number of irms (14%) encouraged specialized tasks and appreciated innovative skills. Most of these companies were managed by innovators themselves. Learning as a result of dynamic capability building at a cost to an organization [high-level cognitive learning]: A signiicant number of irms (28%especially medium to large-size irms) had formal learning systems that develop a distinct learning culture. Learning as a result of task sharing and team building at no cost to an organization [interactive learning]: Several irms (16%) developed informal mechanisms for the exchange of knowledge and learning, and have provided encouragement in the way of building learning and innovation cultures in the organization.
c.
d.
However, intra-organizational learning often takes place with the maturation of a particular knowledge and as a result of technology transfers (Crossan & Inkpen, 1995). Learning was strong among the business units of some irms, especially in cases where knowledge was passed on from one project to another, and also in instances where collective action was needed for speciic knowledge applications. Although managers emphasized the development of knowledge systems, learning was generally relegated to individuals. The managers also strongly supported effective decision-making and responses that required inter-organizational events. This would link participants with their practical objectives and satisfy their discursive relevance criteria. Learning is essentially grounded within a social context where the need to know is based on an employees fear of obsolescence and future requirements. The social grounding of learning entails different knowledge perspectives and heterogeneous knowledge inputs. A signiicant number of employees interviewed undertook higher theoretical learning, which had some relevance to the organization but was not entirely in line with the core business of the irm. Not all irms are equally prepared to engage in the acquisition of new knowledge and integrating such knowledge into already established organizational routines. The attempt to identify an organizations available knowledge and its role in assimilating new knowledge has been recently explored through the concepts of a irms absorptive capacity and prior knowledge (Zahara & George, 2002). In the process of acquiring new knowledge and absorbing such knowledge to renew prior knowledge, irms display varying degrees of agility toward developing learning capabilities. All irms engage in some form of learning. Learning is an essential process in the conversion of knowledge into intellectual capital (Argyris & Schoen, 1996; Senge 1993). Our research indicates that different irms respond differently to developing knowledge and learning capabilities. These responses are usually determined by government policy and production. Deep-seated cultures of production in different countries do not fully realize that learning builds different types of capabilities which allow the consolidation of certain business routines. Our research also suggests that those irms that were able to institutionalize learning processes tended to draw on the knowledge base of individuals by empowering them to make decisions. Learning increases the culture of knowledge sharing, which is, in turn, important for selecting and developing knowledge-based competencies.
30 Liyanage
Styles of Learning
Theories of learning have so far focused on learning as an exogenous process to knowledge development, absorption and utilization. This view, however, is inadequate to explain the way people actually generate new knowledge and contribute to prior and future knowledge. Types of learning undertaken by individuals vary according to the domain of learning that guides each individual to design and deliver certain types of knowledge. Some knowledge types remain in a cognitive domain whereas others are in an empirical domain. Ways of learning are comprised of the direct transmission of information and knowledge, direct acquisition of knowledge by the learner from available and potential sources, observation and engagement and research to develop new patterns, designs and new ways of doing things (Figure 3). The learner encounters several types of knowledge sources and should be in a position to evaluate and access the relevance and usefulness of information iltered through knowledge diffusion processes. The learner is also confronted with a choice of formal or informal learning through engagement with reliable knowledge sources, as well as through knowledge-sharing behaviors in the organization and other forms of knowledge sources. The delineation of areas of study in organizational research into the categories of knowledge and learning has marginalized the actual role played by individual learning and relection. In fact, all organizations implementing knowledge management systems have to carefully consider the discursive rules that delineate knowledge from learning. Because of its pervasive role, learning needs to be analyzed independently but also in conjunction with the contingent variables related to knowledge development, assimilation and absorption. The recognition of inluential variables that lead to a greater and more informative link between
Figure 3. Components of learning process
Reliable Knowledge
Diffusion
Learner
Knowledge Construction
Knowledge Sharing Behaviour
Experiential Knowledge And Action Other forms of Knowledge
learning and knowledge processes is required. Certainly, learning theory must be integrated with knowledge and organizational theories. Experiential learning integrates individual's intellegence and creativity with new forms of intelectual capacity and creative performance (Amabile, 1996). Several learning processes in organizations have been identiied, as seen in Malecki (1997). These processes can be subdivided into the following: Task- and product-related: Learning by doing: Conduct a speciic task and learn from it. Learning by using: Use a particular tool or method and learn to do it. Learning by operation: Changing; system performance feedback; training; hiring; searchingengage in speciic function in a production system and learn from it.
Group- or community-related: Learning by trying: Undertake a task that has not been done before and learn from it. Learning by interacting: Interact with and observe someone who knows to carry out a function and learn from it. Learning by selling: Learn by engaging in the exchange of goods and services and know how to deal with others and your own product.
Individual-related: Learning by imitation: Learn to perform a task by observing and imitating exactly how it was done. Learning by failing: Learn by understanding the root cause for failure and learn how to avoid mistakes that leads to failures.
Although some of these categories can be interrelated, essentially they provide problembased learning situations that demand certain knowledge and competence. Our research shows that irms use all these forms of knowledge. However, direct learning is the most commonly practiced form advocated by irms. Near learning refers to the immediate learning environment, which compensates for immediate knowledge and competitive gaps. Most of such learning processes attempt to understand current systems and their operation. Learning by interaction is largely grounded in personal networks. Learning by transfer occurs as knowledge is transformed into intellectual capital and assimilated into various outcomes. Sources of learning for a irm include: clients or customers; other irms within the irms group; fairs and exhibitions; suppliers; competitors; professional conferences and journals; universities and higher education institutions; computer-based information networks; consultancy irms; government and non-proit institutions; and patent disclosures (OECD, 2000). However there are some distinctions that can be made about learning that depend on the different means by which information can be transmitted; for example, using codiied
32 Liyanage
or tacit knowledge. The formation of networks or industry clusters is helpful for integrating learning through the transfer of both codiied and tacit knowledge within organizations.
Technologies for Types of Learning

Various tools have been used to measure the effectiveness of learning and learning performance (Watkins & Marsick, 2003). Over recent years, the work force, the way we work and how we deal with job security have all changed dramatically. Most young people tend to change their jobs regularly, breaking the traditional system of secured employment. Such changes have also impacted on the way learning takes place among individuals, groups and organizations. The movement from traditional teaching methods to new ways of learning and teaching are necessary for providing state-of-the-art knowledge in the management of ever-changing technological landscapes and industry structures. Electronic media have introduced new ways of training and learning by providing accessibility, lexibility and relevance of learning. E-learning is an Internet-related solution to organizational learning. Content is delivered in multiple formats and combined with learning experiences that are connected via a network of practitioners, learners and scholars. E-learning methods hope to provide faster and more effective solutions to the learning needs of individuals and organizations by acting as a conduit to stimulate a learning space for the learner, teacher and instructor. The development of the effectiveness of e-learning methods requires extensive infrastructure and support facilities such as telecommunications, access to the Internet, development of course material, and the implementation of e-learning portals. In addition, it is necessary to supplement learning with real-life experiences in order to make learning relevant and appropriate. One of the dificulties of e-learning is the selection of methods and approaches that are suitable for a variety of people who have different experiences and learning capabilities. E-learning tools are particularly appropriate for relaying information for scrutinization in real-time. Today, all irms disseminate information related to products, processes and services through intranet and Internet technologies. E-learning platforms provide fast access to: a. b. c. Basic learning, which focuses on building effective, problem-deining and problemsolving competencies; Higher-level learning, which focuses on knowledge and skills conversion for subsequent actions; Meta-level learning, which focuses on collective learning: the subjects taught allow greater interaction through activities, such as research projects and bulletin board discussions, which connect students, staff and industrialists, thereby encouraging organizational learning. Typically, this involves relecting, probing, formulating questions and hypotheses, evaluating and synthesising information.
Figure 4 illustrates the basic skills attempted in a typical technology management training program that combines cognitive, interpretative, analytical and collaborative skills. E-learnCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
Figure 4. Basic learning skills needed for e-learning

Cognitive Skills
Problem solving Capacity ca pacity Creative and lateral thinking Challenging and argumentative Organising Organii ng and discourse
Interpretative Skills
Strategy and Business analysis business analysis Value chain analysis Benchmarking Re-engineering
E.g.. idea, concepts and metaphors
E.g. lifecycle and industry structure
Analytical Skills
Financial analysis Valuation Costing Risk management
Collaborative /Social Skills Team building Leadership Communication

E g Networking , collaboration and relationship
E.g. project and operations
ing programs have the capacity to elevate management education to a meta-level. In their study of competitive advantage, Lei, Michael and Bettis (1996) proposed a model where the development and outcomes of core competencies are based on organizational meta-learning. Dynamic core competencies depend on the integration of universal and tacit knowledge into systemic meta-learning through information transfers, redeinitions of heuristics and continuous improvements. The understanding of diverse, complex and interlinked technological and non-technological tasks requires meta-learning capabilities. These skills may be dificult to explain and impart using e-learning platforms alone. Because of lexibility in delivery, ease of access to otherwise dificult information, available time for relection, and the evaluation of learning objectives, e-learning methods allow various strategies of learning that are not possible in traditional learning processes. One of the powerful features of e-learning lies in its integration of content with the instructor and student simultaneously available in a single learning platform. The roles of instructor and student, however, are still maintained through various deined activities, including: the instructors role in assessment; support and communication patterns; the selection of cases; and the overall context of learning. The student has a different role to play. This involves: a commitment to learning; enthusiasm for embracing new methods of learning; a disciplined approach toward enhancing communication patterns; and an openness to creativity-enhancing techniques. The implementation of e-learning methods and their evaluations suggest that students need at least three to four weeks to read suficient material before commencing dialogue with teachers and instructors in order for meaningful learning to occur. A typical interactive e-learning program includes components such as extended course outlines, detailed course contents, glossaries, references, case studies, lecture notes, bulletin boards, e-mails, online tutor interactions, chat rooms, class lists, assessment drop boxes, grades and feedback, calendars, navigational tools, bookmarks, search tools, compilers, resources and synchronized collaborative tools. For example, the operation of a chat room will entirely depend on course instructors and lecturers. In some cases, it is possible to create student peer groups to initiate and carry out an online discussion on a topic of their choice.
34 Liyanage
The development of online courses requires instructional support so that readers can be suficiently motivated to engage with explicit material. The selection of the right type of questions, probes, diagrams and teasers to prompt responses is necessary. Such a selection needs to be situation-speciic, current and relevant to a students cognitive environment. It should be suficiently simple and illustrative, so that the underlying meanings can be deconstructed.
Assessment and Evaluation of Learning

The assessment process is vitally important in online learning. The proper assessment tools for a learners commitment; the correct choice of course structures; relevance of courseware; effective communication channels and methods; suficient lexibility in learning environment; reliable and cost-effective delivery modes; and suficient interaction and feedback are all essential for effective learning. The commitment and participation of the learner is a critical success factor. E-learning requires specialist support from curriculum designers, information specialists and interactive learning development teams. The course content needs to be transformed to digital formats, and various methods of information delivery are required to provide access to pertinent material and teaching aids. Various management structures are necessary to offer adequate support to those engaging in e-learning methods. Vast resources are needed to provide instructional design, Web development and information management, and learner access to course material. In addition to this logistical support, the learning also has to take place between the teacher and learner. The interaction between teacher and learner commences with building mutual trust and respect; maintaining certain protocols of communication; iltering and sharing relevant information; building practical examples and case studies for real-time learning; developing interactive exercises to engage the learner; developing teamwork for online collaborations; and encouraging community and networking. A series of activities needs to take place in an online environment. The actual implementation of Internet technology is proven to be dificult because of the complexity of certain materials that cannot always be transferred across digital media. The teacher needs to provide facilities to communicate complex ideas, concepts and relationships in a virtual environment. Various metaphors, maps, illustrations and narratives have to be carefully structured and selected. The choice of language and semiotics often determines what can be communicated effectively. There are, however, distinct advantages of connecting the learner with a variety of learning possibilities through e-learning technologies. Questioning and reconstructing existing theories and concepts, interpreting models and frameworks building decision premises are essential in interactive learning. E-learning technology allows a mixing and matching of skills and capabilities by bringing together various learner groups, resource persons, libraries and other resources. It has also assisted in broadening the focus of study from a narrow irms perspective to trans-organizational systems of study. E-learning, however, is not an effective tool to impart tacit knowledge. Although Internet technology has assisted teachers and students by providing a wide range of options, such technologies will not be able to completely replace the practical and experiential learning in some ields of learning. The most dificult task is converting content into various learning experiences that allow effective
individual and organizational learning to take place. These experiences typically involve intense planning; the explication of learning objectives; an in-depth understanding of theory and their relevance to practice; identiication of key metaphors; the building of learning maps; usage of effective and economic narratives for communication; and dedication and commitment from people willing to change learning and teaching cultures. One of the most important sought-after resources for e-learning is an electronic library. This is required to provide a continuous supplement of information technology support. The learner should have lexibility to develop critical thinking skills. Most learners are used to conventional learning techniques, such as listening to lectures, and ind it extremely dificult to adapt to a new learning space that requires new ways of learning and interacting. The new environment provides faceless communication that can be culturally dificult, and the learner needs to have a good command of language. Without such knowledge, the learner may misunderstand the scope of e-learning, method of assessment and learning objectives. Various tools, such as the bulletin boards, e-mails and chat rooms, have been developed to provide multiple sources of communication and conversation. In an interactive e-learning environment, there is also the danger of over supplying material. A mishmash of unrelated knowledge bases makes it dificult to form a coherent system of learning. In a fast-changing knowledge economy, electronic learning can be an asset, as it can bring real-time information to the most remote corners of the planet. However, e-learning processes need to be managed carefully so that knowledge bases can be continually enriched, veriied, and subjected to scrutiny by both the learner and instructor. The cultural dimensions of the learner and instructor are another critical factor affecting e-learning. Communication protocols need to be established before any effective interaction can take place. Some learners are more comfortable using Internet technologies than others. Due to the constricted learning space, developing teams and groups will be dificult, although some tools, such as simulation games and collaborative projects, can be used to overcome this and build collaborative learning. Our research shows that e-learning development and implementation in a majority of academic institutions suffers from deviation from core competency structures. There had been a tendency to develop home-grown products was noted, which proved too cumbersome to use and partially or inadequately developed to deal with learning processes. Much effort of the academics was spent on ixing inadequacies of the system or working with underdeveloped systems rather than being engaged in actual learning processes. Some of the learning platforms failed to conform to basic IT protocols, such as ease-to-use, reliability, accessibility, simplicity and interactiveness. For example, in a recent implementation of an online text, one of the administrators of the system had to apologise to students and instructors: We understand that there was a problem with online test over the weekend and possibly this morning as well. The problem was most likely to be caused by one of the server playing up, as most of the errors were coming from one server. This caused a lack of faith in the system as well as a poor learning experience for student, in addition to the stress of running text again by instructors. The experiences in e-learning of learners and instructors revealed that:
36 Liyanage
Flexibility of online learning is a distinct attraction for undertaking learning. Not all students were comfortable with the learning experience of an online environment. The learner often needed scaffolding and mentoring to synchronize the learning process. This is a time-consuming process that often requires extra support from tutors. Consistency and quality of communication, interaction and relationship building were major stumbling blocks in implementing e-learning technology. Suficient interactions, leadership and role-playing by instructors were necessary for stimulating learning and soliciting the participation of the learner. The most demanding tasks were the proper assessment of learning processes and providing quality and timely feedback to students assessments and ideas.
For many, the development of an online component enhanced the practical relevance of the learning experience and made collaborative case studies and teamwork possible due to online tools. Such technology also provided a self-managed system for implementing teamwork. On the down side, there were some dificulties in ensuring that the learner could relect on and understand concepts in an online environment. This was, however, due more to design problems of the curricula rather than the techno-managerial issues. The development of an online technology management program addresses the dynamism of management education: what is taught today may not necessarily be relevant for tomorrows businesses. The available information technology techniques and tools provided smarter development of learning and knowledge objectives and the design of experiential learning tools, such as simulations and interactive games, developed content that allowed for high retention through visualization and powerful images and metaphors. In terms of delivery and diffusion, various plug-and-play devices and supplementary material, such as videos and audios, can be provided as scaffolding material for slow learners. These technologies also assist teachers to focus on learner-related and instructor-related interfaces, thereby providing ample opportunity for pedagogical development of learning strategies.
Summary and Conclusion

Learning, like knowledge, empowers individuals and organizations. It enables people to access knowledge and to think beyond the act of learning to do what, why and how. Learning cannot and should not be left alone to an individuals responsibility. It requires careful guidance, organizational support, management competencies, and technologies that encourage individuals and organizations to learn and unlearn. In this respect, there is a fundamental question that begs an answer. What are the management competencies and technologies that assist in individual and organizational learning?
The ways in which people work and relate have undergone considerable changes and reforms. The changes to working habits have shifted from the traditional employer-centered to employee-centered working situations. These changes warrant new ways of managing learning among individuals and organizations. There is, however, limited knowledge of how effective competency management systems and technologies are in individual and organizational learning. The challenge is to provide an effective learning environment that acts as an investment for building the current and future competencies in an organization. Knowledge sharing among individuals and organizations has taken a new twist in the knowledge economy since various management structures and technologies are now available for effective learning. Two important parameters come into play in managing learning processes: humanistic and behavioral and procedural aspects of learning need to be recognized and appreciated in these novel organizational structures and processes. While traditional management structures have been quite averse to organized learning in irms, the recognition of learning as an important tool for knowledge sharing and diffusion is essential. In organizational management studies, learning has been treated as an intervening process. However, it is still often treated as exogenous to the processes of a irms knowledge creation, assimilation and diffusion processes. We regard learning as endogenous to knowledge processes and argue that it forms an integral part of knowledge development, knowledge absorption and interchanges. In the formation of knowledge and intellectual capital, learning plays a critical role, especially in illing out the knowledge gaps of individuals, teams and systems. Learning also facilitates much-needed behavioral changes that allow the appreciation of intellectual capital components residing in individuals. In this vein, building an organizational culture that appreciates learning is a fundamental building block of management competencies that needs to be installed in the organization. Nowadays, modern organizations arrange their production in different ways, with most irms focusing on core business routines while subcontracting or outsourcing non-core activities. As a result, employees have to unlearn some skills and learn new skills to meet organizational demands. A new perspective is required for linking knowledge and learning theories as a cohesive set of inputs to the creation of an organizations knowledge capital. Viewing these as separate activities introduces new dynamics of managing e-learning among individuals and organizations. Learning proceeds with the increase in a irms absorptive capacity; this in turn, leads to the assimilation of new knowledge with prior knowledge. A irms absorptive capacity can be expressed in terms of its learning capability, which is able to advance the knowledge capital of a irm. Because of the rapid adoption of new technologies and the growing industrial economies of countries such India and China, there are very real issues in dealing with the education of the workforce and in providing learning opportunities that assist the continuous improvement to organizational productivity. The managerial competencies need to evolve in order to recognize the learners cognitive needs and behaviour, and to assist the learner with greater retention capability and the ability to recall and use what has been learned in the current business practices. Such learning can be provided with the intervention of new technologies capable of rapid design, development and delivery of new knowledge. While a irm is engaged in building its speciic dynamic capabilities, learning should be treated equally or similarly to knowledge management. In modern irms, e-learning provides an effective
38 Liyanage
tool for organizational learning. However, it needs to be managed carefully and structured according to the individuals cognitive needs and the organizations development needs. As the learning concepts become more complex, learning technologies need to improve in terms of multiple functionality and increased accessibility and interactions, at relatively low and affordable costs.
References
Amabile, T. M. (1996). Creativity in context. New York: Westview Press. Argyris, C., & Schoen, D. A. (1974). Theory in practice. Increasing professional effectiveness. San Francisco: Jossey-Bass. Argyris, C., & Schoen, D. A. (1978). Organizational learning: A theory of action perspective. Reading, MA: Addison-Wesley. Argyris, C., & Schoen, D. A. (1996). Organizational learning II: Theory, method and practice. Reading, MA: Addison-Wesley. Boden, M. (1990). The creative mind: Myths and mechanisms. London: Weidenfeld and Nicholson. Bourdieu, P. (2004). Science of science and relexivity. Oxford: Polity. Brown, J. S., & Duguid, P. (1991). Organizational learning and communities of practice: Toward a uniied view of working, learning and innovation. Organization Science, 2(1), 40-57. Burton-Jones, A. (1999). Knowledge capitalism. Oxford & New York: Oxford University Press. Carroll, J. S., Rudolph, J.W., et al. (2002). Learning from experience in high-hazard organizations. Research in Organizational Behavior, 24, 87-137. Christensen C., Suarez, F., et al. (1998). Strategies for survival in fast changing industries. Management Science, 44(12), 207-220. Galegher, & Cicourel, A. (1990). The integration of distributed knowledge in collaborative medical diagnosis. In G. J., R. Kraut, & C. Egido (Eds.), Intellectual teamwork: Social and technological foundations of cooperative work. (pp. 21-242) Hillsdale, NJ: Lawrence Erlbaum Associates. Cockburn, I., Henderson, R. M., et al. (2000). Untangling the origins of competitive advantage. Strategic Management Journal, 21, 1123-1145. Cohen, W. D., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128-152. Cowan, D. A. (1995). Rhythms of learning: Patterns that bridge individuals and organizations. Journal of Management Inquiry, 4(3), 222-246. Crossan, M. M., & Inkpen, A. C. (1995). The subtle art of learning through alliances. Business Quarterly, 60(2), 68-78.
Daft, R. L., & Lewin, A. Y. (1993). Where are the theories of the new organizational forms? An editorial essay. Organization Science, 4(4), iv-vi. Dodgson, M. (1993) Organizational learning- a review of some literatures, Organization Studies 14(3), 375-394. Easterby-Smith, Snell, M., R., et al. (1998). Organizational learning: Diverging communities of practice? Management Learning, 29, 259-272. ECITB (2006). Guide to the training levy, grants and training programme, 2006. E.C.I.T. Board. Hertfordshire, Engineering Construction Industry Training Board, UK, 1-23. Edmondson, A. C. (2002). The local and variegated nature of learning in organizations; A group-level perspective. Organization Science 13(2), 128-146. Garvin, D. (2000). Learning in action. Boston: Harvard Business School Press. Garvin, D. A. (1993). Building a learning organization. Harvard Business Review, 71(JulyAugust), 78-91. Gherardi, S. (1999). Learning as problem-driven or learning in the face of mystery? Organization Studies, 20(1), 101-124. Glynn, M. A., Lant, T. K., et al. (1994). Mapping learning processes in organizations: A multi-level framework linking learning and organizing. Advances in Managerial Cognition and Organizational Information Processing, 5, 43-83. Inkpen, A. C. (1998). Learning and knowledge acquisition through international strategic alliances. Academy of Management Executive, 12(4), 69-80. Inkpen, A., & Dinur, A. (1998). Knowledge management processes and international joint ventures. Organization Science, 9(4), 454-467. Kant, I. (1933). Critique of pure reason. London: Macmillan. Knowles, M. (1990). The adult learner: A neglected species. Houston, TX: Gulf Publishing Company. Kolb, D.A. (1984). Experiential Learning: Experience as the Source of Learning and Development. Englewood Cliffs NJ, Prentice-Hall. Lane, P. J., & Lubatkin, M. (1998). Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19, 461-467. Lave, J.. & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. New York: Cambridge University Press. Lei, D. H., Michael, A., & Bettis, R. (1996). Dynamic core competencies through metalearning and strategic context. Journal of Management, 22(4), 549-569. Liyanage, S., & Jones, A. J. (2002). Investing in knowledge capital. Singapore: Singapore Institute of Management. Lundvall, B. (1992). A national systems of innovation: Toward a theory of innovation and interactive learning. London: Pinter. Malecki, E. J. (1997). Technology and economic development. Harlow, UK: Addison-Wesley Longman.
40 Liyanage
March, J., & Olsen, J. P. (1975). The uncertainty of the past: Organizational learning under ambiguity. European Journal of Political Research, 3, 147-171. Millar, J., Demaind., A., et al. (1997). Transorganizational innovation: A framework for research. Technology Analysis & Strategic Management, 9(4), 399-418. Nidumolu, S. R., Subramani, M., et al. (2001). Situated learning and the situated knowledge web: Exploring the ground beneath knowledge management. Journal of Management Information Systems 18(1), 115-150. Organization for Economic Co-operation and Development (OECD). (2000). Knowledge management in the learning economy. Education and Skills. Paris, OECD. Ruggles, R. (1998). The state of the notion: Knowledge management in practice. California Management Review, 40(3), 80-89. Sanchez, R., & Heene, A. (Eds.). (1996). Strategic learning and knowledge management. Chichester, UK: Wiley. Senge, P. M. (1993). The ifth discipline: The art and practise of the learning organization. New York: Doubleday/Currency. Simon, H. A. (1987). Satisicing. In Millgate, M., Eatwell J., & Newman, P. (Eds.), The new palgrave: A dictionary of economics. (pp. 243-245). New York: Stockton Press. Simon, H. A. (1991). Bounded rationality and organizational learning. Organization Science, 2(1), 125-134. Watkins, K. E., & Marsick, V. J. (2003). Demonstrating the value of an organizations learning culture: The dimensions of the learning organization questionnaire. Advances in Developing Human Resources, 5(2), 132-151. Wieck, K. E. (1977). Organization design: Organizations as self-designing systems. Organizational Dynamics, 6, 31-46. Winter, S. (2000). The satisicing principle in capability learning. Strategic Management Journal, 21, 981-996. Winter, S., & Szulanski, G. (2000). Replication as strategy. Organization Science, 12, 730-743. Zahara, A. S., & George, G. (2002). Absorptive capacity: a review, reconconceptualization, and extension. Academy of Management Review, 27(2), 185-203. Zollo, M., & Winter, S. G. (2002). Deliberate learning and the evolution of dynamic capabilities. Organization Science, 13(3), 339-351.
A Survey of Competency Management Software Information Systems 41
Chapter III
A Survey of Competency Management Software Information Systems in the Framework of Human Resources Management
Alfonso Urquiza, Francisco de Vitoria University, Spain
Abstract
One of the greatest surprises of the Internet economy is that far from replacing people, the use of advanced technology is conirming that talent is the most valuable asset in todays organizations. In this context, competency management (CM) software automation practices become the most valuable business approach to deine, measure, and manage talent needs, the human capital of the organization.This chapters position is that CM process automation in competitive, knowledge-intensive e-business oriented organizations requires that information technology (IT) address software strategy in a comprehensive human resources management (HRM) framework. Core competency management-related applications are deployed in current corporate e-business transformation processes in association with the use of innovative employeefacing relations management technology and reengineering most HR transactional domain type of applications in place.The chapter shows the CM software evolution from a previous fragmented market situation to a much more integrated scenario in which best-of-breed single-function oriented products preferences are now swiftly moving to the enterprise resource planning (ERP) type of architecture.
42 Urquiza
Introduction
As the Internet age transforms the way people work and live, organizations keep continuously embracing the new opportunities and challenges generated by this relatively recent and signiicant change, introducing a new knowledge revolution (Nordstrom & Ridderstrale, 2000). Todays economy is creating a new breed of intelligent organizations, where a very high percentage of the total workforce is comprised of knowledge workers. In this context, the ability to effectively manage human capital investments becomes essential to ensure business success. Organizations gain real advantages by applying Internet technology to the measurement and management of their talent needs, the human capital of the organization. The long transition from traditional personnel administration activities to most recent human resources (HR) management, has meant an evolution from a purely functional to a process-oriented approach in which all those activities associated with the management of employment and work relations are included (Boxall & Purcel, 2003). Traditionally, organizations of any size or activity used to focus primary attention on automating payroll & basic administrative functions. Other administrative-required functions (like recruiting, training, etc.) were largely assumed and performed in a non-automated way, thus creating large staff departmental units in these areas, non-associated with the organizations primary business. IT solutions at the time were not designed to manage knowledge assets; they were focused on managing physical assets. Individual employees and managers used to call upon HR to satisfy different kinds of requests. Individual employees typically required tracking and processing personal information, such as compensation, beneits, or other related data. Managers required HR to provide information on recruitment or training services. In recent HR management, the new e-business context has transformed and automated most HR operations, thus generating additional eficiency: process lows are handled like automated transactions and self-service functions appear, simplifying individual employee/manager relations within the organisation, automating administrative tasks and enhancing task-driven routines formerly performed by HR departments. In the new human capital (HC) paradigm, it is not just about modeling and automating tactical HR functions. Two new dimensions are introduced. The irst to consider is that a new role appears for HR: that of strategic asset (talent) management. The second is that HR becomes just another component in the organisation, like inancial management, supply chain management, customer relations management, or IT, all of them driven to produce a product or service that generates value to the customer (Laudon, 2004). It is in this new, comprehensive management context where competency management fully develops itself, becoming the integrating glue element in HC management systems (Sagi-Vela, 2004), thus reshaping todays and future HR management implementation strategies. In the context of this work, competence is understood as the set of knowledge, skills, and attitudes required in people to perform a speciic task in an eficient way (Sagi-Vela, 2004). CM is a comprehensive HR process that starts by deining the required organizational competencies, assigns them to employees, observes them through behaviour, asses them
according to an organizations deined values, and permanently improves them (Levy-Leboyer, 1997). Unlike in traditional transaction-oriented HR practices, a CM strategy should pursue the following goals: Support business objectives, providing information to acquire, maintain, inluence, develop, and retain the right employees. Align people, processes, and technology around shared values. Measure the strategic value of human capital investments. Anticipate human capital changes. Learn from industry-best practices, leveraging benchmark data.
In todays knowledge-intensive organizations, competency management is not viewed just like a new function associated with a single job topic (learning, employee career development, etc.), nor is it an additional responsibility, to be added to the traditional list of activities that the organization expects to be accomplished by the HR department. Experience shows that effective CM strategies succeed most when all HR processes in operationnot just a portion of them, like career development, performance management, or learning, for exampleare reviewed and aligned to the talent management vision, accomplishing a real e-business transformation in organizations processes. This remains true even in situations where CM strategies are focused to a limited extent in overall taskforce, applied only to strategic level of employees within the organization (i.e., managers), which happens very often, particularly in large-size organizations. The main objective pursued in this chapter relects this business reality: There is not just a simple, unique solution that automates CM process in an organization. CM is not an IT vendor product; it is a full comprehensive strategy that transforms the HR management function from a mere, although mature, administrative level to a strategic business alignment role. This chapters proposed detailed position is that, when it comes to implementing a CM strategy in a competitive, knowledge-intensive e-business-oriented organization, IT components in the three areas showed in Figure 1 should be addressed (to the extent placed upon deined CM deployment requirements) or sometimes re-engineered as a combined operation: The transactional domain type of applications are not obviously related to CM functionality, but some processes in this area require small re-engineering enhancements to work in a CM driven e-business environment. Functions included in the employee life cycle domain are driven to enhance employee value, core in CM deployment. Employee-facing relations management technology deployment usually exceeds the scope of HR, but the use of such tools signiicantly improves CM-related process eficiency: use of paper and forms-oriented types of activities are replaced by Web-based self service processes, generating signiicant return on investment (ROI) in large organizations and employee satisfaction in online experience. In order to fully understand this chapters business-oriented perspective, the rest of this chapter is structured as follows: The next section shows the background in which this work is based, followed by an analysis of IT industry response to the CM evolutionary organizational requirements. Then, a charCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
44 Urquiza
Figure 1. HC key components in CM deployment strategies

Employee-facing, Domain Employeefacing, Relation Relations Domain
Employee Self-Service
Manager Self-Service
Portal Technology, Expert Finder
Employee Life Cycle Management:Strategic CM-related Employee Life Cycle Management: Strategic CM-related Domain Domain
Career Development Career Development
Competency Mangt. Competency Mangt.
PerformanceMangt. Mangt. Performance
E-Learning E-Learning
E-Recruiting E-Recruiting
Workforce Workforce Planning Planning & & Analytics Analytics
Compensation Compensation
HR Basic Transactional Domain HR Basic Transactional Domain

Reporting Payroll Time Management
acterization of most relevant HR processes is presented, depicting inter-process relations, and supporting and clarifying the proposed comprehensive approach to CM implementations. Finally, todays commercial CM-related technology scenario is presented, followed by future trends and inal conclusions uncovered in this survey.
Background
Competency management is considered by some authors (Sagi-Vela, 2004) as a complementary methodology to other related HR management practices, such as emotional intelligence or, more often, knowledge management. Knowledge management (KM) is a very close concept to CM, and technical literature (Alle, 1997) tends to generate some confusion when analysing KM and CM management systems. In the scope of this work, KM management systems are those directed to capture, analyze, apply, and re-use organizations know-how, with the objective of performing higher-quality business processes at a lower cost and generating a competitive advantage. CM systems focus is on employee life cycle, covering competency requirement analysis (Lindgren & Stenmark, 2002) and other related HR management topics, such as personnel selection and compensation (Sagi-Vela, 2004). Contributions on KM designs and product evaluations are widespread in scientiic literature (Benson & Standing, 2001; Friss, Azpiazu, & Silva, 2004; Kamara, Anumbad, & Carrillo, 2002; Rollet, 2003), but technical literature on CM is still scarce and dedicated to speciic areas, such as organizations adjustment to CM (Lindgren, 2005) or competency development (Hardless, 2005). In addition to above, the software capability maturity model initiatives (CMM I, 2002) have been complemented by the Software Engineering Institute with the people capability maturity model (People CMM) (Curtis, Heley, & Miller, 2001) as the foundation for a model
Figure 2. The Process Areas in People CMM
of best practices in managing an organizations workforce. And competency management is obviously present in various signiicant People-CMM process areas: In this context, a CM software information systems survey in the framework of a comprehensive HR management process scenario might be of great value in further CM software research, for example, investigating compliance relationships between the ive People CMM evolutionary maturity levels with CM market products research and development.
Competency Management and the Software Industry

Todays SW products and services marketplace in the HR domain has been shaped in the last 25 years by the business evolution described above. It is no surprise to anyone that until the e-business transformation with the arrival of the Internet age during the late 90s, HR businesses have created a sort of IT industry fragmented market. There exist hundreds (if not thousands) of different HR administration and management solutions (payroll, learning, PM, CM, etc.) in organizations around the globe. Some are global market-oriented, other vertical cross-industry oriented; most are local and country-driven solutions, aimed to satisfy speciic single-market demands. In addition to the above situation, the amount of in-house developments and custom-made implementation solutions in this ield far exceeds expected igures compared to other types of support applications, such as inancials, business analytics, and so forth. But the above scenario is now changing very fast today, and the e-business evolution is driving industry from the previous IT-fragmented scenario to an extremely concentrated one. The vast majority of todays large organizations (those in which CM has the largest deployment potential) are automating HR (or planning to do so) within the scope of one out of three global, integrated ERP products: SAP, PeopleSoft or Oracle.
46 Urquiza
Figure 3. IT portfolio management methodology
Va lu e / Tim in g
Gr ow t h D iscr e t ion ar y En h a nce m e nts N onD iscr e t ion ar y Cor e
D iscre t ion a ry Proj e ct s
Ve nt ure
Risk s
Tran sfo rm the bu sin ess
Grow t h e bu sin ess Risk

N on - D iscre t ion a ry I T Costs
Ru n t h e bu sin ess
2 0 0 3 META Group, I nc., St a m ford, CT- USA, + 1 ( 2 0 3 ) 9 7 3 -6 7 0 0 , m et a group.com
To better understand this evolution that is reshaping the market in which CM software progress takes place and is making many best of breed stand-alone HCMS solutions disappear, lets apply a modern, innovative approach that is useful to map organizations business requirements with information technology decisions. The Meta Group Technology Analysts call it portfolio management: Independently of the economic environment, a key challenge for IT professionals has been to assess and permanently communicate the value of IT investments to the business units. Todays advanced organizations dont just require a functional justiication and a business case study to make a technology decision. The portfolio management approach introduces inancial issues to any IT investment, as for example: Level of risk versus expected beneits/value: Just like a pure inancial investment. In IT, project deployment time, magnitude of investment, and so forth are taken into account. Current fair value, at any time. Expected life cycle: when will investment pay off? Perceived relationships: deining the IT domain, its platform, and architecture. IT assets and projects are categorized as shown in Figure 3. The run-grow-transform classiication deines the primary goal for any IT project:
Figure 4. IT portfolio classiication example
RTB
GTB
TTB
Bajo
Business Value
4 3 2 1
Alto
0 0
Bajo
1
Alto
Risk
2
Run the Business investments are focused to keep business operational (i.e., maintenance contracts, utilities, etc.). Core spending is for business-critical activities, like customer service, sales ordering, and so forth. Non-discretionary relates to organic growth in core IT assets (servers, DBMS, etc.). Business risk is low and expected reward usually medium to high. Grow the business IT expenditure is applied to expand organizations scope, in product or services. Learning activities to develop new skills fall in this category. Transform the business initiatives are related to opening new markets or issues having a major impact on the current business model. Business risks and expected rewards, in the last two categories, are both moderate to high. Organizations sometimes afford the high risk (of unplanned events) and expect also high reward in venture initiatives, where speed is usually the way to proceed (i.e., be the irst to deliver a product). The IT industry in most recent e-business process transformations (that obviously applies to HR) drives SW product development & projects investment decisions, analysing three dimensions: investment business impact, cost and performance expectations, and risk and opportunity appraisal. There are many SW tools in the market to analyse and manage IT portfolios: ProSight (www. prosight.com/solutions/software/), Primavera (www.primavera.com/about/trillion.asp), Niku (www.niku.com/), and so forth. The example in Figure 3 depicts a view of the different SW implementations in place in an organization, associating the value and the risk dimension. Now, if we apply portfolio managements approach to HC management and CM, and clearly understand ITs decision-making process, current technology scenario, and, also, expected trends for the future, it will be much better understood. For this purpose, we will analyse IT-phased evolution, within the proposed HC framework in which CM strategy develops in todays advanced organizations: the employee transactional, life cycle management, and relations management domains.
48 Urquiza
Lets irst focus on the employee transactional domain. Traditionally, payroll and basic employee administration services have been considered the primary, core HR organizations service. Mistakes in this area do have a clear dissatisfying effect, thus damaging HR internal image. And particularly in medium- to large-size organizations, some speciic processing dificulties led to prompt IT automation: earning and deductions complexity derived from different, concurrent labour agreements to apply in single organizations, continuous updates in country regulatory requirements, gross-to-net permanent calculations, and inally, banking reconcilement and automated interfaces requirements. Payroll, therefore, rapidly became the irst core run the business type of investment in HR management, parallel to other basic back-ofice applications like accounting. Business risks were low (processes are quite stable) and beneits were medium. Other transaction-oriented processes followed automation in a segregated manner, just when business required the need due to growth, changes in HR policy, merging events, and so forth. Some of these processes were compensation and beneits or time management, typically non-discretionary enhancements, low business risks, and high beneits. Initial market products scope was local, co-existing with many made-to-measure solutions in place. Outsourcing of services models began to develop, particularly in the low- to medium-size type of organizations. Employee life cycle management is the primary core domain for HC competency management strategy deployment in any organization. Competencies are created to manage organizations required talent across the employee life cycle: Attract talent (e-recruiting) Assign talent (workforce planning) Inluence talent (performance/compensation management) Developing talent (e-learning) and Retaining talent (career & potential development). Although integrated solutions are the most relevant implementation solutions in todays IT industry (see section HC market applications: The CM contribution), stand-alone products are also growing. We will shortly analyse performance management and CM and E-Learning, these two being the most relevant convergent solutions in the CM market today. a. Performance and competency management
Early non-integrated deployed performance management (PM) systems were perceived by employees as non-useful, time-wasting applications. It is not until the e-business transformation process generated during the last ive years that organizations perceived real business impacts in developing values and competencies, enhancing employees business alignment. Medium- to large-size organizations started to create and track measurable skills and competencies through the deployment of advanced e-business-oriented PM solutions. About 40% of large organizations have already deined competencies for some type of jobs, and 10% have a well-deined enterprise-wide set of competencies. And the IT market today still has large expectations: 60% are still non-automated, paper-based solutions, about 25% in-house developments, and 10% product based (PeopleSoft, SAP, Workscape, etc.).
CM solutions are mostly grow the business and, to a lesser extent, transform the business venture-type applications. Business risks and reward are, typically, moderate to high. b. E-learning
Early (1980s to mid 1990s) learning management systems (LMS) were just tools driven to automate instructor-led training activities. E-learning today is somehow different. Cisco chairman John Chambers said in August 2003 that the two great equalizers in life are the Internet and education. More than 70 million people received education over the Internet that year. LMS most relevant change in the Internet age is that employees now become themselves responsible for their own learning requirements. Training is delivered in a personalized way, thus facilitating individual competency development as required and planned by the organization. ITs e-learning products have a singularity compared to the rest of HC applications related to competency management implementations: The LMS market is dominated by stand-alone solutions. Platforms like Saba (www.saba.com/),Centra (http://www.centra. com/education/resources/index.asp), IBMs Lotus (www-306.ibm.com/software/info/ ecatalog/es_ES/products/N105931Y77809P74.html), and 70-plus more represent today 95% of the total market, and just 5% (although growing) are tied to ERP-integrated packages (i.e., SAP, PeopleSoft). E-learning solutions it into the grow the business category. Business risks are moderate and reward is, typically, moderate to high. IT products in the employee relations management domain dont address or directly process CM-related information. They are viewed as communication-integrated tools, driven Figure 5. Generic structure of ESS portal implementation
La Solu cin
? ? ?
Pe r son a l D e v e lop m e nt N e w s & Ev e nt s H olid a y s Sh op p in g
? ? ? ? ?
Pr ocu r e m e n t M arket s Re sou r ce s Bu lle t in Pr oj e ct s
? ? ? ? ? ?
Ag e n d a Com u n ica t ion s Cola b or a t ion Te a m W or k Kn ow le d ge M n g. I n t e r e st Gr ou p s
? ?
Kn ow le d ge Ca r e e r D e v e lop m e nt
? ? ? ? ? ?
M e t r ics E- Le a r n in g Tr a v e l Pa y r oll Com p e t e n cy Mn gt . Ev a lu a t ion
2 4 Oct ob er 20 0 1
Cop y rig h t 2 00 1 I niciat iva Terra Met a4 y ou S.A.
50 Urquiza
Figure 6. Example of a real ESS
to automate most (if not all) HC processes, as in other parallel e-business environments. Typical applications in this domain are employee self service (ESS), manager self service (MSS), worklow, and expert inders. Lets shortly describe them. ESS are portal-driven solutions created to ensure comprehensive and controlled employee access to internal or external information and applications. Employee satisfaction and dramatic reduction in transaction cycle time over HC services are proven beneits in using this tool, typically integrated within HR ERP application. ESS Portal solutions are in the grow the business category. Business risks are moderate and reward is, typically, moderate. MSS technology and purpose is of the same nature as ESS. The only difference is in the nature of the type of applications that are relevant to this type of employees role within the organization. Managers require a single access point to information, otherwise disseminated throughout the organization. They manage people, budgets and groups assigned responsibilities in an effective, proactive way. Again, MSS solutions are in the grow the business category. Business risks are moderate and reward is, typically, moderate. Worklow represents a highly critical enabling technology to automate HR processes, thus critical too for comprehensive CM deployment policies: worklow replaces paper notiications with messages notiication, routing, and approval. Worklow servers solutions are mostly discretionary enhancements within the grow the business category, creating new levels of process eficiency and effectiveness (or process agility). Business risks and reward are, typically, moderate. Expert inders are Web-based tools typically found in integrated ERP HC solutions used to provide the means to ind and locate employees having speciic knowledge or competency proile (who is who-type of approach) and, inally, ensure contact with the expert for the required period of time. Usually works tie to the worklow server functionality. Expert inders solutions are considered discretionary enhancements within the grow the business category. Business risks and reward are typically moderate.
Human Resources Management Processes

The foremost trend in information systems restructuring in recent years is process-based working. This approach enhances knowledge development and sharing, forcing IT professionals and business decision makers to collaborate and share information concerning IT and business trends. Successful corporate e-business transformation initiatives necessarily incorporate process improvement and transformation. Process understanding provides the bridge between corporate strategy and IT implementation. CM software technology is no exception to this reality. As indicated in the introduction, CM market analysis requires a comprehensive approach within the HC domain. This brings about an excellent opportunity to critique and model value delivery mechanisms, thus clarifying business relationships. In todays advanced organisations, a process is deined as a set of activities directed to obtain a service or a product that creates value to the client. Although human capital management its into the corporate management support domain functions (same as inancials, planning, etc.), the same deinition applies. Competency management is the cornerstone in this proposed client-oriented process approach: People drive results. The following is a generic level 1 deinition of all relevant industry-accepted HC management processes representing the framework in which CM software is deployed. Best-process practices indicators are proposed, as well as information concerning the most signiicant internal and external application inter-relationships. This framework will facilitate the understanding of the competency management business ecosystem. Some HR management processes are very much tied to CM strategy deployment, and others are not. This comprehensive view will have signiicant value to further evaluate and assess the various software system architectures and market products available against the People CMM model.
Persons Administration and Payroll

Payroll processing represents the primary HR automated function in most organizations. It ensures that the workforce gets accurately paid on time. Critical activities are usually linked to speciic country regulatory requirements, the amount and structure of labour-related accounting models to consider, and other speciic reconcilement tasks. Best practices key process indicators in this area are the following: Employees administration and information low is integrated within HR or global corporate ERP. Administrative tasks are performed using ESS methods, implementing portal/intranetdriven solutions and worklow technologies.
52 Urquiza
Table 1. Process reference model for payroll & administration

Activity Name Description Workforce data and changes in employee status, relevant to payroll, are introduced in HR system. Sometimes, employee time reporting is the front end of payroll overall process. Taking into account monthly changes, backdate payment rules, corporate and country regulatory requirements, gross and net earnings, and deductions are calculated. Post-payroll process report writing is made to check possible errors prior to HR results formal approval, prior to post-information further processing.
PA_01
Employee data maintenance
PA_02
Payroll process
PA_03
Checking of calculated results
PA_04
Individual payroll forms generation Legal Social Security report generation
Payroll forms are generated and made available to employees in printed or electronic format for individual ESS access. Adequate automated monthly iles (FAN type or other) are generated, together with established payment forms (TC1s) According to company type and local regulations, adequate report (i.e. 110, 111, 190) is periodically generated. Earning and retention information is prepared for personal direct consultation (va ESS). System interfaces with agreed social beneit provision companies Files generated for bank use with net earning to be transferred to employees account. As agreed with the inancial department for accounting processing (general ledger, etc.)
PA_05
PA_06
Tax reports generation Social beneits information delivery Prepare bank transfer iles Accounting File generation
PA_07 PA_08 PA_09
The Process Reference Model for Payroll & Administration is shown in Table 1. The internal and external process relationship is shown in Figure 7. The external-related applications are inancials, ESS, travel management, and CRM (to directly manage incentives typically associated to sales or commercial goals).
Time and Attendance Management

In the context of knowledge persons management, TM (time control of employee time) is not a high-priority process and often deployed to a very limited extend. See Table 2. Figure 8 shows internal and external process relationships.
Figure 7. Internal and external process relationships
ESS
PA_04 Payroll Forms PA_05 SS Reports TRAVEL MNGT PA_06 Tax Reports
PA_01 Empl. Data Maint.
PA_02 Payroll calculation
PA_03 Check Results
PA_07 Social Benefit Rep PA_08 Bank Files FINANCIALS
CRM
PA_09 Accounting
Table 2. Process reference model for time and attendance management

Activity TM_01 TM_02 TM_03 TM_04 Name Time schedule deinition Time off request Time off authorisation Collect clock-in data Description Relevant labour calendar information is prepared: work hours, lexible schedules, and so forth. Relevant only if clock in control is in place. Applicable to vacation periods or special permits. Authorise or reject employee request and inform HR department. Exit and entry information is registered by external clock-in device, located in work premises. Based upon calendar and work hours information, a contrast task is performed with clock-in data in order to identify eventual deviations. The employee justiies detected work absence, then approved or rejected by superior. Time off accruals, vacation, sick leave type of information for HR information and management. Only the payroll-relevant information is transferred for adequate processing. Periodically, HR obtains statistic information to be delivered to employees via ESS.
TM_05
Identify deviations Deviations justiications Registry of time incidences Inform payroll Statistics
TM_06 TM_07 TM_08 TM_09
54 Urquiza
The only signiicant interface to consider here is with any clock-in system (when required). Notiications are often made via ESS or simple worklow dialogs with authorising persons. Best practices key process indicators in this area are the following: Presence control management integrated within HR or global corporate ERP. Decentralization of absence and time off registry. Absence/permit and vacation request approvals low is fully automated.
Organizational Development and Charting

Structure management is often a centralized process that creates, organizes, and manage organizational directory information. See Table 3. Figure 9 shows the internal and external process relationships. Best practices key process indicators: Structure deinition and job description information processed within HR or global corporate ERP. Decentralization of organizational changes across corporation.
ESS
TM_01 Time Sched.Def.
TM_02 Time-off Req.
TM_03 Time-off Autorize
TM_04 CLOCK IN CONTROL Collect Clock-in
Presence Control ?
TM_07 Registry incidence TM08 Inform Payroll
TM_05 Identify Deviations
TM_06 Justify Deviations
TM_09 Statistics
Table 3. Process reference model for organizational development and charting

Activity Name Identify organisation change request Description Every year each department performs a business strategic and functional review identifying recommended changes as business develops. Structure updates proposals are then submitted for approval. Change proposal is aggregated at organisation department level, then approved by board of directors. Once new structure has been approved (includes new assignments, positions, etc.) HR assumes control to update organisation-dependent information. Updating structure often requires changes in HR master data and cost centres allocation. Internally (via B2E mechanisms) and externally if required. Quality assurance manuals also updated.
OM_01
OM_02
Change request approval
OM_03
Inform on change to HR
OM_04 OM_05
Update structure New structure is made public
WORKFLOW APP,
PORTAL/ INTRANET
OM_01 Identif.Org.Ch.Req
OM_02 Change Approval
OM_03 Inform Change
OM_04 Update Structure
OM_05 Structure is publ.
Change organization approval process is managed using worklow mechanisms via ESS/Intranet.
Workforce Planning and Analytics

HR requires automated processes to evaluate generic and speciic income change proposals, beneits, or any other HR cost-related issue that may be considered. Simulation and analytical tools are essential to assume control of present and future HR budget responsibilities. See Table 4. Figure 10 shows the internal and external process relationships. Best practices key process indicators:
56 Urquiza
PAYROLL & ORGANISAT.
CP_01 Staff Planning
CP_02 WF Budget Analy.
CP_03 Scen. Simulation
CP_04 Final Budget Appr.
CP_05 FINANCIALS CONTROLLING Inform Controlling
CP_06 Budget appraisal
Table 4. Process reference model for workforce planning analytics

Activity CP_01 Name Staff planning Description Departments and business unit staff requirements and status requirements are processed. Aggregated information is then validated and budgeted. Current payroll cost is used to estimate cost for next period. Planned staff increase cost is estimated based upon current igures. Payroll cost models are simulated based upon previous calculations. Cost and budget effects due to organizational changes, new statutory requirements, or updated payroll and beneits conigurations are simulated. Final proposal is approved and submitted to management committee. New approved budget is incorporated in accounting. Permanent budget appraisal.
CP_02
Actual cost evaluation
CP_03
Workforce budget analysis
CP_04 CP_05 CP_06
Final coniguration approval Accounting control is updated Budget appraisal
Organization HR budget calculated through HR or global corporate ERP. Full integration with payroll and payroll simulation and analysis. Full integration with the organization module (to simulate organisational changes). Automated budgetary analysis and measurements, identifying unusual patterns in data through technologies such as data mining and understanding correlations between measurements.
Competency Management
CM processes low across organizational divisions and units as the most challenging practice in workforce management activities. Managing knowledge assets like managerial behaviours, problem-solving skills, and so forth is somehow different than managing physical assets. The ultimate CM process goal is to provide the means to ensure continuous development of employees and organizations competencies. It implies observing them through behaviour, assessing them according to organizations deined values, and permanently improving them. Competencies are usually linked to career planning, so employees permanently have a view of the potential career paths they are offered by the organization. CM strategies are as good or effective as the information available. Even though CM is about managing intangibles, without the right information measuring or predicting how employees or the organization will be affected is just hit or miss. See Table 5. Figure 10 shows the internal and external process relationships. Best practices key process indicators: Corporate competency catalogue deined within HR or global corporate ERP. Integration with all relevant applications, supporting HR processes. Automated gap analysis process, matching employee proile with proile required. Competency proile information through B2E communication-based links.
Employee Recruitment
Even though procedures vary from one region and organization to another, the process is driven to ensure the best tracking of applicants (internal or external to organization), based upon skills and competencies to match into the required job proiles. See Table 6. Figure 6 shows internal and external process relationships. Best practices key process indicators:
58 Urquiza
Table 5. Process reference model for CM

Activity Name Description Core and other required competencies, plus proiciency level required for each job proile within the organisation, are established and catalogued as a corporate dictionary. Once competencies are assumed by management, an assessment process deines procedures and levelassignment rules. Competencies are then matched to employees. This is often a time- and cost-consuming non-automated set of activities. Individual employees may leverage competency and potential career paths within organization. Gap between competency level deinition and employee status is identiied. This information is the basis for employee continuous improvement process, usually linked to CM. Individual persons development is supported in learning (e-learning) programs, providing the means to develop required competencies. Supported on KM information and individual development, proiles matching job vacancies are identiied within organizational scope. Succession planning information is analysed. Continuous matching employee-assigned business objectives with actual performance based upon the CM model. Evaluation information is created, then managed by HR evaluation process. Individual salary and beneits compensation is triggered by evaluation, thus closing the periodically updated employee development cycle.
CM_01
Catalogue deinition
CM_02
Competency assignment
CM_03
Employee enrollment
CM_04
Competency gap identiication
CM_05
Employee-continuous improvement activities
CM_06
Career management
CM_07
Performance evaluation
CM_08
Compensation
Recruitment process supported by HR or global corporate ERP. Full integration with administration application. Automated CV information data registry (through B2E).
Automated mail service.
CAREER DEVELOPM.
PA_08 Compensation
PERFORMANCE MNGT.
PA_07 Evaluation CM_06 Career Mngment
EMPLOYEE DB
ESS
RECRUITMENT & SUCCESSION PLANNING
CM_01 Catalogue Def.
CM_02 CompetencyAsign
CM_03 Enroll.Employee CM_04 Gap Identification
CM_05 KM LEARNING Continuous Impr.
Training
This process helps managers to evaluate and plan organizations training requirements and events. Information is processed on prerequisites, objectives, content, schedules, and locations, as well as appraisal courses results and additional management required data. See Table 7. Figure 7 shows internal and external process relationships. Best practices key process indicators: Training management supported by HR or global corporate ERP. E-learning integration within ERP. Integration with other related ERP processes. Automated mail service activities required to manage training courses.
Employees Career Development

This process creates a roadmap for success, determining future development activities. With CM, focus is made on high-potential employees and leadership-oriented careers. Figure 8 shows internal and external process relationships. Best practices key process indicators:
60 Urquiza
Table 6. Process reference model for employee recruitment

Activity ER_01 ER_02 ER_03 ER_04 Name Vacancies identiication Organization internal publication Internal candidates reception Internal candidates iltering Description Next-period strategic HR planning is taken as input information to identify vacancies to be assigned. Required position and expected candidate proile are made public within organization. CV is usually updated by employee prior to candidature submission. HR analyses candidatures and ilters position proile compatibility against candidate proile. HR and required technical and managerial personnel are involved during the selection process. Tests are designed in accordance to position. Technical knowledge and persons proile suitability tests are executed. Applies when no internal candidate gets position assignment. Scheduling. The selection application inputs candidates information for further assessment and tracking. Selected candidate receives notiication and formal offer to accept position. Once formal offer is accepted, starting work date is ixed. Organization-required administrative information is entered and legal employment agreement is created.
ER_05
Selection tests
ER_06 ER_07 ER_08
External recruitment Prequaliication and CVs reception Recruitment proposal HR administration incorporates selected candidate information
ER_09
ORGANISATION
PORTAL / INTRANET
ER_01 Identify Vacancies
ER_02 Internal Publicat.
ER_03 Receive int. Req. ER_04 HR ADMIN.
COMPETENCY MNGT
Filter int.candidat. ER_05 Selection process ER_09 HR Data update
ER_08 Selected candidate N ES_0R CV Reception

INTERNET
Make offer
ER_06 Extennal publ.
Table 7. Process reference model for training

Activity Name Description A training plan is established every business year. It includes all educational methodologies applied within the organization and is inally approved by executive committee. Following inputs are considered: Business units plans. Cross-organization corporate initiatives. Training topics generated by management, derived from evaluation results. HR corporate-deined improvement needs. Executive committee approves plan and budget. According to budget, priority, and expected employee usage. Internal and external providers, content and methodology evaluation against organization requirements, and cost and budget negotiations through established businessto-business commerce chain management procurement process. Assignment is based upon proiles. E-learning initiatives are usually managed by organization based upon employees self-initiative. Arrangements are made according to employees job schedules, presence, or distant learning method used, instructors and logistics availability. In accordance with plan. HR receives assistance and evaluation report. Following aspects are evaluated: Quality: content, instructor (by attendance). Students attitude, knowledge acquisition (by tutor/ teacher). Employee contribution to business (after training). By HR and management, after complete evaluation.
ET_01
Create training plan
ET_02
Identify education requirements
ET_03 ET_04
Training plan assessment and approval Courses & training units deinition
ET_05
Select training and education providers
ET_06
Assign employees to training units
ET_07
Training call
ET_08
Perform training
ET_09
Training evaluation
ET_10
Update employee competency record
Employees development plans deined in HR or global corporate ERP. Individual development initiatives managed from ERP results registry . HR ERP integration with other relevant processes. Employees potential and preferences registered in ERP.
Performance Management
Associated to any CM strategy, organizations require employees daily execution to be aligned to corporate strategy, committed to achieve measurable business results. HR focus
62 Urquiza
EMPLOYEES DEVELOPMENT
EVALUATION CCM & PROCUREMENT
ET_01 Create Training Plan
ET_02 Ident. Ed. Requir.
ET_03 Asses&Aprov.TP
ET_04 Courses Definition ET_05
COMPETENCIES
TECH/CORP.REQ.
TRAINING
Select T.Providers ET_06
ESS ET_10 Update E.Record ET_09 Evaluate Training ET_08 Train Employees
Employees sign on ET_07 Training Call
ADMIN.
E_LEARNING
is on skills tracking, targeting leadership and high-potential employees. It includes skills and competency personal assessment, and, sometimes, 360-degree type of evaluations and feedback. Best practices key process indicators: Evaluation deinition and results registry deined in HR or global corporate ERP. Individual self-evaluation made via ESS and worklow-based approval process. Integration with other relevant ERP modules.
Compensation and Incentive Management

Compensation is critical to ix adequate wage structures within the organization. Incentive management is required to manage employees variable retribution and beneits. Complexity varies according to extent and type of beneits (pension plans, private medical services, insurances, etc.). Usually third-party required information tracking and reports are used. Best practices key process indicators: ERP-based variable retribution, calculated for employee evaluation. ERP-based market salary survey.
Table 8. Process reference model for employees career development

Activity Name Description Based upon employees potential and strategic HR CM development objectives (includes target employees segment, criteria to identify potential, application period, etc.). Succession planning information is iltered. Following actions are taken: Identify employees matching proile employee appraisal. Update HR master employee data with employees information to ensure that opportunities are considered: job vacancy opportunities, special task assignments, and so forth. Individual career plans for competency development. Usually based upon employee performance evaluation, training opportunities, special task assignments, etc. Periodic checks to conirm suitability and compliance to organizations business objectives.
CD_01
Career development plan design
CD_02
Employee potential analysis
CD_03
Employees potential inventory
CD_04
Generate personalized development plans
CD_05
Developments plan appraisal
Market Applications: The CM Contribution

HCMS is considered today a mature market for the IT industry. Although the administrative, transaction-based type of applications have very limited growth potential (only outsourcing, hosted services in small and medium size business), the strategic talent management-type of applications, driven primarily by competency management business interest, is on the rise. Similarly to other enterprise-wide applications, where process integration becomes essential, the vast majority of medium to large organizations are now automating HC process in one out of three products: SAP, PeopleSoft, or Oracle, and two companies, because Oracle inally acquired PeopleSoft during 2005. Around 40% of the worldwide HCM market is concentrated in these two companies. This merger and acquisition wave is not an isolated exception (Adonix acquired Meta4, Cyborg merged with HR Services, Ininium with SSA, etc.). Microsoft (the worlds leading IT corporation but still not an important player in HCMS) and SAP announced merging conversations last year. Best-of-breed products previous business preferences are now swiftly moving to the ERP (trough acquisitions). Market fragmentation in HC management is therefore ending, and many small, single-function-oriented products are being incorporated in ERP suites or just disappearing. Maybe the LMS market is the exception to the rule, with an approximately 15% growth expectation in a $450M world market (2005).
64 Urquiza
Figure 14. Internal and external process relationships:
LEARNING
RECRUIT & SUCCESSION PLANNING
HR ORGANIZATION
CD_01 Prog.Dev.Design
CD_02 Potential Analysis
CD_03 Potential Inventory
CD_04 Indiv.Dev.Plan
CD_05 Plan Appraisal EVALUATION

COMPETENCIES
It is quite dificult to quantify which percentage of the global HCMS market corresponds just to CM processes. Most organizations enter ERP-based solutions (acquiring licences), automating just part of the functionality. Usually, deployment corresponds to most critical business functionality, with limited CM process automation. Companies invest in ERP suite licences and deploy functionality over the years with the conidence that the ERP vendor will ensure a long process integration. Worldwide HRMS and payroll administration revenues in 2004 reached around 3.900 M de with an estimated growth rate for 2005 of 7%. About 30% of the revenue corresponds to payroll and 70% to HR management processes. This survey assumes that, although not all product licences are devoted to comprehensive competency management deployment strategies, about 40% ($1,300 M) is the market size for HC direct CM-related process. The estimated igures reported in this study refer only to product licence fees and maintenance costs. Consultancy and implementation engineering costs are not included. In order to obtain
Table 9. Process reference model for employees performance management

Activity Name Competencies and business objectives deinition Description Frequently mix of global, business unit, and personal objectives plus competencies and shared values. According to persons perceived achievement. Performed by management, and employee agrees. HR validates and ensures organizations coherence. Information is obtained to shape next periods educational requirements and assigns variable beneits.
EV_01
EV_02 EV_03 EV_04
Self-evaluation Formal evaluation Formal approval
EV_05
Results analysis
COMPETENCIES
N EV_01
Stablish Obj&Competencies
EV_02 Self-evaluation
EV_03 Formal evaluation HR OK ?
Y ESS EV_04 Eval.Analysis EVALUATION

COMPENSATION
TRAINING
Table 10. Process reference model for employees compensation and incentive management
Activity CO_01 Name Compensation plan Description Plan is established according to strategic retribution policy, considering internal and market reference information. Competency development and business objectives and accomplishments are evaluated. Based upon results and time schedule to be applied to individuals. HR approves and submit schema for approval. Management board approves inal plan.
CO_02
Evaluation results analysis
CO_03
Compensation schema elaboration and review Management approval
CO_04
Market Survey
HR EVALUATION
CO_01 Compensation Plan
CO_02 Anal. Ev Results
CO_03 Comp.Schema
CO_04 Managt. Approval
Int. references
PAYROLL
66 Urquiza
Figure 17. Best practice cost distribution for competency management

HW 16% Consult 33%
SW 22%
Internal 29%
Concept
% 22% 16%
Ratio 1,50 1,00 0,73
Consultancy & Engineering 33% Product Licences Hardware
Figure 18. OS Environments in CM Implementations (2005)

10% 2% 8% 4% 33% Unix Windows OS/400 Main Frame Linux Others 43%
overall comprehensive IT-related cost distribution statistics, the following experience-based statistics method may be applied: Figure 17 shows current OS environment usage for CM-related solutions. Trends indicate a signiicant increase of Windows 32 and 64b. and reductions in AS400, others, and mainframe environments. Figure 18 shows that, for each Euro spent in the software product licence, 1.5 Euros are spent on consultancy and engineering services and 0.73 Euros in hardware.
Technical Architecture Reference Characteristics

The next point in this survey shows a generic evaluation model used to asses IT and HR business specialists in selecting the most adequate technology to deploy a competency
management strategy in any medium- to large-size organization. As we will see later on, the products technical architecture is an important issue in this process. This section depicts the technical architecture to be considered as a reference in CM strategy deployments. Which are the relevant aspects deine a good (or poor) technical architecture? Quickly and lexibly. Facilitating the introduction of new performance and scalability mechanisms. In a true multi-tiered approach, using comprehensive standards-based interfaces:
The client side is basically a user terminal (Web browser, PDA, etc.). Technologies: HTML, JavaScript, WML, Servlets, JSP, and so forth. The Server side consist on the Server elements, as a Web or WAP Server, that send information to the User Terminal. The business logic is the core application intelligence that makes the CM logic possible (application server with Java support, EJBs, servlets, etc.). Data Logic is responsible for data storage and transactional services (DBMS, etc.). Enhancing SW reusability and development skills specialisation, using component-based software.
Competency Management Market Products

Product solutions vendors that automate CM processes fall in one of these four categories: Large integrated ERP suites: These vendors dominate the high-end HCMS market; also competitive in SME Markets. Organizations select these solutions usually to automate more than one corporate business processes (in the areas of inancials, CRM, HC management, e-procurement, etc.). Even recognising that these vendors do not always deliver the best solution for any given particular function, their value proposition is founded on process integration and proven, reliable comprehensive process automation. Most market CM strategy deployment demand comes from large-size organizations to an extent that in the coming years, CM may be considered the most relevant business driver that will facilitate HCM market growth. SME integrated ERP suites: Similar functional scope to previous high-end ERPs with simpler parametrization options, much faster to deploy, tailored to SME-type organizations. Standalone HCMS: Large- and SME-driven solutions from vendors exclusively dedicated to payroll & HCMS, not providing support for other corporate business processes (as supply chain management, CRM, inancials, etc).
68 Urquiza
Standalone CM solutions: They automate CM processes, sometimes in combination with performance management and eventually providing support for some other related CM process, in a non-comprehensive, integrated way.
The following part of this chapter shows for each CM product-type category previously indicated, IT- and business-relevant vendor information, evaluating their product offering according to the next proposed evaluation method. Some additional markets are based on four evaluation criteria, as indicated hereafter:
Figure 19. Deliver Comprehensive E-Business functionality, across and between Multiple cChannels
Mobile Employee Connected Employee Thin Client Wireless Client Handheld Client
Intranet/Internet
CM Application
Employee DB
Figure 20. CM Development Architecture
Business Logic (Legacy Systems)
Presentation (Client Side)
Presentation (Server Side)
Business Logic (Web Technologies)
Data Logic
Conclusions and Outlook

Following are the most relevant market conclusions and outlook uncovered in this survey: HCM and CM application market scope is strong, clearly on the rise. Organizations have to adapt fast to changing, unstable business conditions in a way that CM paper and spreadsheet manual-based processes will have to evolve to a much more lexible, automated employee management-type practice. Market preference for ERP-integrated CM solutions instead of best-of-breed standalone products will continue in the future. ERP integration ensures that HR and business managers can eficiently align, measure, and automate individual employee competencies and organizational goals and objectives.
Table 11.
Evaluation EvaluationCriteria Criteria Key Key Factors Factors Financial stability strategy and vision Partners and alliances Research and development Professional services Product current and previous versions References CM functionality references Competencies deinition Competencies diagnosis Competencies development Competencies monitoring Rest of CM-, HCrelated processes Technical architecture E-business support Flexibility Integration facilities with required organizations Applications Security compliance Description Vendors with high potential to survive in the IT market and strong product maintenance and support are highly rated.
COMPANY
MARKET PRESENCE
Evaluate product stability, installed base and market credentials in previous implementations.
FUNCTIONALITY
Matching of vendor product standard functionality vs. organizational requirements in current needs and expected required support in the future.
PRODUCT TECHNOLOGY
Although e-business engineering practices are widely used, product architecture, scalability, and coniguration remain an important differentiator.
SAP AG
Table 12. Large Integrated ERP Suites Software Products
70 Urquiza
Evaluation Evaluation Criteria Criteria
Considerations
COMPANY
SAP AG (Waldorf, Germany) is the worlds third-largest independent SW company, devoted to enterprise management, and is the worlds leading ERP vendor. More than 28,000 clients, 1,500 partners, and 34,000 employees with local support in more than 50 countries. Worlds leading vendor in combined administration and HC management, with a 650 M year revenue (2004), second in HC management (560 M ), close behind PeopleSoft, although No. 1 in Europe.
MARKET PRESENCE
SAP started in the 1970s with the irst market product to automate inancial applications (R/1). Evolution from the irst R/1 installations to the presents comprehensive corporate processes (inancials, HCMS, CRM, SCM, e-procurement, business warehouse, portals, etc.), applications under R/3, in which HC and CM solutions are integrated, is a history of sound evolution, incorporating worldwide process excellence and real customer migration support in an all-new product version (currently, once a year).
FUNCTIONALITY
All CM-related processes identiied in this survey are included in the integrated, SAP R/3 Solution. Its international focus has ensured extensive local regulatory worldwide support for administrative functions (i.e., payroll). Some processes remain rigid to implement (compensation, beneits) compared with bestof-breed solutions, but overall CM associated functionality rating is very good.
PRODUCT TECHNOLOGY
SAP NetWeavers architectural design comes from a rigid client-server Model, supported on a vendors proprietary environment (ABAP). Product evolution in the last three years offers a reasonable Java support and interoperability with .NET and IBM WebSphere (J2EE). Development, deployment and execution environments run on its own SAP Web Application Server (WAS), providing combined ABAP, J2EE, and Web services support.
PeopleSoft
Table 12. Continued
Evaluation Criteria Evaluation Criteria
Considerations Considerations Acquired by Oracle in 2005, worlds second-largest independent SW company (See Oracle, next company).
COMPANY
MARKET PRESENCE
PeopleSofts roots within the enterprise management business are in HR, then extended to inancials, SCM, CRM, and e-commerce applications. Leader in the US, far behind in European sales due to SAPs dominance and lateness in incorporating country-required regulations in HHRR administrative and payroll processes in some countries.
FUNCTIONALITY
PeopleSoft Enterprise suite represents the most advanced, integrated process automation solution for HC management on the market. Last product versions have introduced enhancements in CM management and workforce analytics, integrating inancial, transactional HR, and external market data to provide strategic performance management. Workforce rewards manages compensation and retention policies. PeopleSofts own Balance Scorecard is effectively used to manage HR effectiveness under workforce scorecard module.
PRODUCT TECHNOLOGY
PeopleSoft architecture is a very lexible combination of database, application, Web, and ile Server, fully e-business compliant. The PeopleTools Application Server runs the business logic and facilitates access to most client types, using the Internet architecture and allowing interfaces to visual basic-visual C-based user programs.
Oracle
Table 12. Continued
72 Urquiza
EvaluationCriteria Criteria Evaluation
Considerations Considerations
COMPANY
Oracle is the worlds second-largest independent SW company. The combined Oracle + PeopleSoft + JDEdwards (previously acquired by PeopleSoft) has made Oracle the worlds second-largest business application vendor (behind SAP) with an estimated annual revenue (iscal year 2004) in this business of 450 M , and irst vendor in HC and competency management-related solutions, largely because of PeopleSoft enterprise market share. Extensive global partner support and country-assigned services in most locations.
MARKET PRESENCE
Oracles HCMS has 3,000-plus installations in more than 80 countries.
FUNCTIONALITY
Full HCMS process support as part of e-business suite, integrated suite; includes all required functionality to deploy CM-related strategies. Its Daily Business Intelligence supports excellent HC analytical service. Performance, compensation, and e-learning management are excellent and eficient CM-related modules.
PRODUCT TECHNOLOGY
Oracle has now three independent suites to manage: its original ERP, PeopleSoft Enterprise, and JD Edwards Enterprise One/World. The biggest challenge for Oracle is to ensure a product strategy that combines strengths from all different products, delivering customers a reliable and affordable evolutionary roadmap from its current products diversity. During the next two to three years, it is expected that Fusion Oracles applications will incorporate many of the current features from PeopleSoft Enterprise.
Lawson Software
Table 12. Continued
Started operations in the U.S. in 1975 in the corporate analytics and management applications market to become a global ERP solution provider, with particular focus on inancials, HR, and procurement. Its business growth strategy has been based on best-of-breed product acquisitions (Account4, ijob, Armature, etc.) and most recently (June 2005), Intentia (Swedish leading SW company, small-size organizations ERP vendor), to create a 3,500employee company, with 4,000 customers distributed in 40 countries. Large U.S. technology partners list but lacks international, long-term local support and consulting partners outside U.S. Lawson software sales in HC- and CM-related business reached 44 M in 2005 with inancial growth dificulties in last three years.
COMPANY
MARKET PRESENCE
Until recently, business was focused in the United States, Canada, and the U.K. markets. Once merging with Intentia concludes sometime next year, more European technical support may be expected. Product stability and roadmap evolution needs to be ensured in following years.
FUNCTIONALITY
Offers HR suite as part of Lawson. Insight series application includes some CM-type functionality within its personnel administration module but lacks a comprehensive CM functionality (poor performance management, no e-learning part). Strong workforce analytics that compare internal metrics with external sources (provided by Saratoga) in areas such as compensation, stafing, and organizational effectiveness.
PRODUCT TECHNOLOGY
E-business compliant, Web-based architecture, with interesting software extension tools to facilitate integration with other applications (business component integrator).
Table 13. SME-Integrated Suites Northgate Information Solutions
74 Urquiza
COMPANY
U.K.-based software applications and outsourcing services vendor. Created in 1969, has become leading U.K. HR and payroll supplier in the U.K. trough, small and leading state-of-the-art companies (Prolog, PWA, CaraPeople, Rebus HR Group). Apart from HR, includes SME-driven, ERP-type solutions for inancials and CRM. Currently employs 3,300 people, having 5,000-plus SME clients, 58 M revenue in HC management, mainly in the U.K.; presence in the U.S., Australia, and New Zealand.
MARKET PRESENCE
In the U.K., long history of success in SME organizations, offering comprehensive product licence sales, engineering, and outsourcing services.
FUNCTIONALITY
It offers wide functionality in HR, starting with basic processes and product-based automation. Usually requires extensive speciic software development, particularly in areas as workforce management. Includes recruitment, training, beneits, and ESS-type access. Most CM implementation requires product made to measure changes integration with external applications.
P R O D U TECHNOLOGY
Open, scalable, Web-based, non-integrated, standard ERP-type of architecture.
Microsoft Business Solutions
Table 13. Continued
Considerations Considerations Microsoft is the worlds largest SW company, providing products and services to individuals and organizations around the globe. Its business solutions division (Microsoft Dynamics) is an international provider of integrated solutions for most SME functional areas. Microsoft entered the ERP market through acquisition of small leading market vendors (Great Plains, Solomon Applications, Navision, etc.). The most signiicant ones in HRMS are Navision and Axapta, the latest one incorporating an advanced competency management solution for SME.
COMPANY
MARKET PRESENCE
Microsofts market share in HCMS is low (30 M , representing just a mere 1.3 % of the worlds market) in relation to company proile and business market size, mainly because of current SME-driven strategy, until now. In this segment, implementations history is consistent with good customer references, although each product within the family (Axapta, Navision, Great Plains) follows individual development road maps.
FUNCTIONALITY
Microsoft Axapta is easy to implement and operate with integrated HCMS (plus inancials, CRM, project management, etc.) in SME-type organizations. Excellent CM functionality, driven to detect training requirements among individual employees, as well as career plans and organizations re-structuring functions.
PRODUCT TECHNOLOGY
Open, scalable Web-based architecture .NET and Java, e-business-compliant solutions.
Kronos
Table 14. Standalone HRMS Products
76 Urquiza
Evaluation Criteria Criteria Evaluation
COMPANY
Kronos incorporated began operations in 1977 as a hardware time-clock vendor and has successfully evolved into an HR payroll and management company, reaching $519 M in sales in iscal year 2005. Its the worlds third-largest vendor in HC management (irst non-global ERP provider), just behind SAP and Oracle, with total 2004 sales of 210M . About 2,900 employees, offering comprehensive licences sales, consulting, implementation and learning services around the companys HR products. Included in Forbes 200 Best Small Companies.
MARKET PRESENCE
An excellent record of customer retention through service satisfaction, combined with visionary product evolution from a hardware to the most dynamic software HC management business, sometimes by acquisitions (i.e., AdOpt, excellent workforce planning module). Business focus is driven primarily to the U.S., and some customers are in Canada and U.K.
FUNCTIONALITY
Kronos offers a comprehensive HR administration and management solution for SME and some large-size organizations, including most relevant CM-related processes. Probably the most successful WF management solution in the U.S. market. Its workforce central suite includes core CM functionality. Focus in aligning skills and competencies to corporate objectives, tracking skills, and certiications, and competency enhancements trough training initiatives.
P R O D U TECHNOLOGY
Open, scalable Web-based architecture. Deploy Java applets to facilitate dynamic user interfaces, as well as HTML access.
Meta4
Table 14. Continued
COMPANY
Spanish-based company, founded in 1991, acquired by Adonix in November 2004, strengthening previous companys inancial situation and clearing future market development. In November 2005, the Sage Group Plc announced the acquisition of Adonix under an operation that explicitly excludes the business of Meta4 (leaving control of French entrepreneur Emile Hamou). Large and medium-size extensive customer list (1000-plus) in 20 countries. Leading HC, CM market product in Spain with a strong competitive position in other European markets (France, Portugal, U.K.) and South America, with implementations in large as well as mid-size organizations. In October 2005, Meta4 reopened operations in the U.S., supporting its new operations for the Americas in Miami, through partnership with independent HR consultant Vision 3. Extensive technology, consultancy, and service, outsourcing providers in operational Markets. Meta4 PeopleNet/MindSet suite has an excellent administrative process functionality, except in beneits administration, and a powerful competency management strategy and set of tools, particularly competency assessment and performance management. Full CM-related processes automation, including employee/manager self-service and comprehensive integrated knowledge management environment. State-of-the-art e-business technology design and implementation. 3-Tears components-based architecture: Supports pure HTML, Java (applets), and Windows (Active X) Clients, application server and DB Server.
MARKET PRESENCE
FUNCTIONALITY
PRODUCT TECHNOLOGY
Ultimate Software
Table 14. Continued
78 Urquiza
COMPANY
Ultimate Software is a successful payroll and HCMS dedicated company, expanding its business from a mere licence sales vendor to an outsourcing service provider, reaching 37 M in iscal year 2005 in HC management sales. Currently employs 500 professionals and has about 1,200 customers, mainly in the US. HROA (Human Resources Outsourcing Association) 2005 Provider of the Year.
MARKET PRESENCE
An excellent record of customer satisfaction, combined with visionary product to service business orientation. Business focus is driven primarily to the U.S., and some customers are from Canada and U.K.
FUNCTIONALITY
Its Ultipro Workforce Management Suite incorporates most required CM required process automation functions.
PRODUCT TECHNOLOGY
Open, scalable Web-based architecture. Earned certiication from HR-XML Consortium in 2003, creating HRspeciic XML vocabularies.
Mindsolve
Table 15. Standalone CM Products
COMPANY
Fast-growing company created in 1994, exclusively dedicated to the employee performance management business, licensing its Mind Solve Visual Product and providing associated required consultancy services. Last year business revenue estimated at 13 M .
M A R K E T PRESENCE
Niche player, with consolidated HR expertise support to deploy PM functionality in heterogeneous U.S.-based (some outside US) medium- and large-size organizations. Product licence sales and MVP Express suite, tailored for outsourcing, easy to deploy Web-based solution.
FUNCTIONALITY
Includes end-to-end performance appraisals, 360-degree performance assessment, competency alignment and accountability, and talent and development planning. Skills and competency management support, requires interfaces with other external tools for full CM strategy deployment (e-learning, workforce management, etc.).
P R O D U C T TECHNOLOGY
Open, scalable Web-based architecture. It uses proprietary technology (MindSolves Visual Proiler) of unique drag-and-drop system to deliver accurate, trustable information. Visual Proilers ability to comparatively display multiple employees on a single screen produces ratings that better differentiate superior, good, and unacceptable performers.
Geo Learning
Table 15. Standalone CM Products
80 Urquiza
COMPANY
Geo Learning is an SME-type company, U.S. based, dedicated to e-learning and associated educational activities. It sells learning products (LMS, authoring tools, etc.) and other stand-alone SW products, as its competency plus CM tool. Started business in 2000; last 3 years, annual growth rate of 125%; has more than 300 SME-type clients, mainly in government and inancial services.
MARKET PRESENCE
Good record of customer base. Works with small reseller and systems integrators partners.
FUNCTIONALITY
Exxceed competency plus is a stand-alone CM product that includes competency modeling (using proiles matching jobs, projects, etc.), career management (job proiles matching employees competencies), individual development planning (based on competency gaps, assigning learning objects, imported from external LMS), and comprehensive 360 competency assessment process. The foundation of Exxceed functionality is based upon a competency dictionary wherein individual competencies are stored and organized in the form of families, identifying assessable behaviours linked to competencies. Typically records and employee information is obtained using crystal reports or other Web-based reporting tools.
PRODUCT TECHNOLOGY
Microsoft, PC, SQL-type application, thin client Web-based architecture. Low cost and maintenance. Delivered with preconigured competency DB (10,000 competencies in areas such as business, management, engineering, etc, and 400-plus industry predeined job proiles).
In this context, IT managers estimate that the integration value provided by global ERP suites far exceeds the beneit of using best-of-breed solutions for each HC business function. The e-business technological transformation and the processes model depicted in this survey clearly show the inter processes and software dependencies required in CM deployment strategies. And again, the fact that the trend has been widely adopted creates additional conidence in large vendors ability to deliver HC-proven process automation. This loop affects CM implementations, where process automation is critical. In some particular CM tasks (like competency assessment), process eficiency, when adopting a stand-alone option, increases fairly moderately compared to the asset they replace. This situation also increases pressure in IT managers to adopt the comprehensive, integrated ERP suite way. Large, multinational companies require comprehensive multi-region, often centralized, HC deployment solutions. Global processes and local country localizations makes the three large ERP solutions analysed in this survey the strongest CM option in the future for this type of organization, thus displacing fragmented solutions. Small and mid-size organizations (less than 4,000 employees) and large, single-country type installations have many more options. According to required functionality, together with associated licence and engineering costs, they may select in the high-end or mid-market type of products, thus accessing a wider menu of options. Hosting and outsourcing HRMS services keep growing at a fast pace, particularly in small- to medium-size market companies. This trend represents an intelligent business option for small best-of-breed vendors, having experienced competency assessment experts and staff (apart from being acquired or just disappearing). Some product vendors, apart from the traditional, large-size HRMS service providers (i.e., ADP), are successfully, adopting this delivery model(i.e., Ultimate Software, Ceridian, etc.).
References
Alle, V. (1997). The knowledge evolution: Expanding organizational intelligence. Boston: Butterworth-Heinemann. Benson, S., & Standing, C. (2001). Effective knowledge management: Knowledge, thinking and the personal corporate knowledge nexus problem. Information Systems Frontiers, 3(2), 227-238. Blain, J., & Dodd, B. (ASAP Internacional Group) (1999). Administering SAP R/3: The HR-human resources module, QUE. Indiana. Boxall, P., & Purcell, J. (2003). Strategy and human resource management. New York: Palgrave Macmillan. Curtis, B., Heley, W. E., & Miller, S. A. (2001). People capability maturity model (P-CMM), (Version 2.0) (Tech. Rep. No. CMU/SEI-2001-MM-01). Pittsburgh, Carnegie Mellon University, The Software Engineering Institute.
82 Urquiza
Friss de Kereki, I., Azpiazu, J., & Silva, A. (2004). Knowledge management in learning environmental design. In Proceedings of the 34th ASEE/IEEE Frontiers in Education Conference, Savannah. Hardless, C. (2005). Designing competence development systems. Unpublished doctoral thesis, University of Gteborg, Gteborg, Sweden. Hartman, A., & Sifonis, J. (2000). Net ready. Strategies for success in the e-conomy. McGraw-Hill. Kamara, J. M., Anumbad, C. J., & Carrillo, P. M. (2002). A clever approach to selecting a knowledge management system. International Journal of Project Management, 20(3), 205-211. Levy-Leboyer, C. (1997). Gestin de las competencias. Cmo analizarlas, cmo evaluarlas, cmo desarrollarlas, ediciones gestin 2000. Barcelona: SA. Lindgren, R. (2005). Adopting competence systems in fast growing knowledge intensive organizations. Journal of Information & Knowledge Management, 4, 1-13. Lindgren, R., & Stenmark, D. (2002). Designing competence systems: Towards interestactivated technology. Scandinavian Journal of Information Systems, 14, 19-35. Means, G., & Schneider, D. (2000). Meta-capitalism. The e-business revolution and the design of companies. Nordstrom, A. K., & Ridderstrale, J. (2000). Funky business. Talent makes capital dance. Stockholm, Sweden: Book House Publishing.and Markets in the XXI century. Pricewaterhouse Coopers, Deusto Ediciones. Rollett, H. (2003). Knowledge management processes and technologies. Boston: Kluver Academic Publishers. Sagi-Vela, L. (2004). Gestin por competencias. El reto compartido del crecimiento personal y de la organizacin. Madrid: ESIC Editorial, Pozuelo de Alarcn. The Hay Group. (1996). Las competencias: Clave para una gestin integrada de los recursos humanos. Bilbao, Spain: Ediciones Deusto SA.
Competency Management using the Competence Performance Approach 83
Chapter IV
Competency Management using the Competence Performance Approach:

Modeling, Assessment, Validation, and Use
Tobias Ley, Know-Center, Austria Dietrich Albert, University of Graz, Austria Stefanie Lindstaedt, Know-Center, Austria
Abstract
This chapter introduces a new approach in organizational competency management, which is based on Korossys competence performance approach and which rests on the central idea of connecting competencies to actual job performance. Such an approach has several important beneits when compared to more traditional approaches. First, it brings competency modeling and assessment more closely to the actual work processes and tasks. Second, it makes possible validation of the models and the assessment within the modeling and assessment procedure. Finally, it is better able to cope with more dynamic and knowledge-based settings. Three case studies in engineering, human resource management, and research and development illustrate how the approach is introduced, evaluated, and applied. The purpose
84 Ley, Albert, & Lindstaedt
of the chapter is to inform researchers in e-learning and knowledge management of how competencies can be used to support work-integrated assessment and learning.
Competency Management in Organizations

It has been argued that the success of todays businesses increasingly depends on their intellectual assets as opposed to their tangible resources (Stewart, 1997; Sveiby, 1997). Among other things, these assets include knowledge and skills of the workforce. Several organizational disciplines have attempted to ind ways to leverage these assets. From a strategic management point of view, the question has been how organizations are able to use the assets to secure a persisting competitive advantage. Ever since the inluential publication of Prahalad and Hamel (1990), who coined the term core competence, human expertise has been seen as an important ingredient in the mix of a companys systems, technologies, physical location, and infrastructure that make up this competence. Therefore, managing individual competencies (knowledge and skills) is one important element in the management of strategic competitive advantage, if individual competencies are in line with organizational core competence. On the other hand, strategic management research has usually not focused on concrete instruments that deal with individual competencies. Most certainly, human resources management (HRM) has been the primary discipline concerned with operationalizing these strategic objectives. By deinition, HRM is concerned with optimizing the workforce in line with strategic objectives. Traditional HRM instruments (like job analysis, selection, training and development, and the like) have been in use in organizations for years. However, their use has recently been criticized for neglecting the strategic connection. It has also been questioned whether these instruments are able to cope with the new productivity challenge in the knowledge-based economy, namely to enhance the productivity of the knowledge workers who now make up a large share of the workforce (Elkjaer, 2000). More recently knowledge management (KM) has introduced new perspectives. Being driven by innovative information technology (IT) applications, the goal has been to enhance access, sharing, use, and creation of knowledge in organizations. However, just providing employees with a better access to available information or to communication channels has not always produced better outcomes. Instead, peoples ability to make use of and apply knowledge they are generating is becoming a key issue. Michellone and Zollo (2000) argue that the very paradox of knowledge is that irms possess knowledge only if they are able to transform it, and the primary ability to transform knowledge resides in people and their competencies (p. 137). This has been acknowledged by some KM advocates by putting an emphasis on implicit knowledge and the techniques necessary to deal with it from an organizational point of view. It appears that all three disciplines have struggled with the question of how to better leverage human competencies in organizations. They have approached the question from different perspectives and have used different techniques to deal with it. And they all have limitations as well.
The term competency management (CM) will be used subsequently to refer to a class of instruments that have been discussed in all three of the disciplines. CM is taken to encompass all instruments and methods used in an organization to systematically assess current and future competencies required for the work to be performed and to assess available competencies of the workforce. Competencies are deined as the cognitive (e.g., knowledge and skills), affective (e.g., attitudes and values), behavioral, and motivational (e.g., motives) characteristics or dispositions of a person that enable him or her to perform well in a speciic situation. Competencies are dynamically combined by the performer to meet situational requirements. CM has been widely adopted in practical contexts, especially in the U.S. Based on data of several surveys between 1996 and 1998, Schippman et al. (2000) conclude that 75-80% of responding companies have some competency-driven application in place (p. 704). It is a goal of this chapter to develop and apply an instrument for competency management that alleviates some of the drawbacks of the approaches mentioned above. A central concern will be to more closely connect competencies to actual performance in the workplace. The next two sections deal with current approaches of CM by introducing a irst case study and then reviewing existing approaches. From these, challenges and shortcomings that have not been suficiently addressed in current approaches will be identiied.
Current CM Approaches: A Frame of Reference and a Case Study

The following frame of reference is used throughout this paper to refer to typical processes of CM in organizational settings. While the methods that have been developed for CM are rather distinct, several general process steps can be identiied that are mostly employed when CM methods are being used (see Figure 1). The steps 2-4 can be seen as the core processes in competency management. The other two provide the interface to the organization (e.g., its strategy and other processes). The steps do not necessarily have to follow this strict sequence; they should rather be seen as focus areas. The sequence, however, provides a good heuristic for the start of a competency management project.
Figure 1. Process steps involved in a competency management approach
This process model will now be illustrated by a irst case study (Case Study I) that was conducted in the engineering unit of a large automotive company (see also Wls, Kirchpal, & Ley, 2003). The case study also exempliies a typical approach to how CM is currently being conducted. Later, in the third section, we will more systematically review current approaches to CM and derive challenges for future research.
Analyzing Setting and Purpose for CM

The view of CM advocated in this paper maintains that implementing CM involves the construction of an organization-speciic model that relates competencies and performance outcomes. This makes it necessary that the irst step in the implementation process asks for the purpose the CM approach has for the particular organization and for aspects of the setting that inluence the modeling approach. The company is a large engineering and production facility that acts as a supplier to the automotive industry. In several business units, different parts of cars and trucks are developed and produced. The company employs more than 9,000 employees worldwide. The engineering department employs about 200 persons. The purpose was to support the training function in the assignment of development measures for individual employees (HRM purpose) and to provide the basis for an expert search to be developed at a later stage (KM purpose). It was decided to follow a job-based approach; that is, to analyze several jobs in terms of the competency requirements and the available competencies of the respective job incumbents. The positions that were analyzed within the case study were four department head positions in the engineering unit and one group leader position also in the engineering unit. Furthermore, a top-down approach was chosen; that is, the supervisors of the position were interviewed. One reason was that one supervisor could provide information about several jobs and several job incumbents. Also the main purpose of the initiative was decisions about personnel development, which at that time was the responsibility of the supervisors. The implementation was conducted by a project team made up of three employees in the company (from the human resources (HR) and the engineering unit) and one researcher (the irst author).
Deining Competencies
This step is concerned with providing a model of which competencies should be measured and how they relate to organizational performance. In the case study, three methods were used, the irst being a competency elicitation interview with the supervisors to derive competency requirements, the second being job proiles in which results of the interviews were documented, and the third being a competency catalogue that was used to document all competency names and descriptions. In the interviews, the supervisors were asked for the main tasks of the jobs they supervised, and they were asked to derive the competencies required for these tasks. Competencies were classiied as either technical, management and methods, or personal and social (see, for example, Erpenbeck & Heyse, 1999). For each job, each competency was rated on a
four-point scale in terms of what the job required: (1) basic knowledge, (2) standardized application, (3) complex application, or (4) expertise. This scale relects the fact that a competency usually is composed of knowledge in a certain domain as well as skills needed for application. The scale also presupposes that the development of a competency in a certain domain follows an inherent order. The scale implies that basic (declarative) knowledge is developed irst (1), skills in the form of contextualized rules are developed later and lead to increasingly lexible forms of application (2) and (3) (Anderson, 2000; Hacker, 1998). Expertise then refers to the highest stage of application using highly contextualized and implicit skills in the particular domain (4). The interviews were supported with a lipchart visualizing the job proile (see Figure 2). The competency catalogue was used by the project team to document all competency names and descriptions after the interviews had taken place.
Assessing Available Competencies

At this point, a measurement process takes place that assesses the competencies deined in the previous step among a certain group of employees. In the case study, both supervisors and job incumbents independently rated the scale level of the job incumbent for each competency in his job proile. The same four-level scale was used as in the previous step. This means that both competency requirements of the job and actual competencies of the incumbent were rated on the same scale. This is also suggested by Faix, Buchwald, and Wetzler (1991) (see the also the section Current CM Approaches: Reviewing Purposes and Models ). The idea behind this is that both ratings (required and actual) can directly be compared, and necessary training measures can be derived when the level of required competencies is higher than the level of actual competencies. Following this rating, the differing views (between supervisor and job incumbent) were discussed in a moderated meeting involving the job incumbent, the supervisor, and two members of the implementation team. This meeting was meant to imitate the annual ap-
Figure 2. Schematic illustration of lip chart used for competency elicitation interviews
praisal discussion between supervisor and job incumbent. The result of this phase was an employee proile that both supervisor and job incumbent agreed on.
Evaluating Models
The model that has been deined in step 2 and the measurement performed in step 3 both constitute hypotheses about the affairs within the speciic organization under consideration. The question remains whether these hypotheses correspond to the true affairs. In the case study, these hypotheses were not evaluated in a systematic manner, as the focus was on the feasibility of the implementation method. For this reason, a lessons learned workshop was conducted among the members of the project team after the implementation in order to discuss positive and negative aspects of the implementation and ind ways to improve it. An important topic of this meeting was the question of implementation effort, which in the current model was quite substantial both from a modeling perspective (maintenance of the competency descriptions and deinitions) and assessment perspective (yearly appraisal of all employees).
Using the Models

Using the models within organizational practices leads back to the purpose in step 1 and makes the collected data available for the practices deined there. In the case study, the proile data was used in the appraisal discussion to ind suitable development measures. The data was then communicated to the HR department together with the list of development measures generated for the employee. Also, a concept was designed to connect the competency descriptions with an existing training catalogue.
Current CM Approaches: Reviewing Purposes and Models

This section will offer a more systematic overview of CM methods currently in use. We have chosen to offer an application-oriented view of the ield of research by suggesting two ields in which CM methods are being applied, namely HRM and KM. This view puts the methods in the context of the aims and traditional practices of the two organizational functions and of the research disciplines that are connected with them. For example, the two functions are distinct with regards to the elements on which they put the primary focus of intervention, namely personnel on the one hand and IT on the other. Thus in this section, traditional methods of CM and their respective theoretical foundations will be reviewed with relation to the functions from which they have evolved. The strategic management function is not explicitly covered because the focus there is more on organizational level concepts (such as core competencies). Nevertheless, a strategic perspective is taken both within KM and HRM approaches, and will be covered at certain points of the review.
The purpose of the review is to identify elements of existing approaches that have proven to be applicable in organisational contexts and to identify research challenges for CM, which a new approach needs to address.
Competency Management and HRM

History of terms and method. There is now a rather long tradition of competency management approaches within HRM research and practice. Many researchers have credited the start of the competency movement within the HR ield to an article by McClelland (1973) in which he criticized the widespread use of intelligence tests in college admissions and occupational selection. Instead, he argued that selection should be based on competence tests that take into account criterion sampled behavior, for example, by looking at what kind of behaviors actually differentiate good from poor performers in the ield that the test is supposed to select for. (If you want to test who will be a good policeman, go ind out what a good policeman does, McClelland, 1973, p. 7). McClellands work has since been elaborated on and adapted by himself and his colleagues to suit the needs of the business world (Boyatzis, 1982). This includes methods to derive job requirements for any position (Klemp & McClelland, 1986) or providing readymade competency models for speciic positions (Spencer & Spencer, 1993). Overall frameworks. The approach has been termed competency-based HRM, which is concerned with deining human competencies needed to fulil job requirements and identifying competencies that people have acquired. This information is used for all HRM processes, like selection, training, performance management, succession planning, and so on. Several overall frameworks have been suggested that are intended for use in the whole organization, covering a large number of jobs and people. These frameworks advocate a certain model of what competencies are and how they are related to other concepts, and they establish a procedure for implementing the approach in a speciic organization. Typically, they also give examples of speciic competency models for certain jobs. Typical overall frameworks include the one by Green (1999), who emphasizes the connection between individual level concepts (technical knowledge and job skills and performance skills and competencies) and organizational level concepts (core competencies and capabilities and core values and priorities respectively). Green also presents several ways to link competencies to HRM instruments, like selection interviewing, performance appraisal, coaching, and training. Lucia and Lepsinger (1999) present another framework that emphasizes the differentiation between different types of individual competencies. Also, they present a detailed procedural model for introducing competency-based HRM into an organization. Two German conceptualizations include a very early one by Faix et al. (1991), who suggest storing two types of proiles, one for each job (required competencies) and one for each employee (available competencies). Recently, Erpenbeck (2003) presented another framework, which draws on a sound psychological model of individual competencies (Erpenbeck & Heyse, 1999; Erpenbeck & Rosenstiel, 2003). Expertise development. While the overall frameworks usually take a very centralistic stance, psychological expertise research has long examined the development of superior
performance in the workplace (Sonnentag & Schmidt-Brae, 1998). Results indicate that conceptualizations of expertise should not merely be based on quantitative performance differences. Instead, when looking at human performance dispositions, qualitative differences in comparing novices and experts are prevalent (like types of task representations, elaborateness of the strategies used, and reliance on procedural knowledge). Also, research in expertise points to a difference between domain-speciic and domainindependent competencies, and to the fact that situations in which performance is exerted plays a decisive role. Consequently, analyzing competencies has to include analysis of performance and of the situation in which it takes place. Finally, it has been pointed out by many researchers that although there is a need to involve experts in determining their competencies, this is also problematic because knowledge about competencies is mostly tacit and therefore dificult to verbalize. Work-integrated CM. It has long been realized that most learning in which new competencies are acquired happens in informal situations and not in formal training or education. Building on this and on the results of expertise research, attempts have been made to study workintegrated competency assessment and development (Baitsch, 1998; Erpenbeck & Rosenstiel, 2003; Garavan, Morley, & Gunnigle, 2002; Staudt & Kriegesmann, 2002). Assessment of competencies then uses self-report measures, simulations of actual work tasks, the analysis of work samples, or situative questioning.
Competency Management and KM

Building on Prahalad and Hamel (1990) (see above), Probst, Deussen, Eppler, and Raub (2000) devise a process model for developing organizational core competencies that incorporates a broad array of knowledge management instruments. For identiication of individual competencies, they cite competency-based HRM (see previous section), knowledge mapping and competencybased project stafing. The latter two will be more closely considered at this point. Knowledge mapping. Knowledge maps map a domain of knowledge in a corporate domain by providing a visual representation of available knowledge sources or expertise. They are used to facilitate access to expertise in large companies (Eppler, 2001). The map consists of a conceptual framework that provides a common context (e.g., a representation of the companys business processes), and single elements that usually contain reference information to knowledge sources (documents) or expertise (persons). Knowledge source maps contain references to knowledgeable persons. The information contained in them is equivalent to a collection of employee proiles (available competencies). Knowledge structure and application maps, on the other hand, are based on a mapping between tasks and competencies. Project team stafing. Creating lexible teams is an important instrument for building up new knowledge in organizations (Probst et al., 2000). In the literature, these approaches are usually conceptualized as internal project markets in which the competencies of employees are made transparent so that stafing of teams can be based on a matching of available competencies to project requirements (Davenport, 1997; Deiters, Lucas, & Weber, 2000; Hiermann & Hfferer, 2003; Lang & Pigneur, 1999). Stafing decisions can be supported by matching algorithms (e.g., Colucci et al., 2003).
Expertise location. The aim of expertise location is to make experts accessible in a large community (like a company) based on what they know, not whom they know (Ehrlich, 2003). To this end, expertise location usually seeks a technological solution that makes a persons skills, experience, or background available in a proile that can be searched for by others. According to a review by Yimam-Seid and Kobsa (2003), expertise locator systems differ in the following aspects: (1) The basis for expertise recognition may be explicit (e.g., provided by the expert) or implicit (e.g., derived from document authorship or projects worked in). (2) When expertise indicators are then extracted from implicit sources, this can either be domain-knowledge driven (e.g., using a domain ontology) or domain-knowledge independent (e.g., analyzing word co-occurrences in documents). (3) They also differentiate different methods for expertise modeling. The model may be generated in query time (e.g., by using information retrieval techniques), it may be personal agent-based (that is, building local models for each expert), or associated with a centralized model.
Future Challenges for CM

From the review of CM approaches, we contend that there are several shortcomings with traditional approaches and thereby challenges for future research. The present work is an attempt to deal with theses challenges both from a theoretical and an empirical point of view. The following four research challenges have been identiied. Challenge 1: Disconnection between competence and performance level. Current approaches usually do not suficiently take into account the difference between competencies (as theoretical constructs) and performance (as observable behaviour). Speciically, automated approaches are based on an analysis of performance outcomes (e.g., word co-occurrences in written documents). However, there is usually a lack of theoretical conception of what it is they are actually measuring. On the other hand, centralized databases (such as the competency catalogue used in the Case Study I) usually lack the connection of the conceptual level (competency descriptions) to the performance level (tasks). This disconnect leads to unclear meanings of the competency descriptions and therefore to unclear assessment results. It also means that assessment is not connected to the working situations in which competencies are being applied. Challenge 2: Lack of empirical quality criteria. The lack of validation research in CM has long been criticized by several reviewers of the CM scene (Barrett & Depinet, 1991; Laber & OConner, 2000; Sarges, 2001). The reason for this apparent lack can be seen in the disconnection between competence and performance level. Without this explicit connection in the approaches, validation is dificult to achieve. A second reason is, of course, the additional resources that validation requires whenever it needs a separate evaluation study. Because CM is largely set in practical contexts, validation has always been second priority. However, as one can see in the case of assessment center research, validation is possible without additional effort, if competencies and performances are formalized to a certain degree.
Challenge 3: Lack of lexibility in the models and implementations. A review of CM practices by Athey and Orth (1999) brings them to suggest a trend in CM systems that allows them to respond more quickly to changing business needs. Quite clearly, current approaches, especially centralized database approaches, bring about large efforts for update and maintenance of the models (see also Case Study I). On the other hand, todays workplaces are increasingly based in an IT infrastructure. Artifacts that are produced are electronic, workplaces are networked, and work processes are modeled in computer systems. As a consequence, it becomes feasible to employ automated routines for competency modeling and assessment. As a recent review of CM systems shows, however, such systems are scarce (Lehner, 2004). Challenge 4: Convergence of HRM and KM approaches. This review has shown that HRM and KM approaches to CM have evolved separately. Especially in practical contexts, many arguments for a better alignment of HR and KM practices have been put forward (Carter & Scarbrough, 2001; Hansen, Nohria, & Tierney, 1999). In light of the increasingly scarce resources available for initiatives not directly related to the value-creating processes, the need to achieve a broader impact in an organization is of paramount importance. Finally, because they have traditionally emphasized different perspectives, an exchange of methods employed may prove to be beneicial.
Aim of the Present Approach

The aim of this research is to address the research challenges pointed out in the previous section. In addressing challenge 1, the present approach will apply a theoretical framework that is irmly based in behavioural theories of human competence and performance (see section Theoretical Framework). At the same time, the framework employs rigorous formalization so that validation becomes possible. We will develop several quality criteria that may be employed to test validity and reliability of the resulting models (challenge 2, see Evaluating CM). Because of the degree of formalization, the resulting models may be easily integrated in an IT environment and can be used to model and assess competencies in a work-integrated fashion, thereby offering automation potential (challenge 3, see section Implementing CM). Finally, the framework can be lexibly adapted to different purposes and organizational settings. Speciically, HRM and KM goals will both be accounted for (challenge 4). In order to facilitate practical application of the results, the seventh section (see Using CM) will deal with a scenario that illustrates transferability of the approach to organizational practice in work integrated informal learning. We close with an outlook and suggestions for further theoretical development (see inal section, Conclusion and Outlook).
Theoretical Framework
Human Competence and Performance
Behavioral research has long been concerned with the distinction between competence and performance. In the following discussion, we will especially focus on the conceptualization that has been established in organizational psychology. By looking at the it between persons and organizational conditions (such as attributes of the job, organizational structure or culture), organizational psychology seeks to predict individual performance from characteristics of the performer (Borman, Klimoski, & Ilgen, 2002). Such characteristics can either be rather stable and apply to very broad situational classes or be variable and narrower in their focus. Discussing the former ones, Brandsttter (1998) distinguishes motives, temperament, ability, and attitudes. The latter ones are usually said to encompass knowledge and skills (Schuler & Funke, 1998; van den Berg, 1998). In conceptualizing these characteristics, organizational psychology is built on research in cognitive, social, and personality psychology. In an applied setting, the KSAO model (knowledge, skills, abilities, and other characteristics) has been a widely accepted in personnel research (Lucia & Lepsinger, 1999; Schmitt & Chan, 1998). The KSAO model offers a distinction between the most basic human characteristics in terms of performance determinants. They are person-based; that is, they describe characteristics of people. Jobs are described in terms of KSAOs, but only in terms of the characteristics that people need to bring to the job in order to perform successfully. Knowledge and skills are usually believed to be more variable (and trainable) than are abilities and other characteristics. In the current approach, we deine competencies as personal characteristics of job holders that they bring to bear in different situations. Competencies are hypothetical constructs that determine performance in a job. The term performance is understood to encompass all behaviors relevant for the accomplishment of a certain task in a speciic situation (Schmitt & Chan, 1998). The term competency implies that it is used in a number of situations for a number of tasks. Performance is then produced by dynamically combining competencies according to the demands of the situation (e.g., Erpenbeck & Rosenstiel, 2003). We will differentiate competencies into more stable characteristics such as personality traits (or temperaments), motives, and cognitive abilities, and more variable characteristics, such as skills and knowledge. This differentiation is in line with a large body of research into KSAOs (see above), which we embrace especially because of its practical relevance (see Bina & Newkirk, 1998; Hegge-Klaiser, 1998; van den Berg, 1998). Relating to the process model presented above (see Figure 1), deining competencies (step 2) then is concerned with modeling the competence performance relationship; that is, selecting descriptors for competencies and relating them to performance. This results in an organization-speciic model. Assessing competencies (step 3) is the process of ascribing a certain competency to an individual that will be understood as a measurement task. Finally, evaluating the models (step 4) will be conceptualized as an evaluation activity that evaluates the competence performance model derived from step 2 as well as the measurement derived from step 3.
A Competence Performance Framework for CM

Although the models created in this approach are organization-speciic, it makes sense to derive these models from a general framework that formalizes the ideas presented in the previous section. The framework that has been chosen is based on the competence performance approach introduced by Korossy (1997, 1999) as an extension of knowledge space theory (Albert & Lukas, 1999; Doignon, 1994; Doignon & Falmagne, 1999; Falmagne, Doignon, Koppen, Villano, & Johannesen, 1990). Knowledge space theory was developed in the 1980s and 90s as an attempt to model a persons knowledge state as close as possible to observable behavior. It is predominantly concerned with the diagnosis of knowledge and has been applied in adaptive testing and tutoring scenarios and system (e.g., ALEKS Corp., 2003; Hockemeyer, Held, & Albert, 1998). The fundamental idea of knowledge space theory is that a persons knowledge state in a certain domain is seen as the set of problems this person is able to solve. Since solution dependencies exist among the problems, it is possible to present a person only a subset of all problems of a domain in order to diagnose his/her knowledge state. Knowledge space theory is concerned with uncovering and formalizing these dependencies. This is accomplished by introducing a surmise relation that relates each problem to the set of all of its prerequisite problems: The prerequisite set of problem a contains all problems for which the solution can be surmised form a solution of problem a. The surmise relation reduces the number of solution patterns that can be expected to occur to a smaller amount of knowledge states. The collection of all knowledge states is called a knowledge space. Knowledge spaces are closed under union, which is a reasonable proposition as one might reasonably expect that if two persons are in the knowledge state K1 and K2, respectively, that there exists a person whose knowledge would encompass both that of the other persons, and hence his/her knowledge state would be denoted by the union K1 K2. A possible limitation of knowledge space theory that has been put forward by Korossy (1997) is its sole focus on the behavioral (or performance) level. Knowledge space theory thereby neglects progress in cognitive psychology or educational sciences that have advanced theoretical understanding of the reasons for different levels of performance. Such cognitive theories of the underlying skills that shape performance offer better ways to suggest educational measures. As a result, Korossy suggests that in addition to the set of problems, one should look at the set of competencies, that is, knowledge, skills, and abilities needed to solve the problems. Similar to the set of questions, competencies are also structured in a competence space that results from a surmise relation on the set of competencies. The relationship between the two sets (questions and competencies) is formalized by an interpretation function that maps each problem to a subset of competence states that are elements of the competence structure. This subset of competence states contains all those competence states, in each of which the problem is solvable. The interpretation function induces a representation function that assigns to each of the competence states all problems that are solvable in that competence state. Which problems are solvable is determined by the interpretation function. We will next introduce an application of the competence performance approach to organizational competency management. This approach will be illustrated by a second case study.
Applying the Competence Performance Framework to Competency Management: An Illustrative Case Study
In applying the competence performance framework, we will make use of the distinction between competencies and performance that were deined in the previous section. These concepts will be used within the competence performance framework. Thus, the set of questions (performance) in the sense of Korossy will be the tasks that a certain employee performs in the workplace, or the tasks that are needed in a certain position. Tasks entail a number of goal-related behaviors that are exerted by employees and that lead to outcomes relevant for some organizational success criteria. The set of competencies encompasses all knowledge, skills, abilities, and other characteristics that are needed to successfully perform in the tasks. For formalizing the relationship between competencies and performance, the method of a competence performance matrix is introduced. This matrix assigns to each task all competencies needed to perform that task. This matrix thereby provides the interpretation function in the sense of Korossy, and a competence performance structure can be derived from it. To illustrate this procedure, consider the following case study (Case Study II), which took place in the same setting as the one reported previously, namely the automotive company. This time, however, the case study was conducted in the personnel unit. The interview was held with the head of the human resource development department, who supervises (among others) six human resource developers, who are in charge of implementing HR interventions in the other business units. The interview resulted in a competence performance matrix (see Figure 3). Hierarchical task analysis suggested by Armstrong (1999) was chosen for deriving a task list for the position of HR developer. In the speciic case, the method was used to identify 5 areas of overall responsibility for the job (see irst row in the table), which were then broken down into 23 single tasks (second row, denoted by numbers 1.1 to 5.5). From this list of tasks, a list of 18 competencies needed to perform the tasks was derived. These are given in the rows 3-25 and are denoted by letters A to R. From the interview data (tasks and competencies named by the respondent), a competence performance matrix was derived by the irst author and later cross-checked with the respondent and with one of the job incumbents. This matrix assigns to each of the tasks all competencies needed for successfully performing the tasks (denoted by the crosses in the table). In the terminology of the competence performance approach, these assignments correspond to the interpretation function (see the Theoretical Framework section). To illustrate how a competence performance structure is constructed from this matrix, we are using an excerpt of the total matrix (see Figure 4, left part). We have a set of 5 tasks Q={1.1, 2.1, 4.1, 4.2, 5.4}, for example, 2.1 Talk to managers of business unit about future needs, and a set of 5 competencies E={A, B, I, J, L, M, O, P}, for example, B. Structuring and solving problems. As mentioned above, the crosses in the matrix specify the interpretation function. More precisely, it is the minimal interpretation of the tasks in the set of competencies, i.e. the minimal set of competencies needed for successful performance. Note that this is a simpliication of Korossys approach, as the matrix allows only one set of competencies for the
Figure 3. Competence performance matrix for the position of HR developer in the automotive industry
minimal interpretation. As the sets of the minimal interpretation are competence states, each column denotes one competence state. We obtain the following collection of competence states: A={{O}, {B,J,L,O}, {J,O}, {A,B,J,O},{J}}. In order to obtain a competence space, A has to be closed under union, and the empty set and the set E have to be added. We obtain a competence space K={{}, {J}, {O}, {J,O}, {B,J,L,O}, {A,B,J,O}, {A,B,J,L,O}}. The competence space can also be depicted as a directed graph with the nodes representing the competence states and the edges denoting the subset relation. In Figure 4 (right part), the white boxes denote the competence states (the empty set has been omitted). The lines between the boxes denote the subset relation. The competence space is not closed under intersection: While the elements {B,J,L,O} and {A,B,J,O} are part of the competence space, the element {B,J,O} is not. The surmise system for the set of competencies is given by Table 1. The grey boxes below the white boxes in Figure 4 (right part) denote the performance state that can be accomplished in the competence state. These specify the representation function. Note that the performance structure is not a performance space as it is not closed under union, e.g. the element {1.1,5.4} is not part of the structure. This has occurred because the task 4.1 was assigned to a competence state, which is the union of two existing states. The competence performance structure depicted in Figure 4 (right part) can now be used for competency assessment and for development planning. If an employee performs well in task 4.1. Be present for employees in the business unit, it can be concluded that he/she will probably also perform well in tasks 1.1 and 5.4. If the same person was judged to perform poorly in task 2.1. Talk to managers of the business unit about future needs, then competency development measures might be suggested for competencies B. Structuring and solving problems and L. Resistance to uncertainty and stress. Accordingly, the advantage of the competence performance approach is that the competencies help to predict performance outcomes and provide an explanation for discrepancies in performance. For example, missing competencies can help to explain why an employee
Figure 4. Part of a competence performance structure for a human resource manager
was not able to accomplish a certain task. Hence, development programs can be created that focus on theses underlying competencies. Furthermore, the high level of formalization allows for the framework to be utilized in an IT environment. In such an environment, it becomes possible to draw inferences. The advantage of such formalization in our case is also that it does not require high measurement scale levels. This is especially important for empirical evaluation. As will be shown later on, all data analysis techniques are based on non-parametric procedures.
Conclusions
When comparing Case Study I in the second section (see Current CM Approaches: A Frame of Reference and a Case Study, a traditional approach to competency management) and Case Study II in the present section (a competence performance approach), it should be noted that the competence performance approach was introduced into this setting by simply adding an additional step into the traditional competency management process, that is, the construction of the competence performance matrix. This matrix was derived from data that was essentially available when following the traditional approach, namely tasks performed in the position and competencies required. The only thing needed in addition was a small amount of further formalization. This points to the relatively low additional demands when modeling for the irst time. Additionally, there is considerable adaptive potential when applying competence performance structures more frequently or when updating existing structures. Since in both modeling and assessment manual work is involved, practical applications might beneit greatly from (semi-) automated inferences made possible by this approach. This may be especially beneicial for more dynamic settings, that is, situations in which task requirements constantly change over time. In such settings, it is often the case that the work performed is to a large degree knowledge based or based on soft technology (Michellone & Zollo, 2000). As a result, mental operations become increasingly important, which makes both the assessment and the training of competencies more dificult. Furthermore, the management of competencies relies to a large degree on information from job incumbents themselves, as knowledge workers more often than not work in a self-directed manner and in team settings rather than being centrally controlled (Pinchot & Pinchot, 1996).
Table 1. Surmise system for the competence space in Figure 4
The applicability of the competence performance framework in more dynamic and knowledge-based settings will be explored with a further case study. First, competence performance structures are introduced into an organization (see next section). In terms of the CM process model, this intervention is concerned with steps 1-3. An evaluation of the results is then performed (see Evaluating CM: Quality Criteria for Competence Performance Structures). This corresponds to step 4 in the CM process model. The seventh section, Using CM: Work-Integrated Informal Learning, then deals with an application scenario for workplace learning.
Implementing CM: Introducing Competence Performance Structures

The work setting chosen was that of project managers in the research and development (R&D) ield (see also Ley & Albert, 2003b). The investigation (Case Study III) was conducted at the Know-Center, a competence center that carries out cooperative research projects with industry partners in the area of knowledge management. Topics of these projects are introducing knowledge management or e-learning systems in companies and providing necessary consulting services, composing technology studies, or developing software applications for knowledge management purposes. All topics covered have a strong IT component. In terms of introducing the structures, the case study employed two different strategies: First, the competency models were derived bottom-up, that is by asking job incumbents about their own jobs. Then, a top-down method was employed, that is supervisors of the jobs were involved in the modeling process. The contrast between top-down and bottom-up modeling has been broadly debated in job analysis research and has been identiied by Schmitt and Chan (1998) as one of the four most pressing issues in this domain. Case Study III also investigated the use of an alternative assessment procedure (step 3 of the process model). A standard procedure in competency management is such that supervisors rate employees on competency scales (see also Case Study I reported in the Current CM Approaches: A Frame of Reference and a Case Study section). Since many biases have been reported with this method (for example, low convergent validity in Assessment Center research), the present case study suggests a different approach. Supervisors were asked to rate employees on how they performed in different tasks (performance level) and the competence state was then derived through the use of competence performance structures as exempliied in Theoretical Framework. This result was compared to a standard procedure of directly rating employee competencies.
Step 1: Analyzing Setting and Purpose of CM

In the present investigation, this step relied on questioning the Know Center management team about purpose and setting, as well as the goals for competency management. The result of this phase was a business case that saw three purposes for a CM initiative: CM was
supposed to (1) provide input for strategy formulation and align projects and processes to this strategy, (2) enable competency transparency across different groups, and (3) provide insight into the job requirements of project managers to enable a more effective development of their expertise. Important elements of the setting include: (1) the work setting, which can be characterized as being knowledge-based, project-based, and dynamic, that is, project requirements are constantly being negotiated between clients and Know Center staff and only few of these requirements can be planned in advance, and (2) the personnel development setting, which can be characterized as being informal, self-initiated, and work-integrated. It was decided to focus on the project managers for these investigations. It appeared to be necessary to involve job incumbents themselves in the modeling process, as they usually know better about their speciic ields of expertise. On the other hand, it also seemed to make sense to question the supervisors of these job incumbents (i.e., the department heads), as they should have a better understanding about strengths and weaknesses of different job incumbents in the position.
Step 2: Deining Competencies

For this step, interviews were held with seven job incumbents (project managers) and two supervisors of the position (department heads). As a irst step, performance outcomes had to be deined. In the present investigation, we chose to use documents that had been written by the project managers as the performance outcomes. At the Know Center, as well as in many other similar companies, the documents produced by the workforce give a good indication of past job requirements. Because work outcomes are mostly knowledge-based, the result is often a written document that gives account of the work performed. Documents also relect the dynamic work assignments. In the case of the project managers, documents they had produced themselves were used for the analysis. Table 2 gives the document types that were used and the number of documents named by job incumbents. In case of the department heads, documents they had supervised were used. Supervisors were asked to think of one or two best practice examples from the document types mentioned in the table, that is, documents they thought of as especially good examples of those they had supervised in the past. Competencies were elicited by asking for knowledge and skills used for creating the documents. These competencies went well beyond pure text production skills. Rather, it was always stressed that all knowledge and skills should be considered that were needed in the process of producing the document (e.g., research skills, communication skills, or the skills necessary to analyze organizational processes if this happened to be the topic of the particular document). For elicitation purposes, repertory grid technique (RGT) was used. RGT is well suited for ideographic research designs. It has been used for competency elicitation in previous investigations (for a review, see Jankowicz, 1990) as well as in the context of the competence-performance approach (Schweitzer, 2001). Each respondent then considered all of his/her documents and all competencies he or she had named. To each document, the respondent assigned all competencies he or she had used
when creating the document (document-competency assignment). Accordingly, a matrix was obtained for each respondent, which consisted of n m assignments, where n was the number of documents (performance) and m is the number of competencies named by that person. The cells of the matrix are illed with an X, which indicates that a certain competency was used in creating the document. These assignments establish the minimal interpretation of the documents in the set of competencies in the sense of the interpretation function k. A content analysis was then performed with both the job incumbent as well as the supervisor competency descriptions. Table 3 shows the resulting competence performance matrix for the job incumbents, and Table 4 shows the results for the supervisors. In both tables, competency categories are the result of the content analysis. The document types denote the ones mentioned in Table 2; the type Other has been omitted. Comparing the tables, it can be observed that all competency categories derived from job incumbents interviews were also replicated in the supervisor interviews. Twenty-six of the 32 uniquely identiied competency descriptions (81%) from supervisors had already been a topic in the interviews with job incumbents. This shows a general tendency of agreement about the important competencies needed by incumbents of the job. Differences can be found in ive categories (Nos. 30-34 in Table 4) that were obtained only from the supervisors data. Additionally, differences in the importance placed on certain categories can be observed by looking at the number of concepts that were named in each of the categories (comparing column 3 in Table 3 and Table 4). Accordingly, the competency category more important to supervisors is result and application orientation (#30), which seems to relect the particular viewpoint of the department heads in putting a strong focus on the external or internal clients. The categories more important to employees are information structuring, synthesis, and organization (#11), project management (#18), logical argumentation and reasoning (#14), and self and time management, work eficiency (#17).
Table 2. Types of documents used in case study III
From the matrices, the surmise relation on the set of competency categories can be derived that shows the prerequisite sets of the competency categories. The relation can be depicted as a directed graph (see Figure 5), which shows the two relations based on the same set of 16 competency categories for job incumbents and supervisors (#10-25). The boxes denote the competency categories and the lines between the boxes denote the surmise relation, that is, a competency that is connected to another by lines going upward; the irst one is a prerequisite for the second. It is readily apparent from the two graphical representations that the surmise relations are quite different. In fact, the two groups only agreed on six, namely 12S19, 13S19, 15S24, 20S24, 18S25, and 22S21. For job incumbents, 36 pairs of competencies are in the relation. For supervisors, the number is 63which is almost twice as high.
Table 3. Categorized competency document-type matrix
Table 4. Categorized competency document-type matrix obtained from supervisor data
Step 3: Assessing Competencies

A performance assessment task was employed as an alternative to the standard assessment process (direct competency rating). Speciically, supervisors were asked to rate the performance of all of their employees in producing the document types mentioned in the previous step. Because the document types are interpreted in the set of competency categories by means of the competence performance structures (see previous step), it is possible to derive a persons competence state from knowing his/her performance state. Performance in this case means which kinds of document types the employee can successfully produce.
In order to compare this performance assessment to standard practices, department heads were also asked to produce a direct competency assessment. They were asked to indicate for each employee and each competency whether the employee as a project manager fulilled minimal requirements for this competency, or whether he/she did not or over-fulilled the requirements. So here a rating in relation to implicit job requirements was utilized. It was hypothesized that for each employee, the performance assessment would yield a competence state that corresponds to the set of competencies assigned to this employee in the direct competency assessment task. The results of the performance assessment can be found in Table 5. The ranking of the employees (A to H) are given for each of the two department heads. The value 0 denotes that the employee was not thought to fulill requirements of this task; 1 means that the employee did fulill the requirements, but was not found to be suitable. The rest of the employees were then ranked starting from 2 (lowest rank).
Evaluating CM: Quality Criteria for Competence Performance Structures

Evaluation is concerned with establishing the quality of the models and the assessment obtained in the previous step. We are employing two different evaluation rationales for the two types of quality assessments involved in competency management: the question of the degree of quality of the model (step 2: competency deinition) and the question of the degree of quality of the assessments (step 3: competency assessment). In the irst case (competency deinition), we are relying on techniques involved in validating a model or validating a theory (Brewer, 2000; Cook & Campbell, 1979). Speciically, techniques of statistical conclusion validity (e.g., reliability of the modeling techniques) and construct validity are being employed. Criteria for measuring the quality of the models will be suggested in the next section.
Figure 5. Surmise relations on the set of competency categories
In the second case (competency assessment), the validity rationale is that used in classical test theory (John & Benet-Martinez, 2000; Rost, 1996). Objectivity, reliability, and validity of the measurement are being investigated. Criteria for the quality of the assessment will be discussed in Evaluating CM: Quality Criteria for Competence Performance Structures. In addition to these theoretical requirements, selecting suitable criteria requires adherence to the practical demands of organizational settings. The most important issue to be raised is the amount of effort that is required for evaluation. Practical considerations demand that measuring the quality criteria makes use of existing data and does not require additional effort for the organization. Ideally, the criteria can be assessed directly within the modeling process and do not require any additional data to be collected.
Evaluating Competency Deinition

To evaluate competency deinition, reliability and validity of the document competency assignments in the document competency matrix are explored. Furthermore, we suggest a way to compare the competence spaces of two groups (job incumbents and supervisors in the present case). Reliability of the document competency assignments. Reliability of competency assignments in the matrices concerns the question of how stable the assignments are over time or between different raters. In Case Study III, two document competency assignments had been obtained from the respondents. The irst one was obtained in the interview using the competencies generated by each respondent. The second one was obtained after a six-month interval when respondents assigned competencies to their documents from the set of all competencies generated in all the interviews. So for a subset of competencies, a test-retest reliability coeficient can be obtained. To calculate the reliability coeficient, r suggested by Goodman and Kruskal (1954) for reliability estimates in cross classiications was used. It is based on the probability of correctly guessing the second assignment when the irst one is known. The results for the seven job incumbents vary between 0.125 and 0.659. With 0.494, the overall coeficient is of moderate magnitude.
Table 5. Ranking provided by two supervisors
In the case of the supervisor data, a test-retest coeficient was calculated from assignments obtained from within the repertory grid interview and the assignments obtained from the assignment after the interview. Of the 138 assignments (69 for each respondent) in the interviews, 108 (77.7%) were found to be identical in the document competency assignment later on. This resulted in a highly signiicant r=0.524 (z=6.203, p<0.001). The value shows a fairly high amount of internal consistency of the raters. Finally, inter-rater reliability could be calculated between the two supervisors. Six of the eight document types were rated by both respondents, and they were rated on 32 competencies. Therefore, 192 assignments (6x32) were used to calculate this coeficient. It turned out that 128 assignments (66.7%) were identical between the raters. This resulted in r=0.179 (z=2.058, p<0.05). While being signiicant, the coeficient is quite low. It seems that as long as assignments within a rater are considered, the reliability is satisfactory. Reliability between raters is much lower. Validity of document competency assignments. Validity can be established by relating competencies and performance to constructs that have already been established for the organization (like organizational structures or processes). Thereby, we relate the results of the investigations to the organizational view of competencies and performance, that is, the view shared among members of the particular organization. To give an example: The organization has created several business units that can be differentiated on the basis of the topics of their projects. The business units, the associated projects as well as the documents produced within these projects constitute elements of the shared view within the organization. If the competencies that have been elicited within the investigation have any signiicance for the organization, then it should be possible to replicate which documents correspond to which business units from the document competency assignments. We have chosen to consider construct validity, which looks at different operationalizations of the same construct and seeks to establish convergent and discriminant evidence (see Brewer, 2000; Cook & Campbell, 1979). Convergent validity shows a high correlation between different operationalizations of the same construct; discriminant validity shows low correlations between operationalizations of different constructs. Speciically, we have chosen two ways to categorize the documents used in the investigation. The irst way categorizes the documents according to the business units that have produced the document. This categorization should be related to the domain-speciic competency assignments, since these should be speciic to the business units. On the other hand, we have categorized the documents according to document type, namely to whether the document was a formal project documentation (document types 6-8 in Table 2), or whether it was a project result (document types 1-5 in Table 2). This categorization should be related to the general job competency assignments, since for producing formal project documentation similar competencies should be used across business units because the requirements are the same. For testing whether the competency assignments corresponded to the a-priori classes of documents, a classiication evaluation was performed. Since only one sample was available for evaluation, the leave-one-out method suggested by Lachenbruch (see Bortz, 1993) for evaluating cluster analytic solutions was used. This is performed in the following way: The clusters are made up of the total set of documents classiied according to the a-priori classes (according to business area or document type). One document is taken out of the
total set. This document is then classiied to the cluster to which the distance is smallest. The clusters are made up of the rest (n-1) documents. Distance to a cluster is measured by euclidian distance of the attribute vector of the document to the centroid of the cluster, that is, the mean attribute vector of the cluster (nearest centroid rule). For each document that is taken out, it is established whether the classiication leads to the same as the a-priori class for that document. From this, a 2x2 table can be derived that gives the number of correct and incorrect assignments, with N being the number of documents. Cohens coeficient (Bortz, 1993) was then used to evaluate the solution. This coeficient relates the proportion of all objects correctly classiied by the procedure (po) to that proportion of objects that is expected by chance to be classiied correctly (pe): Table 6 shows the results. As expected, the business area classiication could be well replicated, especially by using domain-speciic competencies that establish convergent evidence. is still signiicant but much lower when general job competencies are used for classiication. This establishes divergent evidence. It also seems to indicate, however, that in the different business units, different kinds of general competencies are reportedly being used when documents are created. In fact, a closer analysis on a per-competency level (conditional probabilities for assigning a competency given a document type) shows that for the knowledge management area, two general competencies related to economic utilization (deriving economic utility from results and knowledge of Know Center strategy) are more likely to be assigned across all documents than in the e-learning area. This suggests that document competency assignments are sensitive to the general orientation of a business unit that might relect organizational values and priorities (Green, 1999; see Current CM Approaches: Reviewing Purposes and Models). The document type classiication could not be replicated by either domain-speciic nor general competencies, and in both cases was not signiicant. Apparently, use of general job competencies does not differentiate between formal documentation and project results. In fact, when looking at the single competencies, it appears that only very few differ with respect to the frequency of occurrence between the document types. Validity of competence space: Comparing job incumbents and supervisors. While what has been discussed so far dealt with document competency assignments, what follows will be looking at the validity of the competence space or the surmise system. The measure suggested will help to explore whether hypotheses about the relationships among the set of competencies, which are speciied through the surmise relation, are valid. Validating these hypothesized relationships is especially important when one is concerned with employing adaptive methods for competency deinition or assessment. Such methods are only permissible if the hypothesized relationships among the set of competencies are in fact valid. Comparing job incumbent to supervisor ratings of jobs has often been used as a way to validate assignments in job analysis research (e.g., Wilson, 1997). The present investigation permits drawing comparisons between the results on the job-incumbent level and the supervisor level. A irst rough comparison of the surmise relations (for competencies generated by job incumbents and supervisors) was already presented in Figure 5. In addition to this, an index measuring the suitability of the surmise relations in terms of the it to the data makes use of the index suggested by Goodman and Kruskal (1954). This is achieved by confronting the surmise relations derived from the two groups with the answer patterns (document competency assignments) from the other group.
Table 6. Results for the leave-one-out
This is done by looking at all pairs of competencies that are in the relation (e.g., the one derived from job incumbent data; see right part in Figure 5). For each of these pairs, each vector from the supervisor assignments can be analyzed as to whether the document competency assignments correspond to the relation (concordant pair) or whether they are in conlict with the relation (discordant pairs). To estimate the magnitude of the overall suitability of the surmise relation, the number of concordant and discordant pairs (Nc and Nd respectively) can be related using the index. This order index is used to determine whether two classiications are ordered likely or unlikely:
Nc Nd Nc + Nd
Table 7 shows the results that clearly support the earlier assertion about the poor it between the views of job incumbents and supervisors. In both cases, the number of concordant assignments does not signiicantly exceed the number of discordant assignments. Conclusions. When we consider the results reported in this section, it becomes clear that the question of the quality of the structures obtained in the investigation is not a straightforward one. When the degree of reliability and validity of the assignments is considered, we may regard it to be adequate. Looking at the validity of the competency relationships (formalized through the surmise relation), the picture is less clear. Comparing the structures obtained from supervisors and job incumbents does not result in signiicant agreement. It seems that the relations are much more susceptible to variations than are the assignments. For example, it is well known in the competence performance approach that the surmise relation on the set of competencies is heavily dependent on the set of documents used.
Evaluating Competency Assessment

The present section deals with evaluating the competency assessment, which is based on the models evaluated in the previous section. As we have discussed previously, two types of assessment judgments had been required from the department heads in Case Study III. One was a direct assessment of the competencies of the employees, which can be considered as a standard procedure in competency management. The other type of judgment was an assessment of their performance, that is, the quality of the documents produced by the workforce. The results of the two judgments were compared by collecting the following three types of evidence: (1) consistency between the two judgments, (2) distance of competency assessment and competence space, and (3) differences in direct and derived competency assessment. For validation purposes, all three procedures seek to establish convergent evidence, that is, a large correlation between the two assessment procedures. Consistency between competency and performance assessment. A irst analysis looks at whether these two assessment types were consistent. It is important to ensure that the performance assessment was based on employee competencies (and not on some other criteria), and that the most important competencies to make the performance ranking were present. This can be done by looking at whether the employee ranking obtained from the department heads (performance assessment) can be reproduced by the competency assessment for the same employees. For each document type, a ranking of employees was obtained from the competency assessment. This ranking was then correlated with the ranking of employees from the performance assessment. Spearmans rank correlation coeficient r resulted in correlations ranging from 0.94 to 0.35. Six out of the eight correlations were signiicant. This means that generally speaking, the employee ranking provided by the supervisors was based on the competencies assigned to the employees. Consistency of competency assessment and surmise relation. The second analysis looked at the surmise relation derived from the department heads document competency assignments (see previous section on Validity of competence space: Comparing job incumbents and supervisors) and compared the competency assessment to this relation. Again, the index was used to examine whether the two department heads provided competency assessments in line with the relation. The resulting index was -0.08, which shows that the competency assessment was not consistent with the surmise relation. The per-participant indices range from 0.33 to -0.63. These indices also suggest that judgments of one of the supervisors it the surmise relation better than those of the other.
Table 7. value for the it of the surmise relations
Difference between direct and derived competency assessment. Because performance is interpreted in the set of competencies by means of the competence performance structure, an employees competence state can be derived from the performance assessment. This is done for each employee by taking the union of all competence states in which successful performance is observed, that is, the employee is judged able to produce certain documents. The result is again a competence state, since it is a union of other competence states. This state is the one hypothesized to be the employees competence state. In Case Study III (see Implementing CM: Introducing Competence Performance Structures), the suitability of this procedure can be validated by comparing the outcome with the direct competency assessment. The direct competency assessment for each employee consists of the set of competencies that he/she is judged to possess. The set difference between this set and the hypothesized competence state should be small. When the set difference is obtained for the present investigation for each employee, the numbers range between 0 and 16 (from a maximum of 32). The average set difference is 9.6, which indicates that the two sets differ considerablyon average more than 25% of the competencies in the two sets are different. Because this irst procedure was not promising, an alternative assessment procedure was tested. This time all performance, that is, performance judged to be positive (as above) as well as performance judged to be negative was taken into account. A simple decision rule was employed: any time a competency was more often assigned to the documents judged to be positive than to those judged to be negative, the competency was hypothesized to be possessed by that employee. When the two procedures are compared by means of Cohens coeficient, increases to 0.255 (N=256) from 0.035 (N=256) when both positive and negative performance is taken into account. Although the agreement is not really large, the increase is promising. A problem with this second approach, however, is that it results in sets that are not competence states. Currently, the competence performance approach does not offer a means for dealing with this challenge. Conclusions. Competence performance structures build on the idea of a close connection between tasks and competencies and an explicit modeling of the relationship. In practical applications, this makes it possible to integrate competency assessment into the work processes, as competencies can be inferred from task performance. Results of the current investigations suggest that individual judgments of task performance are, in fact, based on the underlying competencies. However, a closer examination is needed in terms of the agreement of competency and performance assessment. Initial promising results could be gained by considering both positive and negative performance. This makes further developments in the competence performance approach necessary.
Using CM: Work-Integrated Informal Learning

In the previous two sections, questions of implementing and evaluating CM based on the competence performance approach have been introduced. The present section will suggest a scenario of how this approach can be used in integrated informal workplace learning.
The scenario is based on the AD-HOC methodology introduced in Lindstaedt, Farmer, and Ley (2004). AD-HOC seeks to introduce a unifying structure to allow access to content that is already available in a company IT infrastructure but possibly in many different underlying IT systems (e.g., organizational memory systems, worklow systems, e-learning systems, or on ile or e-mail servers). In addition to this technological heterogeneity, the structures used to organize the content in these systems are also heterogeneous. Content may be structured according to work tasks and processes, to learning goals (e.g., e-learning systems), or it might have a thematic structure (e.g., organizational memory systems). This makes it dificult for an employee to ind suitable information. Figure 6 shows an example of an AD-HOC environment that was developed at the Know Center and is currently being used by the project managers for supporting general project management processes. In the example, support for writing a inal project report is shown, a documentation step in the project close-out process. The unifying structural element in this case is the work process (top left). For each process step, documents are available that are needed for executing the task (e.g., templates or checklists) or learning about the tasks at hand (e.g., how-to descriptions or references to reports from former projects). In this step, a worklow is also integrated, in which the result of the process (the inal report) is automatically routed to the department head for review. The department head rates the performance of the employee and attaches a comment to provide feedback to the author. This last point is an example of integrating the assessment into existing work processes. When the tasks represented in this process are formalized in a competence performance structure, competence assessments can be derived and constantly updated from ongoing performance assessment as described in Implementing CM: Introducing Competence Performance Structures. Competency development in this case is achieved by offering the opportunity for the employee to relect on the outcomes of a task and by providing feedback from a more experienced person. Relating back to Case Study III reported in the Implementing CM: Introducing Competence Performance Structures section, a further step can be thought of to provide support for informal workplace learning in the AD-HOC environment. The document competency matrix introduced there was derived by asking project managers which competencies (knowledge and skills) they had used when producing certain documents. Competencies were both general (e.g., communicating in a team setting or with partner companies) and domain speciic (e.g., knowledge of streaming technologies). While the former can be used for feedback purposes as introduced in the previous paragraph, the latter ones may be used in the context of self-directed learning from existing documents as presented next. It is plausible that the AD-HOC environment would provide the facility to not only route formal project documentation but also to archive all project results and make them available to other employees for learning purposes. For example, if an employee has inished a certain project and archives the results in the organizational memory, the environment might pose a few questions similar to the assignment tasks asked from project managers in the interviews. The environment would then place the document into a competence performance structure. This structure might look similar to that in Figure 7, which shows part of the structure that was created from the data provided by the project managers in the interviews (see Implementing CM: Introducing Competence Performance Structures). The part that is shown in the igure focuses on a subset of knowledge used in e-learning projects. The structure was
visualized using formal concept analysis (Ganter & Wille, 1999), which creates concepts (the nodes in the graph) that consist of subsets of objects (documents) and subsets of attributes (competencies). Two documents can be seen in the structure that are given in bold (#5 and #7). The other labels denote competencies. Competencies used for producing a document can be found by following the path upward in the graph. From the graph, relationships between competencies are readily apparent. For example, we ind technological knowledge (Accessing Web services, Structuring metadata, and Streaming technologies) closely related, as these were evidently applied in similar contexts. Also, the two documents are related, as document seven (a publication on adaptive competence testing) used a subset of competencies that was used for document ive (a publication on learner models). When supporting a self-directed search in the document repository, these relationships can be exploited: For the author of document seven who is searching for information on technologies used in e-learning, document ive might be a valuable learning resource. The structures may also be of use when metadata is assigned to newly created documents. The relationships in the structure suggest that when certain competencies are assigned as metadata, certain other competencies seem more likely than others.
Figure 6. AD-HOC Prototype: Support during the project close-put process
In contrast to competence performance structures, formal concept analysis creates structures stable with regard to intersection. We used this type of visualization, as there is visualization software available for visualizing large structures. However, competence performance structures are more related to learning applications where one would argue that learning happens through the acquisition of new competencies. The competence state of a person (i.e., the subset of competencies available to him or her) can be derived from a union of all competencies used for creating his/her documents. Such adaptation to the learners context then facilitates support for competency development by making suitable content available.
Figure 7. Formal context visualizing document competency assignments
Conclusion and Outlook

The review of the state of the art has indicated that there exists a special need for competency management methods in the area of work-integrated competency assessment and development. Currently, most methods that are being employed are separated from the work processes. First experiences with the competence performance approach have attested to the idea that it may ill this gap. The fact that the structures have been shown to relect different situational demands of employees working in the analyzed positions support the claim that it is better able to relect the close connection between tasks and competencies than traditional approaches. In terms of applicability for organizational settings, the investigations suggest that both the implementation and evaluation methods are feasible. Speciically, it could be demonstrated that only little further effort is needed to extend traditional competency management methods when the competence performance approach is introduced. Furthermore, the higher degree of formalization makes it possible to introduce evaluation methods that are also integrated in existing work processes and do not require large additional efforts. In the last section, we have suggested a scenario of how competence performance structures might be applied in work-integrated informal learning. There still remain open issues with regard to this application. We are currently seeking to address these issues within a large European research effort1. We seek to investigate how competence performance structures (among other things) can be used to provide information about a users context when he/she is looking for support while working in a speciic business process and in a speciic domain of work. With regard to future research into the competence performance approach, we suggest the design of interactive questioning tools for modeling competence performance structures and assessing competencies. Both these procedures can utilize the adaptive potential that is offered by the competence performance framework so as to relieve the user from manual modeling tasks. A further direction of research should investigate the utility of automated text mining procedures, such as those built on the vector space model (see Lee, Chuang, & Seamos, 1997). These procedures are complementary to the competence performance approach, as they are data driven and may thereby generate hypotheses about the competencies used in a large collection of documents.
Acknowledgments
The Know Center is funded by the Austrian Competence Center program Kplus under the auspices of the Austrian Ministry of Transport, Innovation and Technology (www.ffg.at) and by the State of Styria.
References
Albert, D., & Lukas, J. (Eds.). (1999). Knowledge spaces: Theories, empirical research, and applications. Mahwah, NJ: Lawrence Erlbaum Associates. ALEKS Corp. (2003). ALEKSA better state of knowledge. Retrieved October 13, 2003, at http://www.aleks.com Anderson, J. R. (2000). Cognitive psychology and its implications. New York: Worth Publishing. Armstrong, M. (1999). A handbook of human resource management practice. London: Kogan Page. Athey, T. R., & Orth, M. S. (1999). Emerging competency methods for the future. Human Resource Management, 38(3), 215-225. Baitsch, C. (1998). Lernen im Prozess der Arbeit - Zum Stand der internationalen Forschung. In Arbeitsgemeinschaft Qualiikations-Entwicklungs-Management (Ed.), Kompetenzentwicklung 98: Forschungsstand und Forschungsperspektiven (pp. 269-337). Mnster: Waxmann. Barrett, G. V., & Depinet, R. L. (1991). A reconsideration of testing for competence rather than intelligence. American Psychologist, 46(10), 1012-1024. Bina, M., & Newkirk, J. (1998). Competencies at the rock - Creating a competency-based integrated human resource system for Prudential health care group. In D. D. Dubois (Ed.), The Competency Case Book (pp. 93-120). Amherst: HRD Press. Borman, W. C., Klimoski, R. J., & Ilgen, D. R. (2002). Stability and change in industrial and organizational psychology. In W. C. Borman, D. R. Ilgen, R. J. Klimoski, & I. B. Weiner (Ed.), Handbook of Psychology: Industrial and Organizational Psychology (pp. 1-17). New York: John Wiley & Sons. Bortz, J. (1993). Statistik fr Sozialwissenschaftler. Heidelberg: Springer. Bortz, J., & Dring, N. (2002). Forschungsmethoden und Evaluation. Heidelberg: Springer. Boyatzis, R. E. (1982). The competent manager. New York: Wiley. Brandsttter, H. (1998). Persnliche Verhaltens- und Leistungsbedingungen. In H. Schuler (Ed.), Organisationspsychologie (pp. 213-233). Bern: Hans Huber. Brewer, M. B. (2000). Research design and issues of validity. In H. T. Reis & C. M. Judd (Eds.), Handbook of Research Methods in Social and Personality Psychology (pp. 3-16). Cambridge: Cambridge University Press. Carter, C., & Scarbrough, H. (2001). Towards a second generation of KM? The people management challenge. Education + Training, 43(4/5), 215-224. Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F. M., Mongiello, M., & Mottola, M. (2003). A formal approach to ontology-based semantic match of skills descriptions. Journal of Universal Computer Science, 9(12), 1437-1454. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation. Boston: Houghton Miflin.
Davenport, T. H. (1997). Knowledge management case study: Knowledge management at Microsoft. Retrieved March 12, 2005, at http://www.mccombs.utexas.edu/kman/microsoft.htm Deiters, W., Lucas, R., & Weber, T. (2000). Skill-Management, ein Baustein fr das Management lexibler Teams. Information Management and Consulting, 15(3), 54-60. Doignon, J. (1994). Probabilistic assessment of knowledge. In D. Albert (Ed.), Knowledge structures (pp. 1-52). New York: Springer. Doignon, J., & Falmagne, J. (1999). Knowledge spaces. Heidelberg: Springer. Ehrlich, K. (2003). Locating expertise: Design issues for an expertise locator system. In M. Ackerman, V. Pipek, & V. Wulf (Ed.), Sharing expertise: Beyond knowledge management (pp. 137-158). Cambridge, MA: MIT Press. Elkjaer, B. (2000). Learning and getting to know: The case of knowledge workers. Human Resource Development International, 3(3), 343-359. Eppler, M. J. (2001). Making knowledge visible through intranet knowledge maps: Concepts, elements, cases. In Proceedings of the 34th Hawaii International Conference on System Sciences (HICSS-34). Los Alamitos: IEEE Computer Society. Erpenbeck, J. (2003). KODEKompetenz-diagnosik und entwicklung. In J. Erpenbeck & L. v. Rosenstiel (Ed.), Handbuch Kompetenzmessung (pp. 365-375). Stuttgart: Schffer-Poeschl. Erpenbeck, J., & Heyse, V. (1999). Die Kompetenzbiographie: Strategien der Kompetenzentwicklung durch selbstorganisiertes Lernen und multimediale Kommunikation (Vol. 10). Mnster: Waxmann. Erpenbeck, J., & Rosenstiel, L. v. (2003). Einfhrung. In J. Erpenbeck & L. v. Rosenstiel (Ed.), Handbuch Kompetenzmessung (pp. IX-XL). Stuttgart: Schffer-Poeschl. Evans, K., & Kersh, N. (2004). Recognition of tacit skills and knowledge. Journal of Workplace Learning, 16(1/2), 63-74. Faix, W. G., Buchwald, C., & Wetzler, R. (1991). Skill-Management: Qualiikationsplanung fr Unternehmen und Mitarbeiter. Wiesbaden: Gabler. Falmagne, J., Doignon, J., Koppen, M., Villano, M., & Johannesen, L. (1990). Introduction to knowledge spaces: How to build, test, and search them. Psychological Review, 97(2), 201-224. Ganter, B., & Wille, R. (1999). Formal concept analysis: Mathematical foundations. Heidelberg: Springer. Garavan, T. N., Morley, M., & Gunnigle, P. M. D. (2002). Human resource development and workplace learning: emerging theoretical perspectives and organisational practices. Journal of European Industrial Training, 26(2, 3, 4), 60-71. Goodman, L. A., & Kruskal, W. H. (1954). Measures of association for cross classiications. Journal of the American Statistical Association, 49, 732-764. Green, P. C. (1999). Building robust competencies: Linking human resource systems to organizational strategies. San Francisco: Jossey-Bass. Hacker, W. (1998). Allgemeine Arbeitspsychologie. Bern: Hans Huber.
Hansen, M. T., Nohria, N., & Tierney, T. (1999). Whats your strategy for managing knowledge? Harvard Business Review, 77(2), 106-116. Hegge-Klaiser, C. (1998). Applying competency-based selection and training to Reengineering. In D. D. Dubois (Ed.), The competency case book (pp. 257-270). Amherst, MA: HRD Press. Hiermann, W., & Hfferer, M. (2003). A practical knowledge-based approach to skill management and personal development. Journal of Universal Computer Science, 9(12), 1398-1409. Hockemeyer, C., Held, T., & Albert, D. (1998). RATHA relational adaptive tutoring hypertext WWW: Environment based on knowledge space theory. In C. Alvegrd (Ed.), Computer Aided Learning and Instruction in Science and Engineering, Proceedings of the CALISCE98 (pp. 417-423). Gteborg: Chalmers University of Technology. Jankowicz, A. D. (1990). Applications of personal construct psychology in business practice. In G. J. Neimeyer, & R. A. Neimeyer (Eds.), Advances in personal construct Psychology (pp. 257-287). Greenwich: JAI Press. John, O. P., & Benet-Martinez, V. (2000). Measurement: reliability, construct validation, and scale construction. In H. T. Reis & C. M. Judd (Ed.), Handbook of research methods in social and personality psychology (pp. 339-369). Cambridge: Cambridge University Press. Klemp, G. O., & McClelland, D. C. (1986). What characterizes intelligent functioning among senior managers. In R. J. Sternberg & R. K. Wagner (Ed.), Practical intelligence: Nature and origins of competence in the everyday world (pp. 31-50). Cambridge: Cambridge University Press. Korossy, K. (1997). Extending the theory of knowledge spaces: A competence-performance approach. Zeitschrift fr Psychologie, 205, 53-82. Korossy, K. (1999). Modeling knowledge as competence and performance. In D. Albert, & J. Lukas (Eds.), Knowledge spaces: Theories, empirical research, and applications (pp. 103-132). Mahwah, New Jersey: Lawrence Erlbaum Associates. Laber, M. E., & OConner, J. K. (2000). Competency modeling: Ready, set, research. The Industrial-Organizational Psychologist, 37(4). Retrieved August 07, 2005, at http:// www.siop.org/tip/backissues/TipApril00/14Rogelberg.htm Lang, A., & Pigneur, Y. (1999). Digital trade of human competencies. In Proceedings of the 32nd Hawaii International Conference on System Sciences (HICSS 1999). New York: IEEE. Lee, D. L., Chuang, H., & Seamos, K. (1997). Document ranking and the vector space model. IEEE Software, 14(2), 67-75. Lehner, F. (2004). Marktanalyse zum angebot an skill-management-Systemen. In P. Dadam & M. Reichert (Eds.), Informatik 2004Informatik Verbindet, Proceedings of the 34th Annual Meeting of the German Society for Informatics (Vol. 2, pp. 317-322). Bonn: Gesellschaft fr Informatik. Ley, T., & Albert, D. (2003a). Kompetenzmanagement als formalisierbare Abbildung von Wissen und Handeln fr das Personalwesen. Wirtschaftspsychologie, 5(3), 86-93.
Ley, T., & Albert, D. (2003b). Identifying employee competencies in dynamic work domains: Methodological considerations and a case study. Journal of Universal Computer Science, 9(12), 1500-1518. Ley, T., & Albert, D. (2004). Quality criteria for competency assignments: Examples from a project management case study. In P. Dadam & M. Reichert (Eds.), Informatik 2004Informatik Verbindet, Proceedings of the 34th Annual Meeting of the German Society for Informatics (Vol. 2, pp. 342-346). Bonn: Gesellschaft fr Informatik. Lindstaedt, S. N., Farmer, J., & Ley, T. (2004). Betriebliche Weiterbildung. In J. Haake, G. Schwabe, & M. Wessner (Eds.), CSCL-Kompendium- Lehr- und Handbuch zum computeruntersttzten kooperativen Lernen (pp. 423-434). Mnchen: Oldenbourg. Lindstaedt, S. N., Ley, T., & Mayer, H. (2005, November 10-11). Integrating working and learning in APOSDLE. In Proceedings of the 11th Business Meeting of the Forum Neue Medien, University of Vienna, Austria. Graz/Austria: Forum Neue Medien. Lucia, A. D., & Lepsinger, R. (1999). Competency models: Pinpointing critical success factors in organizations. San Francisco: Jossey-Bass. McClelland, D. C. (1973). Testing for competence rather than intelligence. American Psychologist, 28, 1-14. Michellone, G., & Zollo, G. (2000). Competencies management in knowledge-based irms. International Journal of Technology Management, 20(1/2), 134-155. Pinchot, G., & Pinchot, E. (1996). The rise and fall of bureaucracy. In P. S. Myers (Ed.), Knowledge management and organizational design (pp. 39-54). Newton/MA: Butterworth-Heinemann. Prahalad, C., & Hamel, G. (1990). The core competence of the corporation. Harvard Business Review, 68(3), 79-91. Probst, G. J. B., Deussen, A., Eppler, M. J., & Raub, S. P. (2000). Kompetenz-Management: Wie Individuen und Organisationen Kompetenz entwickeln. Wiesbaden: Gabler. Rost, J. (1996). Lehrbuch Testtheorie, Testkonstruktion. Bern: Hans Huber. Sarges, W. (2001). Competencies statt Anforderungen: nur alter Wein in neuen Schluchen? In H. Riekhof (Ed.), Strategien der Personalentwicklung. Wiesbaden: Gabler. Schippman, J. S., Ash, R. A., Battista, M., Carr, L., Eyde, L. D., Hesketh, B., Kehoe, J., Pearlman, K., Prien, E. P., & Sanchez, J. I. (2000). The practice of competency modeling. Personnel Psychology, 53, 703-740. Schmitt, N., & Chan, D. (1998). Personnel selection. Thousand Oaks: Sage. Schuler, H., & Funke, U. (1998). Diagnose berulicher Eignungen und Leistungen. In H. Schuler (Ed.), Organisationspsychologie (pp. 235-283). Bern: Hans Huber. Schweitzer, E. (2001). Expertenbefragung auf Basis der Kompetenz-Performanz-Theorie: Entwicklung einer Methode am Beispiel der Rechtschreibung [Questioning experts on the basis of the competence-performance theory: Development of a method for orthography]. Unpublished thesis. Graz: Graz University.
Sonnentag, S., & Schmidt-Brae, U. (1998). Expertise at work: Research perspectives and practical interventions for ensuring excellent performance at the workplace. European Journal of Work and Organisational Psychology, 7(4), 449-454. Spencer, S. M., & Spencer, L. M. (1993). Competence at work: Models for superior performance. New York: John Wiley & Sons. Staudt, E., & Kriegesmann, B. (2002). Weiterbildung: Ein Mythos zerbricht (nicht so leicht!). In E. Staudt, N. Kailer, M. Kottmann, B. Kriegesmann, A. J. Meier, C. Muschik, H. Stephan, & A. Ziegler (Eds.), Kompetenzentwicklung und Innovation. Mnster: Waxmann. Stewart, T. A. (1997). Intellectual capital: The new wealth of organizations. New York: Doubleday. Sveiby, K. E. (1997). The new organizational wealth. San Francisco: Berrett-Koehler. Van den Berg, P. T. (1998). Competencies for work domains in business computer science. European Journal of Work and Organisational Psychology, 7(4), 517-531. Wilson, M. A. (1997). The validity of task coverage ratings by incumbents and supervisors: Bad news. Journal of Business and Psychology, 12(1), 85-96. Wls, K., Kirchpal, S., & Ley, T. (2003). Skills managementAn all-purpose tool? In K. Tochtermann & H. Maurer (Eds.), Proceedings of I-Know 03, 3rd International Conference on Knowledge Management (pp. 138-143). Graz: Know-Center. Yimam-Seid, D., & Kobsa, A. (2003). Expert-inding systems for organizations: Problem and domain analysis and the DEMOIR approach. Journal of Organizational Computing and Electronic Commerce, 13(1), 1-24.
Endnote
1
This is currently being undertaken within the APOSDLE project (Advanced Process Oriented Self-directed Learning Environment; see Lindstaedt, Ley, & Mayer, 2005). APOSDLE is partially funded under the FP6 of the European Commission within the IST Workprogram.
120 Lytras, Mantziou, & Pontikaki
Chapter V
Competencies Management:
Applying Technologies for Strategic Management
Miltiadis Lytras, University of Patras, Greece Maria Mantziou, Open University of Greece, Greece Maria Pontikaki, Intralot S.A., Greece
Abstract
Competencies in the management literature are analyzed from different perspectives. A converging point of the various approaches is the fact that competencies provide a critical context for a uniied treatment of business processes and personal development. In other words, the achievement of critical business objectives is directly related to knowledge and learning management.The objective of this chapter is to analyze competencies management from a managerial perspective and to develop a set of requirements for new knowledge and learning management systems aiming to capitalize on the application of competencies management in to daily business life. The relevant discussion is initiated with the provision of several real-world scenarios that summarize the key business challenges of competencies development, management, and reuse in the education and health domains. The resulting list of requirements is used for a critical overview of limitations of current technology-enhanced learning approaches, especially in the context of real businesses. The concluding section
Competencies Management: Applying Technologies for Strategic Management 121
of this chapter discusses the new insights that competencies management through SW and TEL approaches bring to well-known business problems and reveals several streams of emerging research on the topic.
Competencies: Demystifying the Concept through Real World Scenarios

The deployment of leading-edge approaches to knowledge and learning management systems is not an easy task, not only because of the characteristics of the humans but also due to the extensive analysis of requirements that is needed. In most cases, this analysis results in a matching of different perceptions or expectations of the different entities that interrelate and collaborate within a business system. The same statement stands for the case of competencies management. We decided at this introductory section instead of repeating a general overview of deinitions and models of competencies, to present the down-to-earth context. In other words, we selected some real-world scenarios aiming to depict how competencies initiate critical business decisions related both to managerial and technological matters. Scenario 1: Competencies management for education Maria has been a teacher for 10 years. With a bachelor in pedagogy, and a masters in management of education, she has undertaken a new role in her region. She is responsible for a new program aiming to contribute to the development of teachers in a rural area of Greece. Development in the sense of human resources management. She has developed a list of priorities: To develop an innovative training program aiming to deliver up-to-date knowledge for the 200 teachers who work in the area of her responsibility for the discipline as well as the new technologies. She never liked this term (new technologies). To promote a social network with the support of technology aiming to help teachers in sharing ideas and expertise To analyze the characteristics of the new proile of teachers in the knowledge society. This is a key objective, since Maria realizes that technology can play a substantial role in the schools, not only in the direction of constructive collaborative learning but also toward more effective management. To ind a way to link the return of investment on training in lifelong teacher seminars. To be able, from a managerial perspective, to develop a career planning model for the managers of education.
In this context, Maria undertook a survey on information and communication technologies (ICTs) that could help her in the promotion of her objectives. Due to the limited budget available for investments on infrastructures, she tried to get some guidance from well-known institutions and the net. Riina Vuorikari has a key role in the European Schoolnet (EUN), http://www.eun.org. Maria sent mail to Riina and got a kind reply. Hello Maria, Here are some links for some of the latest projects where EUN with partners have looked into the use of ICT in schools. We are just about to inish the P2P project about peer-reviewing good ICT practices both on the policy and practice level. Below you can ind a link to the project site and some subareas where some school portraits are seen. Other than that, on our sites about e-twinning.net or springday2006.eun.org you should be able to ind some peeks into schools. Hope this helps. When you are done with your report we would be very interested in reading it and also disseminating it on our INSIGHT website. Regards, Riina
http://p2p.eun.org http://insight.eun.org/ww/en/pub/insight/school_innovation/best_practice/14school_visit_reports.htm http://insight.eun.org/ww/en/pub/insight/school_innovation/best_practice/p2pgallery.htm http://insight.eun.org/ww/en/pub/insight/school_innovation/best_practice/ernist_school_portraits.cfm
David Jonassen is a distinguished professor and director of learning technologies and educational psychology at the University of Missouri (www.coe.missouri.edu/~jonassen/). His work in constructivist learning environments is really outstanding. Maria had read many of his articles and tried it. They say that on the Internet, you are just a click away. She sent an e-mail and got an immediately reply from the professor. You are by far the BEST in this area so I would be grateful if you could provide me any links to reference articles, few that you consider as the state of the art (I know that most of them are yours ;). Here are some of mine. Jonassen, D.H. (2006). Modeling with technology: Mindtools for conceptual change. Columbus, OH: Merrill/Prentice-Hall.
Jonassen, D.H., Howland, J., Moore, J., & Marra, R.M. (2003) Learning to solve problems with technology: A constructivist perspective, 2nd. Ed. Columbus, OH: Merrill/PrenticeHall. I attach a paper that describes the rationale for the 2006 book. I would also examine the work of Eliot Soloway at the University of Michigan, John Bransford at the University of Washington, and Sasha Barab at Indiana University, especially Quest Atlantis (http://atlantis.crlt.indiana.edu/). And please if you have any idea on any research tool (e.g. questionnaire or interview agenda for such research context please let us know). [sic] I would have to know more about sampling and goals. All the best, Dave This experiment was really fascinating. Maria realized the power of community. She always believed that, especially in the educational context, the capacity of the community could promote and support many interesting initiatives. After some weeks it was obvious that her management vision directly correlated to two critical milestones. She had to work out the speciication of the desired proile in terms of knowledge that would permit her people (teachers in primary schools) to exploit the merits of new technologies. The problem was that she had found in many books a lot of wishful thinking but very few practical guidelines. She started thinking in terms of competencies and skills, and after a couple days she developed a nice list of must haves. On the other hand, she realized that the supporting ICT environment of her vision had to be linked to a community portal that could be accessed easily through a user-friendly environment. In Greece there where several portals, for example, http://www.e-paideia.net and http://www.pekp.gr/, but Maria felt that something better could be done in terms of functionalities, services, and community building. She had frequent talks with her colleagues and friends on this fascinating project. She was really motivated to deliver value through this proposition, and she liked to make her colleagues involved from the beginning. Many of their comments were to the point. Elena, a 33-year-old teacher, was disappointed: I feel that we forget the critical role of the teacher when we are dealing with new technologies. ICTs are not panacea. I feel that the problem is not on the deployment but mostly on the design of processes that we want to support. E.g. Think [sic] on the e-learning system that we currently deploy for our continuing education programs. It is really not a learning system. It is just a repository of iles. Giannis, 40-year-old teacher criticizing the no-sense of monolithic approaches to learning: I would be happy to use any technology-enhanced system that would be designed to help me become better in my core discipline as well as to the use [sic] of new technologies. But
most of the times these approaches do not consider my current level. They are treating every individual as the same. This is a critical management challenge. To be able to exploit from our people their skills and competencies and also to help them become better. Liana, 23, had the enthusiasm of the new generation of teachers. I use everyday 10s of Blogs, Wikis and other reference points for inding guidance ang [sic] get info for the new trend in my job. I like the most the OLDaily, by Stephen Downes, http://www.downes. ca/news/OLDaily.htm.. [sic] Today I read this amazing posting: [link:11 Hits] We are the music-makers, And we are the dreamers of dreams, Wandering by lone sea-breakers, And sitting by desolate streams. World-losers and world-forsakers, Upon whom the pale moon gleams; Yet we are the movers and shakers, Of the world forever, it seems. Yes, I am oficially back from my hiatus; thanks to those of you who have posted welcomes. Rob Wall kindly says Given the hysteria around Web based e-mail accounts, MySpace and other threats to the educational orthodoxy, the timing couldnt be better. And I do want to make a difference regarding these and other issues. What brings me back is a desire to make things better, to contribute my own unique voice to what I see as a renaissance. Like this: What edubloggers must do is to continue to engage in critical dialogue, relect, and communicate with those around us. We must be the ones who stand up and take responsibility for the struggle (If not us, then Who?). We must relect and act together in a way that offers a new story, a new vision of education can be. Start by looking in the mirror: Meet the new boss; youre not the same as the old boss... Perhapsbut we do not need vision and will, we do not need great leaders. There will be no revolution, no renaissance, until we change ourselves, until we ourselves become the embodiment of the caring and compassionate society we want to create. How hard that is! I return from my time away more aware than ever of how fallible, how ordinary, how human I am. Oh my yes, I have my apologies to give and my amends to make. Still, no matter how hard it is, we need to believe in ourselves, to believe we can make a difference, to believe we matter, to believe we can live freely. This, above all, must be our legacy. Question: Which competencies and under which titles we can describe the proile of the Teacher in the knowledge society? Scenario 2: Competencies management for health Maria has been an information technology (IT) manager for 7 years. With a bachelor in informatics, and two masters in management of information systems and management of
healthcare organizations, she is now chairing a committee in the Ministry of Health investigating propositions for a better performance in public hospitals. In the committee, there are people from many different areas: doctors, managers of hospitals, representatives of professional bodies (e.g., pharmaceutical companies, associations of doctors, ministry policy makers) as well as technostructure. Since the overall context and the relevant problem is really huge, Maria suggested to the ministry the formation of seven working groups under the following titles: WG1: IT-enabled services for citizens WG2: Healthcare management policies WG3: Human resources management WG4: Management of patient records WG5: Integration of information systems/interoperability WG6: Education of healthcare people WG7: Social awareness/communication
After many meeting and talks, Maria was enthusiastic about the propositions of the leaders of WGs. She was surprised that all of the WGs had in the top priorities of their proposed actions the need to exploit the human capital of health, in terms of managers, doctors, other staff, etc. It was obvious that LEARNING toward the development of new skills and competencies would be a critical bet for the success of the government program that she was leading. After this fact, she asked for guidance from technologists of her committee. She was more interested to see some technology-enhanced learning approaches for the support of health people education. In a meeting and small exhibition that was arranged, many IT providers presented the services enabled by their platforms: virtual symposiums, collaborative spaces, context modeling tools, knowledge portal integrated solutions, worklow systems, e-learning suites, mobile learning appliances and systems, semantic Web wikis and blogs. It was obvious that a new emerging era of technologies was here. Nobody can ignore this stream of tools, she said, so this was the bet for her. The challenge was to ind the way from a managerial perspective to integrate these fascinating systems into daily working life of healthcare organizations and health governing bodies. Her studies have resulted to a nice mix of managerial and decision-making behavior. Maria concluded that: Technologies must be deployed for well-deined strategies that the whole system can support and People think, design, implement, control, and revise strategies.
Competencies in Literature
The literature on competencies is really huge. We tried, in Figure 1, to adjust the proposition of Boon and Van der Klink (2001) with several other considerations. Competencies in the management literature are perceived mostly as characteristics of individuals or characteristics of organizations. These complementary views are critical for the strategic approach to competencies management. Strategies like skills and competencies building, knowledge exploitation, personal improvement, career planning, life-long learning, organizational learning, and knowledge management are only a few approaches. Another critical stream of the literature relates to the infrastructure that incorporates social and technical networking, as well as the knowledge, context, and process triptych. In simple words, the down-to-earth objective, or the wishful thinking, is to develop an infrastructure that will exploit a socio-technical network, in which knowledge lows will be initiated by speciic conditions of the working context for the support of critical business process. And we want this ine-tuned system to be able to transform strategies to tangible results. It really sounds like a quite ambitious objective. Before trying to see what technology today does for our ambitious objective, it is interesting to look for some guidance in various competencies frameworks and competencies models. Our point of view is to reveal some implications for management and technology management. It is obvious from the table that the various implications depicted lead to big questions. And perhaps the answers cannot be monolithic. They require the convergence of many
Figure 1.
Management of Competencies Competencies
Views of Competencies
Characteristicsof of Characteristics Individuals Individuals

Skills Building Knowledge Exploitation Improvement Career Planning
Characteristicsof of Characteristics Organizations Organizations
AMode Modeof ofdiscourse discoursebetween between A education& &the thelabor labormarket market education
Life Long Learning Training
Strategies
Knowledge Management Organizational Learning
Competencies Competencies Management Management

Specialization from device-centered view
Processes Processes Knowledge Knowledge Infastructure WorkContexts Contexts Work
NetworkInfrastructure Infrastructure(Social (Social& &Technical) Technical) Network
Table 1.
Model Boon and Van der Klink (2001) Types Characteristics of individuals Characteristics of organizations A mode of discourse between education and the labor market Implications We must specify these characteristics in a formal way. We must see how the development of these characteristics can be a technology support matter or if they require integrated strategies. We must model competencies and develop competency descriptions, and also we must specify level of competence for each distinct type. We must ind ways for judging on the threshold levels and also to develop performance measures. We must ind ways to go beyond the verbalism of distinction of categories and ind ways to support all the various types with systems. We must analyze the working context as well as the business requirements and link learning with career planning. We must understand competencies as a responding mechanism to changing business environment. We must develop systems that relate with the real business needs. Several times there is a huge gap between rigorous academic propositions and tough daily business life. This seems a problem, but in our understanding is a great challenge.
Sparrow and Hiltrop (1994)
Behavioral competencies Managerial competencies Core competencies
Boyatzis (1982)
Competency frameworks threshold and high performance levels
Devisch (1998)
Core competencies Functional competencies Speciic competencies
Kuijpers (2000)
General working competencies Learning competencies Career-related competencies Task-speciic Firm-speciic Industry-speciic
Nordhaug (1998)
Human resources management department practical approaches
First approach: Managerial Generic Technical functional Managerial competencies Job-speciic Competencies General management Corporate speciic
Second approach
Third approach Behavioral anchors
disciplines and also a new approach to the perception of the role and the mode of working of modern businesses. We will try in the next section to comment on the must haves for the exploitation of competencies considerations in real world conditions.
New Insights for the Competencies Management: Must haves from a Business Perspective
It is really exciting to be critical even of your own beliefs, especially when these relate to the application of your discipline to a domain. The most interesting criticism is happening whenever people from the industry confront academic propositions. We decided to organize this session around a futuristic killer-app of competencies management. It sounds provocative, but it is really interesting to think on services, or characteristics of such application. Killer app characteristics: Ontological agreement on competencies types: It is obvious that a system for competencies management must deploy a well-deined and agreed-upon set of competencies. This means that a thorough work at the design stage must conclude the list of competencies. In other chapters of the book, there are several discussions on technology support approaches for the deinition of competencies as well as semantic Web-enabled formalizations. Interpretation: Let us take the case of management of education. If we use this killerapp in the context of life long training of educational personnel (teacher and managerial staff), then we admit that the system will be populated with a list of 100s of competencies with threshold levels and relations that could be used for many services, for example, career planning. In a way, the proile of a teacher could be described in terms of subsets of the whole list of competencies. And of course an instance of this ontology of competencies could be used to describe each individual. So an excellent question is: Is it enough, the fact that we have this predeined (and possibly extended) list of competency types? The answer is deinitely no. The most important issue is to exploit these potential information containers for the population of context-speciic knowledge. And this guides us to the second characteristic of the killer-app. Competency-based annotation/categorization of knowledge: Within a business, the knowledge ecosystem is consisted by various pieces of knowledge that, according to the literature of KM, fall into several blur categories. It would be really interesting to be able to annotate this knowledge wealth with a competency-based point of view. Interpretation: Let us consider the case of management of education. In schools, a critical process is the administration, which relates which knowledge lows between the ministry of education, local educational oficers, and school directors. Competency-based annotation/categorization of knowledge in this context means that the killer app could be used to apply metadata and semantics to various knowledge types
(e.g., documents, reports) in order to increase the system performance. A practical example is the following: Consider the case where new books are published aiming to support a new educational paradigm (e.g., constructivism, collaborative learning, etc.). A key challenge is to support teachers with guidance, examples, and educational material that will help them to promote and transform the new vision into their daily life. If we had the killer-app, then we could do multiple-fold competency-based annotations. For example, we could annotate the guidelines for teachers with semantics that would explain what a positive effect would have on teacher competencies. Or from the other point of view we could annotate new educational material (e.g., new books contents) with semantics that would say the impact of content on competencies of pupils. This guides us to the next characteristic of the killer app. Competency-based (e-)learning: For many years, technology-enhanced learning approaches, as well as e-learning practices, have been criticized for their learning outcome, or the return on investment. With a competency-based approach, the learning low as well as the instructional design can be organized around the development of speciic competencies. Learners can be associated with speciic competencies proiles and a systematic knowledge gap analysis can be implemented as a strategic plan for their learning. The same strategy can be applied to an organization as a whole. The blur vision of organizational learning can ind new insights from the competencies perspective. The core competencies of organizations can be treated as critical milestones where contributions of individual and team competencies play a key role Competency-based social networking: Business networking is an old story. The evolution of collaboration and communication technologies has created unforeseen opportunities for teamwork and (inter)organizational synergies. Competency-based social networking implies that detailed competency proiles of people are used for exploring collaboration opportunities. Several advanced services could exploit this capacity. Interpretation:Let us consider the case of management of education. Within a rural area where 200 teachers are working, their detailed competency proiles can be used for exploring collaboration opportunities, for example, in the context of a European e-learning initiative. Or consider the case of an international effort to develop a who is who in education aiming to collect the info for thousands of teachers worldwide but with an objective to use these characteristics for the participation in a big research and development (R&D) program, which requires the involvement of teachers with speciic characteristics/competencies. Competency-based interoperability/allocation of people to business tasks: One of the biggest problems in business contexts is the contribution of the individual competencies to the organizational core competencies. It is not just a verbalism to claim that the cost of employees professional development is extraordinary these days, not only because training is a continuous process but mostly because the life cycle of knowledge and its utility has shortened dramatically. From another point of view, it is extremely interesting to integrate interoperability, or allocation of people to speciic tasks according to their level of proiciency or their level of competence.
This list is far from exhaustive. And for sure it requires a critical consideration of many open issues. The idea of the killer-app is not far away from the propositions of many innovative research and development projects. But it must be made clear that a system that will exploit competencies management has to integrate the strategic consideration of an organization. Management relates with four critical functions: Planning Organizing Leadership Control
This typical fourfold approach, even though it provides a simplistic context for the discussion of critical management issues, can be used for specifying the research agenda on the strategic convergence of new technologies and business strategy through competencies. The following table summarizes only few of the obvious matters. We must keep in mind that competencies management is not a panacea. We use to say that if our only tool is a hammer, then why consider everything nails? This is a trap in every knowledge- and technology-intensive matter. If you ask yourself, then you will admit that in our self-judging, its too tough to list our competencies and our knowledge gaps. Competencies management is about promoting a more intensive view of human capital within every organization. The competencies-based view of the irm is, in fact, a management challenge for the organizations of the 21st century. And, as always, pioneers get the risk but also enjoy and the fruitful juices of their efforts.
Table 2.
Management Function Planning Competencies Insights Career planning Personal development Recruiting knowledge Gap analysis Competencies Threshold levels Organizing Employees allocation Teams formation Worklow Detailed competencies description Business process and competencies matching Competency-based annotation Social networking Leadership Personal development Collaboration Support Exploitation of individual competencies Inspiring interorganizational synergies Control Performance measurement Competency-based evaluation Management by competencies Worklow control based on competencies Emergency plans according to knowledge gaps
Conclusion
In this chapter we didnt try to give answers. We just tried to show the emerging and changing character of our times. We are living in an interconnected world, where people participate in various networks and provide their expertise and knowledge in various contexts. And we also said that competencies management, like any other hot topic in the management agenda, requires a multi-fold analysis of enabling factors. We do believe that the next milestones of business strategy will relate with the promotion of performance-related with knowledge in social networks. Competencies have to play a key role. For all interested in inding more on business perspectives of competencies, we compiled an excellent list of references.
References
Abraham, S. E., Karns, L. A., Shaw, K., & Mena, M. A. (2001). Managerial competencies and the managerial performance appraisal process. Journal of Management Development, 20(10), 842-52. Ackoff, R. L. (1994). The democratic corporation. New York: Oxford University Press. Albanese, R. (1989). Competency-based management education. Journal of Management Development, 8(2), 66-79. Alderson, S. (1993). Reframing management competence: Focusing on the top management team. Personnel Review, 22(6), 53-62. Allen, C. (Ed.). (2003, February 26). Competencies (1.1) (measurable characteristics). HrXML recommendation. American Compensation Association. (1996, May). Raising the bar: Using competencies to enhance employee performance. Scottsdale, AZ: ACA. American Compensation Association Competencies Research Team (1996). The role of competencies in an integrated HR strategy. ACA Journal, 5-21. Anderson, T., & Whitelock, D. (2004). The educational semantic Web: Visioning and practicing the future of education [Special issue]. Journal of Interactive Media in Education (1). Antonacopoulou, E. P., & Fitzgerald, L. (1996). Reframing competency in management education. Human Resource Management Journal, 6(1), 27-48. Argyris, C. (1993). Education for leading: learning. Organisational Dynamics, 21(3), 517. Argyris, C., & Schon, D. A. (1978). Organisational learning. Reading, MA: AddisonWesley. Armstrong, A., & Foley, P. (2003). Foundations for a learning organization: Organization learning mechanisms. The Learning Organization, 10(2), 74-82.
Arthur, M. B., & Rousseau, D. M. (1996). Introduction: The boundaryless career as a new employment principle. In M. B. Arthur & D. M. Rousseau (Eds.), The boundaryless career: A new employment principle for a new organisational era. (pp. 3-20). New York: Oxford University Press. Ashworth, P. D., & Saxton, J. (1990). On competence. Journal of Further and Higher Education, 14(2), 3-25. Athey, T. R., & Orth, M. S. (1999). Emerging competency methods for the future. Human Resource Management, 38(3), 215-28. Baader, F., Calvanese, D., McGuinness, D., Nardi, D., & Patel-Schneider, P. (Eds.). (2003). The description logic handbook. Theory, implementation and applications. Cambridge University Press. Baker, T., & Aldrich, H.E. (1996). Prometheus stretches: Building identity and cumulative knowledge in multi-employer careers. In M. B. Arthur & D. M. Rousseau (Eds.), The boundaryless career: A new employment principle for a new organisational era (pp. 132-149). New York: Oxford University Press. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120. Barney, J. (2001). Resource-based theories of competitive advantage: A ten-year retrospective on the resource-based view. Journal of Management, 27(6), 643-8. Barrett, G. V., & Depinet, R. L. (1991). A reconsideration of testing for competence rather than intelligence. American Psychologist, 46(10), 1012-24. Bartlett, C. A., & Ghoshal, S. (1989). Managing across borders: The transnational solution. Cambridge, MA: Harvard Business School Press. Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine Books. Becker, B. E., & Huselid, M. A. (1998). High performance work systems and irm performance: a synthesis of research and managerial applications. In G. R. Ferris (Ed.), Research in personnel and human resources management (pp. 53-101). Greenwich, CT: JAI Press. Beckett, R., & Murray, P. (2000). Learning by auditing: A knowledge creating approach. The TQM Magazine, 12(2), 125-36. Beleherman, G., McMullen, K., Leckie, N., & Caron, C. (1994). The Canadian workplace in transition. Ontario: Inc. Press. Bereiter, C. (1985). Toward a solution of the learning paradox. Review of Educational Research, 55, 201-26. Bergenhenegouwen, G. J. (1990). The management and effectiveness of corporate training programmes. British Journal of Educational Technology, 21(3), 196-202. Bergenhenegouwen, G. J., ten Hom, H. F. K., & Moorjman, E. A. M. (1996). Competence developmentA challenge for HRM professionals: Core competences of organizations as guidelines for the development of employees. Journal of European Industrial Training, 20(9), 29-35. Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The semantic Web. Scientiic American, 284(5), 34-43.
Bessant, J., & Caffyn, S. (1996). High involvement innovation through continuous improvement. International Journal of Technology Management, 14(1), 7-28. Birdir, K., & Pearson, T. E. (2000). Research chefs competencies: A Delphi approach. International Journal of Contemporary Management, 12(3), 12-22. Boam, R., & Sparrow, P. (1992). Designing and achieving competency: A competency-based approach to developing people and organizations. London: McGraw-Hill. Boon, J., & Van der Klink, M. (2001, January). Scanning the concept of competencies: how major vagueness can be highly functional. In Proceedings of 2nd Conference on HRD Research and Practice across Europe, University of Twente, Enschede. Boyatzis, R.E. (1982). The competent manager: A guide for effective management. New York: Wiley. Boyatzis, R. E., & Kolb, D. N. (1995). From learning styles to learning skills: The executive skills proile. Journal of Managerial Psychology, 10(5), 24-7. Bratton, D. A. (1998). Develop a framework of corporate competencies. Credit Union Magazine, 64(10), 17-18. Brown, R. B. (1994). Reframing the competency debate: Management knowledge and metacompetence in graduate education. Management Learning, 25(2), 289-99. Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32-42. Brusilovsky, P. (2000). Adaptive hypermedia: From intelligent tutoring systems to Webbased education (invited talk). In G. Gauthier, C. Frasson, & K. VanLehn (Eds.), Intelligent tutoring systems. Lecture notes in computer science (Vol. 1839) (pp. 1-7). Berlin: Springer Verlag. Burgoyne, J. (1989). Creating the management portfolio: Building on competency approaches to management development. Management Education & Development, 20(1), 56-61. Cappelli, P., & Crocker-Hefter, A. (1996). Distinctive human resources are a irms core competencies. Organisational Dynamics, 28(2), 7-21. Cappelli, P., & Singh, H. (1992). Integrating strategic human resources and strategic management. In D. Lewin, O. Mitchell, & P. Shewer (Eds.), Research frontiers in industrial relations & human resources (pp.165-192). Madison, WI: IRRA. Carroll, J. M. (2000). Five reasons for scenario-based design. Interacting with Computers, 13(1), 43-60. Carroll, S. J., & Gillen, D. J. (1987). Are the classical management functions useful in describing managerial work? Academy of Management Review, 12(1), 38-51. Chatman, J. A. (1989). Improving interactional organizational research: A model of personorganization it. Academy of Management Review, 14(2), 333-49. Cockerill, A. P., Hunt, J. W., & Schroder, H. M. (1995). Managerial competences: Fact or iction? Business Strategy Review, 6(3), 1-12. Cockerill, T. (1989). The kind of competence for rapid change. Personnel Management, 21(9), 52-6.
Cockerill, T., & Hunt, J. (1995). Managerial competencies: Fact or iction?. Business Strategy Review, 6(3), 26-34. Cofsky, K. M. (1993, November-December). Critical keys to competency-based pay. Compensation & Beneits Review, 46-52. Collin, A. (1989). Managers competence, rhetoric, reality and research. Personnel Review, 18(6), 20-5. Conway, C. (1994). Developing senior management competences at Ocean Group. Management Development Review, 7(1), 7-12. Cornelius, N. (1999). Human resource management: A managerial perspective. London: Thompson Business Press. Cornford, T., & Athanasou, J. (1995). Developing expertise through training. Industrial & Commercial Training, 27(2), 10-18. Corporate Leadership Council. (1997, January). Developing a strategic HR function: Aligning individual competencies. Washington: The Advisory Board Company. Corporate Leadership Council. (1998, April). Developing individual competencies for HR professionals (pp. 1-13). Washington: The Advisory Board Company. Cranield University of Limerick. (1999). Department of personnel and employment relations, University of Limerick, Munster. Crossan, M. M., Lane, H. W., & White, R. E. (1999). An organisational learning framework: From intuition to institution. Academy of Management Review, 24(3), 522-37. Currie, G., & Darby, R. (1995). Competence-based management development: Rhetoric and reality. Journal of European Industrial Training, 19(5), 11-26. Dale, M., & Iles, P. (1992). Assessing management skills: A guide to competencies and evaluation techniques. London: Kogan Page. DeFillippi, R. J., & Arthur, M. B. (1996). Boundaryless contexts and careers: A competencybased perspective. In M. B. Arthur & D. M. Rousseau (Eds.), The boundaryless career (pp. 116-131). New York: Oxford University Press. Derouen, C., & Kleiner, B. (1994). New developments in employee training. Work Study, 43(2), 13-6. Devisch, M. (1998). The Kioto people management model. Total Quality Management, 9(4-5), 62-5. Dibella, A. J., Nevis, E. C., & Gould, J. M. (1996). Understanding organisational learning capability. Journal of Management Studies, 33(3), 361-79. DPWS (1997). A perspective of the construction industry in NSW in 2005. Department of Public Works and Services, Sydney. DTI White Paper (1995). Competitiveness: Forging Ahead (CM2867). HMSO, London. Dunphy, D., Turner, D., & Crawford, M. (1997). Organisational learning as the creation of corporate competencies. Journal of Management Development, 16(4), 232-44. Engestrm. Y. (1987). Learning by expanding. An activity theoretical approach to developmental research. Oy, Helsinki: Orienta-Konsultit.
Engestrm. Y. (1999). Work, learning and development: Towards an activity-theoretical reconceptualization. Keynote speech given at Zu richer Symposium, ETH, 12 October. Engestrm, Y. (2000). Activity theory as a framework for analyzing and redesigning work. Ergonomics, 43, 960-74. Eraut, M. (1994). Developing professional knowledge and competence. London: Falmer Press. European Commission. (1997). Green paper: Partnership for a new organization of work. document drawn on the basis of COM(97) 128 inal, Brussels. Fay, D., & Frese, M. (2001). The concept of personal initiative: an overview of validity studies. Human Performance, 14(1), 97-124. Feldman, D. (1996). Managing careers in downsized irms. Human Resource Management, 35(2), 145-61. Fielding, N. G. (1988). Competence and culture in the police. Sociology, 22, 45-64. Fiol, C. M., & Lyles, M. A. (1985). Organisational learning. Academy of Management Review, 10(4), 803-13. Freedman, D. H. (1992). Is management still a science? Harvard Business Review, 70(6), 26-38. Frese, M., Fay, D., Hilburger, T., Leng, K., & Tag, A. (1997). The concept of personal initiative: Operationalization, reliability and validity in two German samples. Journal of Occupational and Organizational Psychology, 70, 139-61. Frese, M., Kring, W., Soose, A., & Zempel, J. (1996). Personal initiative at work: Differences between East and West Germany. Academy of Management Journal, 39, 37-63. Friedrich, P. (1992). Kompetensutveckling vid local teknikfrndring (Skills development and technical change activities at workplace level). Unpublished doctoral dissertation, Royal Institute of Technology, Stockholm. Garavan, T. N., Bamicle, B., & OSulleabhain, F. (1999). Management development: contemporary trends, issues and strategies. Journal of European Industrial Training, 23(4/5), 3-12. Garca, E., & Sicilia, M. A. (2003). User interface tactics in ontology-based information seeking . Psychology e-Journal 1(3), 243-256. Gorsline, K. (1996). A competency proile for human resources, no more shoemakers children. Human Resource Management, 35(1), 53-66. Grant, R. M. (1991). The resource based theory of competitive advantage: implications for strategy formulation. California Management Review, 33(3), 114-35. Greenhaus, J. H., & Callanan, G. A. (1994). Career management. London: Dryden Press. Gruber, T.R. (1993). A translation approach to portable ontologies. Knowledge Acquisition, 5(2), 199-220. Grugulis, I. (1997). The consequences of competence: A critical assessment of the management NVQ. Personnel Review, 26(6), 428-44.
Hacker, W. (1998). Allgemeines arbeitspsychologie (General work psychology). Bern: Huber. Hackman, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-Wesley. Hamel, G., & Prahalad, C. K. (1989). Strategic intent. Harvard Business Review, 67(3), 63. Hamel, G., & Prahalad, C. K. (1990). The core competence of the corporation. Harvard Business Review, 68(3), 79-91. Hamel, B., & Prahalad, C. K. (1993). Strategy as stretch and leverage. Harvard Business Review, March-April, 75-84. Havaleschka, F. (1999). Personality and leadership: A benchmark study of success and failure. Leadership & Organization Development Journal, 20(3), 114-32. Henderson, R. (1994). Managing innovation in the information age. Harvard Business Review, January-February, 100-05. Hirsch, W., & Jackson, C. (1996), Strategies for career development: promise, practice and pretence. Brighton: Institute for Employment Studies. Hodgetts, R. M., Luthans, F., & Slocum, J. W. (1999). Strategy and IIRM initiatives for the 00s environment: Redeining roles and boundaries, linking competencies and resources. Organisational Dynamics, 28(2), 7-21. Hoerr, J. (1989). The pay-off from teamwork. BusinessWeek, 56-62. Hofer, C.W., & Schendel, D. (1978). Strategy formulation: Analytical concepts. St. Paul, MN: West Publishing. Hoffmann, T. (1999). The meanings of competency. Journal of European Industrial Training, 23(6), 275-286. Hogg, P. (1994). European managerial competencies. European Business Review, 23(2), 21-6. Holms, L. (1995). HRM and the irresistible rise of the discourse of competence. Personnel Review, 24(4), 16-28. Holsapple, C. W., & Joshi, K. D. (2004). A formal knowledge management ontology: Conduct, activities, resources, and inluences. Journal of the American Society for Information Science and Technology, 55(7), 593-612. Hondeghem, A., & Vandermeulen, F. (2000). Competency management in the Flemish and Dutch civil service. International Journal of Public Sector Management, 13(4), 25-34. Horton, S. (2000). Introduction: the competency movement: its origins and impact on the public sector. International Journal of Public Sector Management, 13(4), 7-14. Huselid, M.A. (1995). The impact of human resource practices on turnover, productivity, and corporate inancial performance. Academy of Management Journal, 38(3), 635-72. Johnston, R., & Sampson, M. (1993). The acceptable face of competence. Management Education & Development, 24(3), 216-24. Jubb, R., & Robotham, D. (1997). Competences in management development: Challenging the myths. Journal of European Industrial Training, 21(4-5), 171-77.
Kakabadse, A. (1991). The wealth creators. London: Kogan Page. Kakabadse, A., & Anderson, S. (1993). Top teams and strategic change. Management Development Review, 11(2), 10-22. Kamoche, K. (1996). Strategic human resource management within a resource capability view of the irm. Journal of Management Studies, 33(2), 213-34. Kanungo, R. N., & Misra, S. (1992). Managerial resourcefulness: A reconceptualization of management skills. Human Relations, 45(12), 1311-32. Ketchen, D. J., Thomas, J. B., & McDaniel, R. R. (1996). Process, content and context: Synergistic effects on organisational performance. Journal of Management, 22(2), 231-57. Kim, D. H. (1993). The link between individual and organisational learning. Sloan Management Review, 35(1), 37-50. King, A. W., & Zeithaml, C. P. (2001). Competencies and irm performance: examining the causal ambiguity paradox. Strategic Management Journal, 22. Klemola, U. M., & Norros, L. (1997). Analysis of the clinical behaviour of anaesthetists: Recognition of uncertainty as a basis for practise. Medical Education, 31, 449-56. Klink, M. R., Van Der Boon, J., & Bos, E. (2000). The investigation of distinctive competencies within professional domains. In Proceedings of the First Conference of HRD Research & Practice across Europe, 105-14. Kolb, D. (1984), Experiential learning. Englewood Cliffs, NJ: Prentice-Hall. Kossek, E. E., Roberts, K., Fisher, S., & Demarr, B. (1998). Career self-management: A quasi-experimental assessment of the effects of a training intervention. Personnel Psychology, 51, 935-60. Krogh, G., & Roos, J. (1995). A perspective on knowledge, competence and strategy. Personnel Review, 24(3), 56-76. Kuijpers, M. (2000). Career development competencies. In Proceedings of the 2nd Conference of HRD Research & Practice across Europe (pp. 309-14). Laabs, Jennifer. (1998, May). Why HR cant win today. Workforce, 62-74. Lado, A. A., Boyd, N. G., & Wright, P. (1992). A competency-based model of sustainable competitive advantage: toward a conceptual integration. Journal of Management, 18(1), 779-791. Lado, A. A., & Wilson, M. C. (1994). Human resource systems and sustained competitive advantage: A competency-based perspective. Academy of Management Review, 19(4), 699-727. Langley, A. (2000). Emotional intelligence: A new evaluation for management development. Career Development International, 5(3), 25-36. Lantz, A., & Friedrich, P. (1999). A guide to the description of transferable skills. Eskilstuna, Sweden: Mlardalen University, Department of Social and Behavioural Sciences. Lantz, A., & Friedrich, P. (2003). Learning in the workplace - An instrument for competence assessment. The Learning Organisation, 10(3), 185-194.
Lave, J., & Wagner, E. (1991). Situated learning. Legitimate peripheral participation. Cambridge: Cambridge University Press. Lawler, E. (1993). From job-based to competency-based organizations. Journal of Organizational Behavior, 15(1), 3-15. Le Boterf, G. (2002). Dvelopper la comptence des professionnels. Paris: ditions dOrganisation. Lei, D., & Hitt, M. A. (1996). Dynamic core competencies through meta-learning and strategic context. Journal of Management, 22(4), 549-61. Leonard-Barton, D. (1992). The factory as a learning laboratory. Sloan Management Review, 34(1), 23-38. Leonard-Barton, D. (1995). Wellsprings of knowledge. Boston: Harvard Business School Press. Leonteev, A. N. (1978). Activity, consciousness, and personality. Englewood Cliffs, NJ: Prentice-Hall. Lepak, D. P., & Snell, S. A. (1999). The human resource architecture: Toward a theory of human capital allocation and development. Academy of Management Review, 24(1), 31-48. Lindgren, R., Stenmark, D., & Ljungberg, J. (2003). Rethinking competence systems for knowledge-based organizations. European Journal of Information Systems, 12(1), 18-29. Losey, M. R. (1999). Mastering the competencies of HR management. Human Resource Management, 38(2), 99-111. Lytras, M., Pouloudi, A., & Poulymenakou, A. (2002). Knowledge management convergence: Expanding learning frontiers. Journal of Knowledge Management 6(1), 40-51. Lytras, M., Tsilira, A., & Themistocleous, M. G. (2003). Towards the semantic e-learning: An ontological oriented discussion of the new research agenda in e-learning. In Proceedings of the Ninth Americas Conference on Information Systems, 2985-2997. Mabon, H. (1995). Human resource management in Sweden. Employee Relations, 17(7), 57-83. MacLachlan, R. (1995). Revitalising people: The key to long-term success. People Management, 1(3), 13-15. Management Charter Initiative. (1990). Management competencies: The standards project. London: MCI. Mansield, B. (1996). Towards a competent workforce. Aldershot, UK: Gower. Martin, G., & Staines, H. (1994). Managerial competence in small irms. Journal of Management Development, 13(7), 23-34. Matlay, H. (2001, January 26-27). HRD in the small business sector of the British economy: An evaluation of current training initiatives. In Proceedings of 2nd Conference on HRD Research & Practice across Europe, University of Twente, Enschede, The Netherlands.
Matthews, P. (1999). Workplace learning: Developing an holistic model. The Learning Organization, 6(1), 18-29. McClelland, D. C. (1973). Testing for competence, rather than intelligence. American Psychologist, 28, 1-14. McClelland, D. C., & Boyatzis, R.E. (1982). Leadership motive pattern and long-term success in management. Journal of Applied Psychology, 67(6), 737-43. Metz, E. J. (1998). Designing succession systems for new corporate realities. Human Resource Planning, 21(3), 31-7. Miles, M. B., & Huberman, A. M. (1984). Qualitative data analysis: A sourcebook of new methods. Beverly Hills, CA: Sage. Miller, D. (1989). Matching strategies and strategy making: Process, content, and performance. Human Relations, 42(3), 241-60. Mintzberg, H. (1990). The design school: Reconsidering the basic premises of strategic management. Strategic Management Journal, 11(3), 171-95. Mintzberg, H. (1994). The fall and rise of strategic planning. Harvard Business Review, 72, 107-14. Mirabile, R. (1997). Everything you need to know about competency modelling. Training & Development, 51(8), 73-77. Mosakowski, E. (1993). A resource-based perspective on the dynamic strategy-performance relationship: An empirical examination of the focus and differentiation strategies in entrepreneurial irms. Journal of Management, 19(4), 819-38. Murray, P. (2001). Developing sustainable learning strategies in the construction industry. Unpublished doctoral thesis. University of Western Sydney, Macarthur. Murray, P. (2002). Cycles of organisational learning: A conceptual approach. Management Decision, 40(3), 239-47. Murray, P., & Donegan, K. (2003). Empirical linkages between irm competencies and organizational learning. The Learning Organization, 10(1), 51-62. Mwanza, D. (2002). Towards an activity-oriented design method for HCI research and practice. Unpublished doctoral dissertation, The Open University, UK. National Academy of Public Administration. (1996, June). A competency model for human resources professionals. National Council for Vocational Qualiications (NCVQ) (1997). Criteria for national qualiications. London: Qualiications and Curriculum Authority. Nelson, R. A., & Winter, S. (1982). An evolutionary theory of economic change. Cambridge, MA: Belknap Press of Harvard University. New, G. E. (1996). Relections: A three-tier model of organisational competencies. Journal of Managerial Psychology, 11(8), 44-52. Newton, R., & Wilkenson, M. J. (1995). A portfolio approach to management development: The Ashworth model. Health Manpower Management, 21(3), 16-31.
Nilsson, M., Palmr, M., & Naeve, A. (2002, May 7-11). Semantic Web meta-data for elearning--Some architectural guidelines. In Proceedings of the 11th World Wide Web Conference, Hawaii. Nonaka, I. (1991). The knowledge creating company. Harvard Business Review, 69, 96104. Nordhaug, O. (1998). Competencies speciicities in organisations. International Studies of Management & Organisation, 28(1), 8-29. Nordhaug, O., & Gronhaug, K. (1994). Competencies as resources in irms. The International Journal of Human Resource Management, 5(1), 89-106. Norros, L. (1990). Development of operators expertise in implementing new technologies: Constructing a model within a FMS case study. Paper presented at the international symposium Work and Welfare, Karlstad, Sweden. Oliveara-Rees, F. (1994). Qualiication versus competence: A discussion on the meaning of words: a change in concepts of a political issue. Beroepsopleiding, 1, 74-9. OReilly, C. A., & Chatman, J. A. (1994). Working smarter and harder: A longitudinal study of managerial success. Administrative Science Quarterly, 39(4), 603-27. Orpen, C. (1997). Performance appraisal techniques, task types and effectiveness: A contingency approach. Journal of Applied Management Studies, 6(2), 139. rtenblad, A. (2001). On differences between organizational learning and learning organization. The Learning Organization, 8(3), 125-133. rtenblad, A. (2001). Organizational learning and learning organization. The Learning Organization, 8(3), 125-133. Overmeer, W. (1997). Business integration in a learning organisation: The role of management development. Journal of Management Development, 16(4), 245-61. Porter, M. (1980). Competitive strategy: Techniques for analysing industries and companies. New York: Free Press. Pottinger, P. (1987). Competency assessment at school and work. Social Policy, 35-40. Prager, H. (1999). Cooking up effective team building. Training & Development, 53(12), 14-20. Prahalad, C. K., & Hamel, G. (1990). The core competence of the corporation. Harvard Business Review, 68(3), 79-91. Quinn, J.B. (1980). Strategies for change: Logical incrementalism. Homewood, IL: Richard D. Irwin. Raelin, J. A. (2000). Work-based learning: The new frontier of management development. Upper-Saddle River, NJ: Prentice-Hall. Raelin, J. A., & Cooledge, A. S. (1995). From generic to organic competencies. Human Resource Planning, 18(3), 24-33. Rainer, F., & Kazem, C. (1995). Strategy formulation: A learning methodology. Benchmarking: An International Journal, 2(1), 38-55. Reagon, P. M. (1994). Transform organisations using competency development. Journal of Compensation and Beneits, 95(5), 25-34.
Reichel, A. (1996). Management development in Israel: Current and future challenges. Journal of Management Development, 15(5), 22-36. Reid, M. A., & Barrington, H. (1994). Training interventions: Managing employee development. London: IPD. Resnick, L. B. (1987). Learning in schools and out. Educational Researcher, 16(9), 13-20. Reynolds, R., & Ablett, A. (1998). Transforming the rhetoric of organizational learning to the reality of the learning organization. The Learning Organization, 5(1), 24-35. Richter, P., Hemmann, E., & Pohlandt. A. (1999). Objective task analysis and the prediction of mental work load: Results of an application of an action-oriented software tool (REBA). In M. Wiwethoff & F. R. H. Zijlstra (Eds.), New approaches to modern problems in work psychology (pp. 67-76). WORC report 99.10.001. Tilburg, The Netherlands: University Press. Rothwell, W., & Kazanas, H. (1992). Mastering the instructional design process. San Francisco: Jossey-Bass. Sandberg, J. (2000). Understanding human competence at work: An interpretative approach. Academy of Management Journal, 43(1), 9-17. Sathe, V. (1983). Implications of corporate culture: A managers guide to action. Organizational Dynamics, 12(3), 5-23. Schein, E. H. (1984). Coming to a new awareness of organizational culture. Sloan Management Review, 25(2), 3-16. Schein, E. H. (1992). Organizational culture and leadership (2nd ed.). San Francisco: Jossey-Bass. Schoonver, S. C. Human resource competencies for the year 2000: The wake-Up call. Schroder, H. M. (1989). Managerial competence: The key to excellence. Dubuque, IA: Kendall Hunt. Selznick, P. (1957). Leadership and administration. New York: Harper & Row. Senge, P. (1990a). The leaders new work: Building learning organisations. Sloan Management Review, 32(1), 7-23. Senge, P. (1990b). The ifth discipline: The art and practice of the learning organization. New York: Doubleday. Senior, B. (1997). Organisational change. London: Financial Times-Prentice-Hall. Shea, G. P., & Guzzo, R. A. (1987). Group effectiveness: What really matters? Sloan Management Review, 28(3), 25-31. Sicilia, M. A., Garca, E., & Alcalde, R. (2003). Fuzzy specializations and aggregation operator design in competence-based human resource selection. In Proceedings of the 8th World Federation on Soft Computing Conference. Snell, S. A., Youndt, M. A., & Wright, P. M. (1996). Establishing a framework for research in strategic human resource management: Merging resource theory and organisational learning. In G. Ferris (Ed.), Research in personnel and human resource management (pp. 61-90). Greenwich, CT: JAI Press.
Soliman, F., & Spooner, K. (2000). Strategies for implementing knowledge management: Role of human resources management. Journal of Knowledge Management, 4(4), 337-345. Spangenberg, H. H., Schroder, H. M., & Duvenage, A. (1999). A leadership competence utilization questionnaire for South African managers. South African Journal of Psychology, 29(3), 117-29. Sparrow, P. R., & Hiltrop, J. M. (1994). European human resource management in transition. Hemel Hempstead: Prentice-Hall. Spencer, L.M. (1983). Soft skills competencies. Edinburgh: Scottish Council for Research in Education. Spencer, L. M., McClelland, D. C., & Spencer, S. M. (1990). Competency assessment methods: History and state of the art. Boston: Hay/McBer Research Press. Spencer, L. M., & Spencer, S. (1993). Competence at work: Models for superior performance. New York: Wiley. Storck, J., & Hill, P. A. (2000). Knowledge diffusion through strategic communities. Sloan Management Review, 41(2), 63-74. Stuart, R., & Lindsay, P. (1997). Beyond the frame of management competencies: Towards a contextually embedded framework of managerial competence in organisations. Journal of European Industrial Training, 21(1), 26-33. Sullivan, S. E., Carden, W. A., & Martin, D. F. (1998). Careers in the next millennium: Directions for future research. Human Resource Management, 8(2), 165-85. Sun, P., & Scott, J. (2003). Exploring the divideOrganizational learning and learning organization. The Learning Organization 10(4), 202-215. Sunassee, N., & Sewry, D. (2002). A theoretical framework for knowledge management implementation. In Proceedings of the 2002 annual research conference of the South African Institute of Enablement Through Technology, 235 - 245. Sure, Y., Maedche, A., & Staab S. (2000). Leveraging corporate skill knowledge--From ProPer to OntoProPer. In D. Mahling & U. Reimer (Eds.), Third International Conference on Practical Aspects of Knowledge Management (pp. 30-31). Tampoe, M. (1994). Exploiting the core competences of your organization. Long Range Planning, 27(4), 66-77. Taylor, F. W. (1911). The principles of scientiic management. New York: HarperCollins and Norton. Teece, D. J. (1990). Firm capabilities, resources and the concept of strategy: Four paradigms of strategic management (CCC Working Paper No. 90-8). Thomson, R., & Mabey, C. (1994). Developing human resources. Oxford: ButterworthHeinemann. Tolley, G. (1987). Competency achievement and qualiications. Industrial & Commercial Training, 19(5), 6-8. Townley, B. (1994). Reframing human resource management: Power, ethics and the subject at work. London: Sage.
Training Standards Agency (2000). Deinition as in S. Horton, Introduction: The competency movement: its origins and impact on the public sector. International Journal of Public Sector Management, 13(4), 7-14. Trice, H. M., & Beyer, J. M. (1993). The cultures of work organisations. Englewood Cliffs, NJ: Prentice-Hall. Tyre, M. J., & Von Heppel, E. (1997). The situated nature of adaptive learning in organisations. Organisational Science, 1, 71-83. U.S. Merit Systems Protection Board. (1993). Federal personnel ofices: Time for change?. Ulich, E. (1991). Industrial psychology. Stuttgart, Germany: Poeschel. Ulich, E., & Weber, W. G. (1996). Dimensions, criteria and evaluation of work group autonomy. In M.A. West (Ed.), Handbook of work group psychology (pp. 247-82). Chichester, NY: Wiley. Ulrich, D. (1997). Human resource champions. Boston: Harvard Business School. Ulrich, D. (1998, January). The future calls for change. Workforce, 87-91. USDA Graduate School. (1998, October). Competency based training task force report. Vaill, P. B. (1997). Learning as a way of being. San Francisco: Jossey-Bass. Van der Wagen, L. (1994). Building quality service with competency-based human resource management. Oxford: Butterworth Heinemann. Vasconcelos, J., Kimble, C., & Rocha, A. (2003). Organisational memory information systems: An example of a group memory system for the management of group competencies. Journal of Universal Computer Science, 9(12), 1410-1427. Virkukunen, J., Engestrm, Y., Helle, M., Pihlaja, J., & Poikela, R. (1997). The change laboratory: A tool for transforming work. In T. Alasoini, M. Kyllnen, & A. Kasvio (Eds.), Workplace innovation: A way of promoting competitiveness, welfare and employment (pp. 22-33). National Workplace Development Programme, report 3, Helsinki, 14(1). Weber, G., & Brusilovsky, P. (2001). ELM-ART: An adaptive versatile system for Webbased instruction. International Journal of Artiicial Intelligence in Education, 12(4), 351-384. Webster, J. (2000). Manual of competencies. Unpublished, Dublin. Wenger, E. C., & Snyder, W. M. (2000). Communities of practice: The organizational frontier. Harvard Business Review, 78(1), 139-45. Wild, R. H., Griggs, K. A., & Downing, T. (2002). A framework for e-learning as a tool for knowledge management. Industrial Management & Data Systems, 102(7), 371380. Winterton, J., & Winterton, R. (1996). The business beneits of competence-based management development (Research Series RS16). Department of Education & Employment, UK. Winterton, J., & Winterton, R. (1999). Developing managerial competence. London: Routledge.
Woodall, J., & Winstanley, D. (1998). Management development: Strategy and practice. Oxford: Blackwell. Wright, P. M., Dunford, B. B., & Snell, S. A. (2001). Human resources and the resource based view of the irm. Journal of Management, 27(6), 701-21.
Section II: Competencies from the Viewpoint of E-Learning and Organizational Memory
146 Abel
ChapterVI
An Organizational Memory Tool for E-Learning

Marie-Hlne Abel, Laboratory Heudiasyc UMR CNRS 6599, University of Compigne, France
Abstract
Learning can be considered an outcome associated with acquiring new competencies (Sicilia, 2005) and adding new knowledge. A competence is a way to put into practice some knowledge in a speciic context. The process of competency acquisition starts from a need in this speciic context. It may induce the search and the selection of relevant resources. Numerous resources may be used during e-learning, their access is a real problem. Different approaches may be adopted to exploit them. This chapter describes the tool E-MEMORAe, which supports an organizational goal-driven approach based on the concept of learning organizational memory. In such a memory, ontologies are used to deine knowledge that indexes resources; the capitalization and the organization of knowledge, information, and resources relating to a speciic context can be realized. End-users have a direct access to the memory. The organizational environment E-MEMORAe was evaluated in the context of two courses taught at the university (algorithms, mathematics).
An Organizational Memory Tool for E-Learning 147
Introduction and Background

Learning can be considered an outcome associated with acquiring new competencies (Sicilia, 2005) and adding new knowledge. A competence is a way to put into practice some knowledge in a speciic context. From an educational point of view, knowledge is deined as all the notions and the principles that a person acquires through study, observation, or experience, which can be integrated into skills. However, studying an encyclopaedia is not suficient to get knowledge; Didactic work has to be made. The process of competency acquisition starts from a need in a given context. It may induce the search and the selection of relevant resources. Numerous learning resources may be used during e-learning. E-learning becomes part of a complex organizational conduct, in which lack of required competencies trigger the search for appropriate contents (Sicilia, 2005). Different approaches may be adopted to exploit such contents. They can be stored in learning objects repositories and then reused, combined and adapted in different contexts. They can also be selected and organized in learning memories that are directly accessed by learners. These approaches offer a goal-driven organizational learning. Over the last few years, many projects aiming at building bases of learning resources, in order to share and re-use them, have been launched. These projects often rely on a network of contributors that feed the base with collaboratively controlled resources. Conversely, each contributor can beneit from resources brought by other contributors. We can make a distinction between learning object repositories (LOR), which usually group many subject matters, and what we call thematic resource bases that contain resources related to only one domain. LOR usually group all subject matters. Their scope can be restricted to one or several universities or to a country; it can also be international. If expected scope is wide, LOR are based on a network of contributors or on a consortium of institutions. The restriction in resources related to a particular domain brings more homogeneity; resources and associated knowledge can be managed more precisely. Thus, relying on knowledge engineering techniques, Paquette (Paquette, 2001) designed knowledge and resources base on tele-learning. As in the case of repositories, the idea is also to share and re-use resources. These resources are not ready to be used by learners; instructional design work is usually needed beforehand. On the contrary, within the MEMORAe project (in French, Mmoire organisationnelle applique au e-learning) supported by the pole Systms et Technologies de lEducation et de la Formation (STEF) of the Picardy area, France, our goal is to let learners directly access the resources of a course memory. Following a knowledge engineering approach, we organise the resources in a learning organizational memory based on ontologies (Abel, Barry, Benayache, Chaput, Lenne, & Moulin, 2004). In fact, it is a course memory, where a course is seen as an organization. This memory is different from a classical organizational memory because its goal is to provide pedagogycal users with content. This content is the result of two pieces of work: (1) the capitalization of knowledge, information, and learning resources relating to the learning context (a course unit); (2) a pedagogical work concerning the choice and the organization of this capitalization.
148 Abel
In order to give learners direct access to the memory, part of the instructional design work has to be made earlier. The advantage is that the memory is ready to be used by learners, provided that pedagogical and didactical choices made earlier are acceptable. This can therefore lead to a loss of lexibility, but we make the assumption that these choices can at least be shared by a teacher community that could act as a community of practice (Wenger, 1998). We realized two pilot applications to evaluate our propositions: the irst one concerns NF01, a course on algorithms and programming at the University of Technology of Compigne, and the second one concerns B31.1, a course on applied mathematics at the University of Picardy (France). In the following, we irst specify links between competencies and knowledge. Then we specify the role of knowledge in organizations and the parallel between knowledge management and e-learning. Afterward, we introduce the project MEMORAe, founded on the concept of learning organizational memory based on ontologies. We show how we modeled the memory. Finally, we present the organizational environment built from this model: E-MEMORAe (http://www.hds.utc.fr/~ememorae/Site-MEMORAe/). All examples come from the B31.1 application.
Competencies and Knowledge

We can ind different deinitions of the concept of competency. Meanwhile, all the deinitions seem to agree on three fundamental characteristics (Harzallah & Vernadat, 2002): resources, context, and objective. A competency is made of resources structured into categories. We can consider three main categories of resources: knowledge, know-how, behaviours (Harzallah, Leclre, & Trichet 2002). Knowledge is something that we acquire and store intellectually. It concerns everything that can be learned in an education system. For example, this category is concerned with theoretical knowledge, procedural knowledge. Know-how is related to personal experience and working conditions. It is acquired by putting into practice knowledge in a speciic context. Behaviours are individual characters that lead someone to act or react in a particular way under particular circumstances. They often condition the way knowledge or know-how is put into practice. According to Baugh (1997), we can distinguish two types of competencies: Hard competencies identify the basic resources that require performing an activity. These resources are generally expressed in terms of knowledge, skills, and abilities. Soft competencies correspond to personal behaviours, personal traits and motives (Wooddruff 1991), for example: working with others, leadership, etc.
The competency context is related to the environment in which the competency is situated. It represents the conditions and the constraints in which competencies should be mobilized.
Competency is related to reaching a goal or to the accomplishment of one or more missions, or tasks. These goals, missions, or tasks constitute the objective of the competency. Finally, acquiring competencies needs to select resources, to manage their combination and to control the way of bringing them into play. In this chapter, we focus on hard competencies and particularly knowledge resource, context, and objective of the competency.
Knowledge and Organization

Knowledge in an organization is the collection of expertise, experience, and information that individuals and workgroups use during the execution of their task (Abecker, & Decker, 1999). A knowledge organisation is one in which the key asset is knowledge (Conklin 1996). An organizations knowledge is built upon experience of their human resources and the lessons they learn. The effective management of this knowledge is addressed through research in the ield of knowledge management: organizational memory. In order to share information in an organization, actors have to use common terminology, especially when they are geographically distant. It is one of the reasons why organizational memories are often based on ontologies.
Organizational Memory
According to Van Heijst, Schreiber, and Wielinga (1997), one of the attempts of knowledge management is to make sure that an organization is able to apply the right knowledge at the right place and at the right time to achieve its business goal. Another attempt concerns the development of a knowledge policy based on business strategy. It necessitates a knowledge infrastructure to implement it and to monitor the functioning of the knowledge infrastructure (is the knowledge policy still adequate? Does the knowledge infrastructure adequately support it?). Two necessary conditions for accessible knowledge are: an eficient way of inding information related to some information need, and an eficient way of investigating the context of information found in order to establish its relevance and reliability (Martin, 1993). A common approach to tackle the knowledge management problem in an organization consists of designing an organizational memory (OM). Such a memory can be seen as an explicit and persistent representation of knowledge and information in an organization, in order to facilitate their access and reuse by members of the organization for their tasks (Rabarijaona, Dieng, Corby, & Ouaddari, 2000, p. 56). The irst role of an OM is to support the growth, the transmission and the conservation of knowledge (Steels, 1993). To that end, organizational memory management is based on the following stages (Dieng, Corby, Giboin, & Ribire 1998):
150 Abel
Needs detection Construction and structuring Diffusion of the contents Exploitation of the contents Evaluation of the objectives Maintenance and evolution of the contents
Organizational memories should comprise the knowledge of an organization collected over time (Klemke, 2000). It includes a model to describe information resources and the context in which these sources are created. It also includes knowledge in the form of personal memories of users (their knowledge, resources, etc.).
Ontologies
The term ontology is borrowed from philosophy, where ontology is a systematic account of existence. For artiicial intelligence (AI) systems, ontology is an explicit formal speciication of how to represent objects, concepts and other entities that are assumed to exist in some area of interest and the relationships that hold among them. What exists is that which can be represented. According to Gruber (1993) An ontology is an explicit speciication of a conceptualization. Guarino gives precision on this deinition, considering that ontologies are necessarily a partial speciication of a conceptualization (Guarino & Giaretta, 1995). We can add with Gruber an ontology is a description (like a formal speciication of a program) of the concepts and relationships that can exist for an agent or a community of agents. Using ontologies in AI systems is important. Indeed, ontological analysis clariies the structure of knowledge. If we dont have the conceptualizations that underlie knowledge, then we do not have a vocabulary for representing knowledge. Another point is that they provide a means for sharing knowledge (Chandrasekaran, Josephson, & Benjamins, 1998). Finally, an ontology is a catalog of the types of things that are assumed to exist in a domain of interest D from the perspective of a person who uses a language L for the purpose of talking about D. The types in the ontology represent the predicates, word senses, or concept and relation types of the language L when used to discuss topics in the domain D (Sowa, 2003). According to (Van Heijst et al., 1997), we can distinguish different kinds of ontologies: Domain ontology: Represents speciic conceptualizations of a domain. These representations are reusable for several applications of this domain. Application ontology: Represents domain knowledge useful for a given application. This knowledge is speciic and not reusable. Generic ontology: Represents valid conceptualisations for different domains. Meta-ontology: Represents language primitives of knowledge representation.
Knowledge Management and E-Learning

Currently, several approaches concern the pedagogical content management for e-learning. They are often based on the use of large sets of learning objects. A set of learning objects can be a repository shared by a network of actors whose goal is to reuse and adapt learning material. It can also be a learning memory where resources are structured and organized in order to support distance learning. Learning object repositories (LOR) usually group all subject matters. Their scope can be restricted to one or several universities or to a country; they can also be international. If expected scope is wide, LOR are based on a network of contributors or on a consortium of institutions. In Europe, the Alliance of Remote Instructional and Distribution Networks for Europe (ARIADNE, 2004) focuses on the share and re-use of hypermedia learning documents. These resources are stored in a knowledge pool system and are indexed by metadata based on the LOM (learning object metadata) standard (LOM, 2002). The re-use of these resources can be made by: Creating new materials from pieces of material to which the author can add new elements, Making a new presentation of an existing course material obtained by a rearrangement of its semantic components.
This implies that each author involved in the knowledge pool experience allows, under citation restrictions, the use and modiication of the components he brings in it. Conversely, he can do the same thing with other components. ARIADNE has thus interesting selection and access features, but one can notice that an instructional design work remains necessary to re-use the resources. The restriction in resources related to a particular domain brings more homogeneity; resources and associated knowledge can be managed more precisely. Thus, relying on knowledge engineering techniques, Paquette (2001) designed knowledge and resources base on tele-learning. To this end they rely on task ontology, based on use cases, and on a domain ontology that allows them to better index the resources. As in the case of repositories, the idea is also to share and re-use resources. These resources are not ready to be used by learners; an instructional design work is usually needed beforehand. On the contrary, within the MEMORAe project, our goal is to let learners directly access the resources of a course memory. Following a knowledge engineering approach, we organise the resources in an organizational memory. In fact, it is a course memory, where a course is seen as an organization. It can rather be seen as a memory of concepts and resources that teachers or designers ind useful in the framework of a particular course. In order to give learners direct access to the memory, part of the instructional design work has to be made earlier. The advantage is that the memory is ready to be used by learners, provided that pedagogical and didactical choices made earlier are acceptable. This can therefore lead
152 Abel
to a loss of lexibility, but we make the assumption that these choices can at least be shared by a teacher community that could act as a community of practice (Wenger, 1998). In using such a course memory, the learner can access the appropriate learning activities. It can be considered a goal-driven organizational learning.
A Learning Organizational Memory: The Project MEMORAe

The project MEMORAe was started from two points: A course unit is based on knowledge and competencies it should provide, on actors (learners, instructors, trainers, course designers, administrators, etc.), and on resources of different types (deinitions, exercises with or without solution, case studies, etc.) and different forms (reports, books, Web sites, etc.). In this sense, a course is an organization. An effective organizational memory is a prerequisite for organizational learning (Balasubramanian, 1995). Indeed, both the demonstrability and usability of learning depend on the effectiveness of the organizations memory.
Within the MEMORAe project, we propose to manage the resources and knowledge of a course by the means of a learning organizational memory based on ontologies (Abel et al., 2004).
Learning Organizational Memory/Organizational Memory

A learning organizational memory is different from a classical organizational memory because its goal is to provide pedagogical users with content. This content is the result of two pieces of work: (1) the capitalization of knowledge, information, and resources relating to the training or course unit; (2) a pedagogical work concerning the choice and the organization of this capitalization. The pedagogical content is composed of the notions to learn, the links between these notions, and the resources they index. Notions are not only chosen because they are related to the course unit, they are also the result of a relection on the course itself. For example, with NF01, why and how to make a link between the loop and array notions? Resources have to be selected relying on pedagogical goals. The choice of their indexing terms is related to this goal, too. It is not an automatic indexing. The course manager is responsible for the relevance of the links. It is not because a document treats of a notion to acquire that it
will be necessary indexed by this notion. The choice is explicit; that is to say, the document must have been evaluated as suficiently adapted to the learning of this notion. These choices are part of the pedagogical scenario the course manager wants to implement. In a classical organizational memory, there is no pedagogical scenario because the objective of this kind of memory is not training. The learning organizational memory we propose aims at facilitating knowledge organization and management for a given course or training, and at clarifying competencies it permits acquiring. An organizational memory allows capitalizing not only on learning resources related to the contents of the course but also on information on actors themselves (speciicities, background, proile, etc.). Administrative management (registration, notes, etc.) of the course can also be realised. Beyond the pedagogical content, the memory capitalizes: any information related to the learning environment that can be useful to users: trainers, staffs, entry form, learners, etc., and users knowledge.
Pedagogical Content of a Learning Organizational Memory

The pedagogical content of a learning organizational memory is mainly composed of the notions to learn, the links relating them, and the resources indexed by them. The manager of a training memory is responsible for its content, that is to say, the choice of the notions to learn and the resources indexed by these notions. In this sense, there is no course design (as it can exist in a linear course) but more precisely pedagogical content selection. The only material created is the set of notions, links, and annotations that can accompany them (for instance, why is this concept dificult). Resources are reused as they are.
Notion to Learn
The design of an e-learning application implies focus on the learner, giving him/her the means to be active, to make him/her understand the resources that are at his/her disposal, and to teach him/her how to search and use them. In knowledge engineering, a notion is any entity of thinking; it is used to structure knowledge and the perception of the world. Notions to learn are used as indexes to access documents related to them. A notion to learn can refer to several resources (giving several means to acquire it), and a resource can be referred to by several notions (giving several means to retrieve it). Note that a digital document can be made of several parts that can be themselves indexed. It will, however, remain a document as a whole for which the author has no writing guidelines to follow.
154 Abel
Learning Resources
Learning resources are generally documents: course texts, course notes, slides, e-books, reports, books presentations, and so forth. They may also be anything useful for training (for example, software, links to Web site). Among the represented resources, some (digital resources) are stored in the memory and others are references or description to physical resources. Resources can get accessed according to different rights. They can be private. In this case, users only store them in the memory and do not want to give other users access to them. They can be annotations, work in progress, or downloaded and not yet analyzed documents. Resources can also be semi-public or public, that is to say, shared by part or all of the users. For example, an annotation of a reader giving his/her motivated impression on a resource can help memory users choose appropriate resources. Moreover, several annotations written by different authors or relying on different notions can be attached to a same resource. Resources can also have different status. They can be terminated and validated resources, or, on the contrary, working resources created by one or more users and therefore shared by them during the time of their realization.
Annotations
Our relection on annotations started from two observations: On one hand, when users of the memory access a notion to learn, there are faced with several resources related to this notion. The choice can be based, as it is presently, on several associated characteristics: author, resource type (book, Web site), but it also could be guided by other information such as comments or remarks on the resources. On the other hand, the role of an organizational memory is to capitalize knowledge. It is then useful to keep track of the reasons that led a course manager to choose a resource, a notion, or a link between two notions. We propose to take into account this information by using annotations. In the MEMORAe project we consider that an annotation: Is a resource, result of an annotation action. Is related to a target that can be a notion to learn (concept), a link between concepts, a resource, a part of a resource, a collection of resources. Has one or several authors and presents its/their comments on the target. These comments are created at a given date with a precise objective and are directed to a precise audience (that can be the author himself in case of a personal annotation). Is not part of the target itself. It is then necessary to make a link between the target and the annotation. Makes sense only in its context (target, author, goal, audience). Can be text, graphic, voice, or illustration.
Note that a target must have a representation in the memory in order to be annotated. As an annotation is a resource, it can be itself annotated. Following this conception, our notions to learn are not annotations; they are metadata. We will now see how we represent them using ontologies.
The MEMORAe Model

The MEMORAe model relies on the expected use of the memory: e-learning. We mainly tried to: Determine and present the notions to learn and resources describing these notions. Offer a natural and easy access to the memory contents.
For this purpose, we were interested, on the one hand, in ontologies to represent the notions to learn and their links (deinition of a common vocabulary) and on the other hand in topic maps (XTM, 2001) as representation formalism facilitating navigation and access to the learning resources. The ontology structure is also used to navigate among the concepts, as in a roadmap. The learner has to reach the learning resources that are appropriate for him.
Ontologies
For navigating through the memory, the end-users (learners, teachers, etc.) need a shared vocabulary and knowledge structured. That is why we decided to model the memory with ontologies. From the different ontology types deined by Van Heijst et al. (1997), generic
Figure 1. Elements of domain ontology

Resourceaccess Digital Audio Video Text Image Solid Image CD Paper Slide Audiotape Videotape
Document Picture Graph Diagram Exercise Message Annotation Lecture Index Slide show Book
Person Administrative Student Technicien Teacher Course NF01 B31.1
156 Abel
Figure 2. Specializations of the set notion

Set Axe 1: cardinality finite/infinite Infinite set Finite set Axe 1.1: cardinal value Empty set Singleton Pair Set operation Complement Union Intersection Cartesian product
Axe 2: countable/uncountable Axe 3: super/sub
Uncountable set Countable set Superset Subset Strict subset Full subset
ontologies, domain ontologies, application ontologies, and meta-ontologies, we only use the second and third categories. We have to consider two aspects for modeling the memory and building ontologies (Breuker & Muntjewerff, 1999). First, the domain of training has its own characteristics. Second, it must be linked to the application domain of a particular training program. The irst ontology (domain ontology) we specify describes the concepts of the training domain (cf. Figure 1). They can be users types (teacher, administrative), document types (book, slides for oral presentation, Web page, site, etc.), and media types (text, image, audio, and video). They can also be pedagogical characteristics (activity type), and they can refer to point of view (annotation). Learning resources are not organized following the way recommended by the learning object metadata standard in the educational category, because we do not agree to associate various activity types, like exercise or simulation, with data representation like a diagram, igure, or graph in the same set. A description of the LOM standard can be found in the document 1484.12.1, http://ltsc.ieee.org/wg12/index.html. The second ontology (application ontology) speciies the organization of theoretical notions that are studied during training session. In the example of B31.1 course, some notions like set or ininite set are explained. It is possible, but not mandatory, to consider ininite set and inite set as sub-concepts of the concept set and to deine the relation has cardinality between the concepts inite set and cardinal (in this case, they are the domain and range value of this relation). According to the Ontospec method (Kassel, 2005), concepts can be specialized according to semantic axes. For example, the concept set is specialized according to three axes: inite/ininite, countable/uncountable, subset/superset (cf. Figure 2). These ontologies are not independent; the second one is necessarily attached to the irst one. For example, to express that a document is an introduction to ininite set, we join
the two concepts introduction and ininite set that do not belong to the same ontology. Pedagogical relations like prerequisite or uses that occur between concepts of the application ontology are deined in the domain ontology. However, speciic roles can belong to the application ontology (for example, for the B31.1 application, has-cardinality).
Topic Maps
The modeling of a training memory, which is presented here, contains three elements: two ontological parts and the way to index documents on them. The modeling must at least allow three operations: The reunion of two ontologies: the generic one and the application one; The substitution of an application ontology by another one coming from another domain; The attachment of document indexing on the reunion of two ontologies.
The modeling must also favour the answering to queries on the memory like: What are the documents (books, presentations, Web pages) that talk about, introduce, and develop a notion that appears in the training? What are the notions associated (pre-requisite, studied in the same time) with a given notion?
The Choice of the Topics Maps Formalism

The choice of the formalism(s) for representing the memory is very decisive. It must go beyond the hybrid aspect of the modeling (ontology and indexing) and favour the interoperability between various tools that have to deal with the memory (edition, updater, consultation, navigation, etc.). Two paths can be followed: either to choose the languages better adapted to the speciic nature of each element of the modeling or to choose a unique language. The irst choice requires a system to easily integrate data coming from the two formalisms. The advantage of the second choice is to unify the description of data. It can be valid only if this formalism allows describing some features that it is not worded for. The second choice can be split into two parts: We can choose a formalism adapted to ontology features representation or a formalism adapted to document indexing. The main aspect of the memory and its main use inluence the choice of the formalism. Even if the indexing of the learning documents is the main aspect of the memory, it is necessary to use a formalism allowing representation of ontological elements. Several formalisms can be envisaged, but we recommend the topic maps (TM) formalism (IEC, 1999).
158 Abel
This formalism is useful to deine and manipulate the information attached to resources. That provides a logical organization a large quantity of resources, keeping them accessible and facilitating the navigation between them. Since 2001, it has been possible to write a TM using the norm XTM 1.0 (XTM 2001) that can be considered a particular extensible markup language (XML). The building of a TM is based on an organization of topics. Each resource is directly attached to one or more topics by an occurrence link. The Association concept allows deined roles between topics. Moreover, the TM standard allows reifying some associations in order to place them in a particular scope. Some resources can also be reiied as a topic when it is necessary to attach other resources or data on them. In this manner, we can add annotations to resources. Annotations are resources that express a point of view on other resources. Overall, we chose the TM formalism because it keeps a semantic level close enough to the model of our memory. With an ontology-oriented point of view, also developed in (Park, 2002), this formalism allows envisaging the important following characteristics: It is possible to consider some topics as generic concepts and other as concept instances. It is possible to consider associations, scopes, and occurrences as roles between concept topics. Associations have no limitation in their member number. The occurrence relation allows to directly attach resources to concepts (the same resource can appear in several occurrence relations and be accessible from more than one concept). Relations (associations, occurrence) and concept labels can be deined inside scopes. This allows simply implementing annotations (or points of view in the memory).
To deinitely adopt this formalism, we veriied it was possible to take into account ontological features, mainly the relation superclass-subclass for building hierarchies of concepts.
An Example
Lets consider the notion inite set; its deinition is a inite set is a set which has a cardinal . It is represented by: three topics: inite set, set, cardinal Two associations: superclass-subclass (subsumption link between inite set and set), possess.
The notion inite set is treated on the Web site http://www.planete-maths.com/html/. Figure 3 presents the XTM syntax of the topics map formalism corresponding to this sentence.
Figure 3. Extrait XTM dune topic map

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. </topicMap> </roleSpec> <topicRef xlink:href=#t-Cardinal/> </membre> </association> </ roleSpec> <topicRef xlink:href=#t-inite-set/> </membre> <membre> <roleSpec> <topicRef xlink:href=#t-numberElement/> </topic> <association id= A-LA001 > <instanceOf> <topicRef xlink:href= #t-Possess /> </instanceOf> <membre> <roleSpec> <topicRef xlink:href= #t-set/> </topic> <topic id= t-Cardinal> <instanceOf> <topicRef xlink:href= #t-B31OBJECT /> </instanceOf> <baseName> <baseNameString> Cardinal </baseNameString> </baseName> <topicMap xmlns=http://www.topicmaps.org/xtm/1.0/ xmlns:xlink=http://www.w3.org/1999/xlink> <topic id= t-inite-set> <instanceOf> <topicRef xlink:href= #t-set /> </instanceOf> <baseName> <scope><topicRef xlink:href=#FR/></scope> <baseNameString> Ensemble ini </baseNameString> </baseName> <occurrence> <instanceOf> <topicRef xlink:href=#Site/> </instanceOf> <resourceRef xlink : href= http://www.planete-maths.com/html//> </occurrence>
160 Abel
The Environment E-MEMORAe

With the organizational environment E-MEMORAe (Abel, Benayache, Lenne, & Moulin, 2006), resources can be reached by navigating through a graphical display of the ontologies on which it relies. It integrates tools to facilitate autonomous learning. We present this environment by using examples coming from the B31.1 application.
The User Interface

E-MEMORAe aims at helping the users of the memory acquire the notions of a given course. To this end, the users have to navigate through the application ontology that is related to the course and access the indexed resources thanks to this ontology. The general principle is to propose to the learner, at each step, either precise information on what he is searching for or graphically displayed links that allow him to continue navigating through the memory. He has no need to use the keyboard in order to formulate a request, even if it is possible through the environment. To be more precise, the user interface (cf. Figure 4) proposes: Entry points (left of the screen) allowing one to start the navigation with a given concept: An entry point provides a direct access to a concept of the memory and, consequently, to the part of the memory dedicated to notions. The person who is in charge of the course has to deine the notions that (s)he considers as essential. Resources (bottom of the screen), the contents of which are related to the current concept: They are ordered by type (books, course notes, sites, examples, comments, etc.). Starting from a notion, an entry point, or a notion reached by the means of the ontology, the user can directly access associated resources. Descriptions of these resources help the user to choose among them. A short deinition of the current notion: The learner has a preview of the notion; he can decide if (s)he has to work it or not. A history of the navigation: The learner can remind and be aware of the path (s)he followed before. Of course, (s)he can get back to a previously studied notion if (s)he wants to. Least but not last, the part of the ontology describing the current resource is displayed at the centre of the screen.
If the learner wants access to a notion that is not an entry point, he or she has to choose the entry point that he or she thinks is the closest point from the searched notion.
Figure 4. Vertical navigation in the memory
Figure 5. A book resource
Display the resource
Send
Print the information
162 Abel
Learning by Exploration through the Memory

Vertical navigation allows exploration of subsumption relations and related concepts. For example, if the user wants to discover the inite set notion, the best entry point is set (population). By choosing this entry point, (s)he has access to the local taxonomy associated with the notion of set. Among the sub-concepts of set, (s)he can ind inite set. By clicking on this concept, a local taxonomy centred on this new concept is displayed (cf. Figure 4). Thus the learner can browse the ontology using its taxonomic form. Some presentation rules are used in order to help the user visually explore this hierarchical organization: The current concept C is at the centre of the screen; all the sub-concepts of C that represent more speciic notions are presented; at last, the super-concept of C, which represents a more general notion, is also presented. We did not ind it useful to extend this representation. Our goal was to keep it understandable. To end with the hierarchical navigation, let us inally note that the representation uses semantic axes. In order to visualize them, we used different colours for each of these axes. At this stage, their meaning is not explicit. Let us suppose now that the learner decides to temporarily stop the navigation and focus on a particular concept. This concept is at irst described by a short deinition. If the user wants to learn more on the selected notion, (s)he has access to a list of resources ordered by type. For example, Figure 4 shows that if the user wants to deepen the notion of Finite Set, (s)he can select among the associated resourcesfor example, a book entitled Mathematics for Computer Scienceby left-clicking on the name of this resource. A description text is then displayed in a new window (cf. Figure 5). Other bibliographic information such as ISBN number, authors, publisher, and so forth are also available. When the resource is digital, it can be displayed or sent to someone by e-mail. A concept can refer to concepts other than those that are displayed in the taxonomy. Access to these concepts is sometimes needed in order to understand some notions. Proximity relations (other than subsumption) are useful for that. Examples of these relations are: prerequisiteof, in-the-deinition-of, suggests, and so forth. Other application-speciic relations such as subset-of, has-cardinal, and so forth can also be considered. We call this kind of navigation horizontal navigation, in comparison with the vertical navigation that we considered before. These relations are accessed by right-clicking on the source concept C; a popup menu contextually displays the available relations starting from C. Let us consider one more time the case of the inite set concept (cf. Figure 4). Among the available horizontal relations, the learner can choose, for example, prerequisite-of and learn more about prerequisite notions such as countable set or cardinal (cf. Figure 6). Choosing the countable set concept in the list of the prerequisite concepts of inite set allows switching back to a vertical navigation centred on this new concept of countable set. Finally, one can see that the navigation through the application ontology is made very easy by combining vertical (left-click) and horizontal (right-click) moves.
Figure 6. Horizontal Navigation
Figure 7. Searching the inite set notion
164 Abel
After each exploration action made by the learner, the history (cf. Figures 4 and 6, right frame) is actualized. This history keeps track of the path followed by the user during his (her) exploration. Of course, it is possible to go back to a previously visited notion by clicking on it in the history.
Learning by Querying the Memory

We have seen in the previous section how to access the contents of the memory by navigation (consult menu). It is also possible to directly access a notion by querying the memory (Search menu). This interface allows searching for a word in the textual data contained in the memory. The word can be all or a part of an author, notion, or resource name, or it can be included in the textual description of the notion, and so forth. For example, Figure 7 shows the search results for the set word. When choosing a notion in the search results, the user has access to the part of the taxonomy related to this notion. He or she can continue to explore the memory horizontally or vertically (cf. learning by exploration through the memory).
Architecture
The architecture of the E-MEMORAe environment is a three-tier architecture: PHP/MySQL, Appach, and JavaScript/HTML+SVG. This type of architecture separates the application in three levels of distinct services: presentation, treatment, and storage.
Figure 8. Functional architecture of the E-MEMORAe environment

PRESENTATION XTM file (import) PROCESSING STORAGE
JavaScript menu Topic maps HTML pages BD conversion PHP/SQL conversion XTM conversion
MEMORAe Web Site
SVG graph PHP/SQL codes
MySQL
DB
XTM File (export)
Figure 8 shows the functional architecture of the E-MEMORAe environment: one can indeed ind a storage part (MySql data base), a treatment part (Topic Maps modeling, etc.), and an information presentation part. The MEMORAe web site is accessible at the following URL: http://www.hds.utc. fr/~ememorae/Site-MEMORAe/.
Conclusion
We presented in this chapter a goal-driven organizational approach to facilitate the search
for appropriate resources to acquire new competencies. Indeed, numerous resources may be used during e-learning. Their access is a real problem. E-learning becomes part of a complex organizational conduct. Learners have to access the right resources at the right time. This kind of problem is a problem of knowledge management, and we propose to achieve it in using the concept of learning organizational memory based on ontologies. This memory is different from a classical organizational memory because its goal is to provide pedagogical users with content. This content is the result of two pieces of work: (1) the capitalization of knowledge, information, and resources relating to the learning context (for example, a training); (2) a pedagogical work concerning the choice and the organization of this capitalization. The opposite of the approach that is generally adopted with learning objects repositories or thematic resources bases, this course memory is bound to be directly used by learners. This implies doing an earlier part of the instructional design work. Let us note, however, that this approach is only feasible with learners having self-regulating abilities. The E-MEMORAe environment we developed is based on the use of ontologies and topic maps. It is used as a support for e-learning. The objective is to help users understand the notions starting from resources selected by teachers. The indexing of the documents is supplemented by pedagogical criteria that help the learner appreciate his or her relevance. We think that using such a memory enhances the activity and the autonomy of the learner. We made a irst evaluation of E-MEMORAe with students in the framework of the B31.1 mathematics training at the University of Picardy in France and of the NF01 algorithms training at the University of Compigne. Our objective was to see how learners could discover alone new notions to learn through E-MEMORAe. In order to assess the understanding of these notions, the learners had to solve some problems related to the notions in a given time. In the two cases, we obtained encouraging results: the majority of students solved the problem and accessed to the right notions (seen in the recorded history). After presenting our prototype to our colleagues, we have received different demands to use E-MEMORAe. We plan to assess E-MEMORAe according to the teachers point of view.
166 Abel
Currently we are working on the extension of E-MEMORAe in order to assess such a memory for collaborative learning, computer support for collaborative learning (Stahl, 2002). We also search partners from industry to test E-MEMORAe in an organizational learning context.
References
Abecker, & Decker. (1999). Organisational memory: Knowledge acquisition, integration, and retrieval issues. In F. Pope (Ed.), Knowledge based systems, survey and future directions. In Proceedings of the 5th German Conference on Knowledge Based Systems, Lectures Notes in Artiicial (LNAI), (Vol. 1570). Springer-Verlag. Abel, M. H., Barry, C., Benayache, A., Chaput, B., Lenne, D., & Moulin, C. (2004). Ontology-based organizational memory for e-learning. Educational Technology & Society Journal, 7(4), 98-111. Abel, M. H., Benayache, A., Lenne, D., & Moulin, C. (2006). E-MEMORAe: A contentoriented environment for e-learning. In S. Pierre (Ed.), E-learning networked environments and Architectures: A Knowledge processing perspective (pp. 186-205). Springer Book. Alliance of Remote Instructional and Distribution Networks for Europe (ARIADNE). (2004). Alliance of remote instructional and distribution networks for Europe 2004. Balasubramanian, V. (1995, May). Organizational learning and information. ISWorld Net. Retrieved from http://www.e-papyrus.com/personal/orglrn.html Baugh, J. (1997, Spring). Rewarding competencies in latter organizations. Competency: The Journal of Performance through People, 4(3). Breuker, J., & Muntjewerff, A. (1999, July 18-19). Ontological Modelling for Designing Educational Systems. In Proceedings of The Workshop on Ontologies for Intelligent Educational Systems, Ninth International Conference on Artiicial Intelligence in Education, AI-ED99, Le Mans, France Chandrasekaran, B., Josephson, J. R., & Benjamins, R. (1998, April 23). The ontology of tasks and methods. In Proceedings of KAW98, Eleventh Workshop on Knowledge Acquisition, Modeling and Management Voyager Inn, Banff, Alberta, Canada. Conklin, E. (1996). Designing organizational memory: Preserving intellectual assets in a knowledge economy. Group Decision Support Systems, Inc. Dieng, R., Corby, O., Giboin, A., & Ribire M. (1998). Methods and tools for corporate knowledge management. In Proceedings of the 11th Workshop on Knowledge Acquisition, Modeling and Management (KAW98), Banff, Alberta, Canada. Retrieved from http://ksi.cpsc.ucalgary.ca/KAW/KAW98/KAW98Proc.html Guarino, N., & Giaretta, P. (1995). Ontologies and knowledge bases, towards a terminological clariication. In E. N. Mars (Ed.), Towards very large knowledge bases: knowledge building and knowledge sharing (pp. 25-32). ISO Press.
Gruber, T. (1993). A Translation approach to portable ontology speciications. Knowledge Acquisition, 5(2), 199-220. Harzallah, M., Leclre, M., & Trichet, F. (2002). In Proceedings of Fourteenth International Conference on Software Engineering and Knowledge Engineering (SEKE02), Ischia, Italy. Harzallah, M., & Vernadat, F. (2002). IT-based competency modeling and management: From theory to practice. Enterprise Engineering and Operations, Computers in Industry, 48(2), 157-179. International Electronical Commission (IEC). (1999, April 19). International organisation for standardization (ISO). Topic map (International Standard ISO/IEC 13250). Kassel, G. (2005). Integration of the DOLCE top-level ontology into the OntoSpec methodology. https://hal.ccsd.cnrs.fr/ccsd-00012203 Klemke, R. (2000, October 30-31). Context framework: An open approach to enhance organisational memory systems with context modelling techniques. In Proceedings of PAKM2000: Third International Conference on Practical Aspects of Knowledge Management, Basel, Switzerland. LOM. (2002). LOM standard, document 2002; 1484.12.1. Martin, P. (1993). Scholarly Web sites as organisational memory system. Retrieved from http://www.cs.nott.ac.uk/~hla/HTF/HTFVII/Martin.htm#Conklin%201993 Paquette, G. (2001). Telelearning systems engineering: Towards a new ISD model. Journal of Structural Learning and Intelligent Systems, 14(4), 319-154. Park, J. (2002). XML topic maps, creating and using topic maps for the Web. Jack Park Editor 2002. Rabarijaona, A., Dieng, R., Corby O., & Ouaddari, R. (2000). Building a XML-based corporate memory [Special issue]. IEEE Intelligent Systems, 56-64. Sicilia, M. A. (2005). Ontology-based competency management: Infrastructures for the knowledge-intensive learning organization. In M. D. Lytras & A. Naeve (Eds.), Intelligent learning infrastructures in knowledge intensive organizations: A Semantic Web perspective (pp. 302-324). Hershey, PA: Idea Group Inc. Sowa, J. F. (2003). Ontology. Retrieved from http://www.jfsowa.com/ontology/ Stahl, G. (Ed.) (2002). Computer support for collaborative learning: Foundations for a CSCL community. In Proceeding of CSCL 2002, Boulder, Colorado. Steels, L. (1993). Corporate knowledge management. In Proceedings of the International Symposium on the Management of Industrial and Corporate Knowledge (ISMICK93) (pp. 9-30). Van Heijst, G. (1997, September). Knowledge infrastructures: Re-engineering for learning. In Proceedings of Workshop on Knowledge-Based Systems for Knowledge Management in Enterprises, The 21st Annual German Conference on AI 97, Freiburg, Germany. Retrieved from http://www.dfki.uni-kl.de/~aabecker/Freiburg/Heijst/index.htm Van Heijst, G., Schreiber, A., & Wielinga, B. (1997). Using explicit ontologies in KBS development. International Journal of Human-Computer Studies, 46, 183-298.
168 Abel
XTM. (2001, 3 March). TopicMaps.org XTM authoring group, XML Topic Maps (XTM) (1.0): TopicMaps.org Speciication. Wenger, E. (1998). Communities of practice. Learning, meaning and identity. Learning in doing: social, cognitive, computational perspectives. Cambridge, MA: Cambridge University Press. Wooddruff, C. (1991). Competent by any other name. People Management Journal.
An Ontological Representation of Competencies as Codiied Knowledge 169
Chapter VII
An Ontological Representation of Competencies as Codiied Knowledge

Salvador Snchez-Alonso, University of Alcal, Spain Dirk Frosch-Wilke, University of Applied Sciences, Germany
Abstract
In current organizations, the models of knowledge creation include speciic processes and elements that drive the production of knowledge aimed at satisfying organizational objectives. The knowledge life cycle (KLC) model of the Knowledge Management Consortium International (KMCI) provides a comprehensive framework for situating competencies as part of the organizational context. Recent work on the use of ontologies for the explicit description of competency-related terms and relations can be used as the basis for a study on the ontological representation of competencies as codiied knowledge, situating those deinitions in the KMCI lifecycle model. In this chapter, we discuss the similarities between the life cycle of knowledge management (KM) and the processes in which competencies are identiied and assessed. The concept of competency, as well as the standard deinitions for this term that coexist nowadays, will then be connected to existing KLC models in order to provide a more comprehensive framework for competency management in a wider KM framework. This paper also depicts the frameworks integration into the KLC of the KMCI in the form of ontological deinitions.
170 Snchez-Alonso & Frosch-Wilke
Introduction
Models of knowledge creation inside organizations are considered as dynamic processes of development that evolve over time (Cavaleri & Reed, 2000). These models provide a breakdown of the creation process in terms of concrete processes and elements that drive the overall production of knowledge as targeted to satisfy organizational expectations. For example, the knowledge life cycle (KLC) model of the Knowledge Management Consortium International (KMCI, http://www.kmci.org) distinguishes the knowledge processing environment (KPE) from the business processing environment (BPE), describing the latter as the context of actual usage and ield assessment of the claims formulated, produced and evaluated in the former. As the KPE is divided into two sub-processes, namely knowledge production (KP) and knowledge integration, the existence of a BPE emphasizes the fact that knowledge codiied in artefacts as part of KP processes and disseminated as part of KI processes will be subject to further validation in actual business experience. Previous work has shown KLC models as a comprehensive framework for situating learning-oriented artefacts in an organizational context (Sanchez-Alonso & Frosch-Wilke, 2005; Sicilia, 2005). The work of Sicilia (2005) has demonstrated that the design and creation of learning resources as described by Downes (2004) is not essentially different from knowledge production. The integration processes, in particular, might be considered to subsume programmed organizational learning activities. Thinking about learning as an outcome of the need to acquire new competencies, learning activities inside the organization can be considered enablers of knowledge acquisition activities. In this context, the concept of competency becomes essential in the KLC model, both as a prerequisite to perform knowledge acquisition activities and as an outcome of these kinds of activities. Furthermore, meta-claims about the knowledge producedin the case of competenciesmay be interpreted as the recording of usage conditions, hypotheses, and assumptions on the acquisition of the competencies evaluated. In consequence, the concepts related with competency management can be put in connection with existing KLC models, in an attempt to provide a comprehensive framework for reuse-oriented competency management and KM. In this chapter, we approach the integration of concepts related to competencies into the framework of the KLC. This would clarify the relationships between knowledge management and competency deinition standard efforts. The method to develop the conceptual integration is that of engineering an initial ontological description for the main concepts, connecting them to existing ontological databases. This continues existing work described by Sicilia, Lytras, Rodrguez, and Garca (2006) regarding the ontological description of learning activities as an extension of the ontology of KM described recently by Holsapple and Joshi (2004). Formal ontologies (Baader, Calvanese, McGuinness, Nardi, & Patel-Schneider, 2003) are a vehicle for the representation of shared conceptualizations that is useful for technologyintensive organizations. Ontologies based on description logics (Gruber, 1995) or related formalisms provide the added beneit of enabling certain kinds of reasoning over the terms, relations, and axioms that describe the domain. A pragmatic beneit of the use of formal ontologies is that it is accompanied by a growing body of semantic Web (Berners-Lee, Lassila, & Hendler, 2001) tools, techniques, and knowledge. Previous work considered here as a point of departure (Sicilia, Garca, Snchez-Alonso, & Rodrguez, 2004) has described the integration of e-learning technology concepts with the OpenCyc knowledge base, the open source version of the Cyc system (Lenat, 1995).
The rest of this chapter is structured as follows. The second section describes the knowledge life cycle of the KMCI, as this is the framework for the subsequent discussion. The third section includes a brief discussion on some current deinitions of the term competency and details the most interesting efforts in the standardization of competency deinitions. The fourth section shows how competencies can be integrated in the knowledge life cycle of the KMCI, while the ifth section provides a preliminary mapping of competency-related concepts to terms in upper ontologies. Finally, conclusions are provided in the last section.
The Knowledge Life Cycle of the KMCI

Knowledge management is an area built on the assumption that each and every organization has a certain amount of valuable knowledge that is worth being captured, catalogued, and preserved with the main aim of sharing it whenever it is necessary. However, irst-generation KM, as it is referred by McElroy (1999) has not been considered fully satisfactory, which is probably due to an excessive emphasis on both knowledge integration and on the technology side as the answer to most questions. To many, this irst generation of knowledge management has supposed little more than document management and imaging becoming the reason why some feel that KM is an idea that amounts to little more than yesterdays information technologies trotted out in todays more fashionable clothes. Hopefully, a second generation of KM has emerged. This second generation of knowledge management is not so focused on the technology side but instead on the participants, the processes involved, and the social interactions and initiatives among them. The arrival of this second generation has introduced a number of new concepts and ideas, such as the knowledge life cycle, nested knowledge domains, containers of knowledge, organizational learning, the open enterprise, social innovation capital, and sustainable innovation, among others. While an in-depth discussion of this and other key ideas of this second generation of KM is out of the scope of this chapter, the interested readers are directed to the book by McElroy (2003). The knowledge life cycle, one of the previously mentioned ideas introduced as part of the second generation of KM, is a new view of KM that emphasizes knowledge production in detriment of the knowledge integration. The following explanation by McElroy points out the differences between the irst and the second generation of KM taking as criteria of comparison the KLC: While practitioners of irst-generation KM tend to begin with the rather convenient assumption that valuable knowledge already exists, practitioners of second generation KM do not. Instead, theyor wetake the position that knowledge is something that we produce in human social systems, and that we do so through individual and shared processes that have regularity to them. We can describe this process at an organizational level in the form of what is now being referred to as the knowledge life cycle, or KLC. (McElroy, 2003)
Figure 1. The Knowledge Life Cycle (KLC) Model of the KMCI
From this perspective, the Knowledge Management Consortium International, a non-proit association of knowledge and innovation management professionals from around the world (www.kmci.org) based in the U.S., has developed a model of KLC that is shown in Figure 1 taken from McElroy (n.d.). This model shows how the knowledge of an organization is held both subjectively in the minds of individuals and groups, and objectively in recorded or expressed form, shaping what is known as the distributed organizational knowledge base (DOKB) of the organization. The use of this knowledge in speciic business environments can lead to outcomes that either satisfy expectations or fail to do so. The former outcomes, known as matches, reinforce knowledge previously used, thereby leading to its re-use, whereas the later ones, known as mismatches, lead to adjustments in a business processing environment. Adjustments triggered by a mismatch introduce what is known as the single-loop learning. This single-loop learning means that the assumptions, or choices, made from within a range of pre-existing knowledge in the DOKB should be studied and probably corrected in the light of the results of the revision. Successive failures from single-loop learning to produce matches in expected or desired outcomes is understood as a problem and could lead to doubt about and probably reject pre-existing knowledge. Problems like these trigger knowledge processing efforts to produce and integrate new knowledge, in what is known as double-loop learning (Argyris & Schon, 1996). Double- loop learning starts with a problem claim formulation, an attempt to learn and state the speciic nature of the detected knowledge gap (or problem), followed by a process of knowledge production. The outcome of this process is a knowledge claim evaluation, which leads to surviving knowledge claims, falsiied knowledge claims, or undecided knowledge claims, as well as additional information about each of these outcomes (this information is known as metaclaims). The record of all the previously mentioned outcomes will be part of the DOKB after a number of activities in a process of knowledge integration. When the knowledge has successfully been integrated in the DOKB, the new claims and metaclaims are ready to be used in new business processing. The life cycle described is the framework for all the subsequent discussion. The following section will provide a brief introduction to the concept of competency, as well as detailed information on current efforts of standardization (IMS-RDCEO and HR-XML) intended to make it easier to integrate competency management into worklow and decision-support frameworks such as the KLC of the KMCI.
Competency: Deinition and Standards

At present, several different deinitions of the concept of competency coexist. Although most agree on a few core characteristics, it is interesting to provide a brief discussion about some of the most closely related work. The notion of competency is often considered a placeholder for knowledge, skill, abilities, and other characteristics (Sicilia, 2005). However, this view can be judged an excessive oversimpliication of the many facets of the use of the term (Hoffman, 1999). In a general sense, a competency can be deined as an underlying characteristic that leads to successful performance, which may include knowledge and skills as well as bodies of knowledge and
levels of motivation (Rothwell, n.d.). Another broad deinition is that included in the IMSRDCEO best practices and implementation guide (Cooper & Ostyn, 2002c): All classes of things that someone, or potentially something, can be competent in. Some authors believe that competencies encompass more than just knowledge and skill, as they focus on what is unique about individuals doing the work rather than what people must know or do to perform the work alone (Rothwell, n.d.). In this sense, the deinition included in the HR-XML seems to cope with this approach, as this is a much more inclusive deinition: A speciic, identiiable, deinable, and measurable knowledge, skill, ability and/or other deployment-related characteristic (e.g., attitude, behaviour, physical ability) which a human resource may possess and which is necessary for, or material to, the performance of an activity within a speciic business context. In the rest of this section, the most prominent approaches to competency standardization are studied. It should be remarked that, as it has been stated earlier, most agree on the core characteristics of competencies, even though all include their own deinitions and consequently refer to the term competency from their own perspective.
IMS-RDCEO
The IMS consortium (http://www.imsglobal.org) provides a speciication for competencies called reusable deinition of competency or educational objective (RDCEO). IMS-RDCEO deines an information model for describing, referencing, and exchanging deinitions of competencies, primarily in the context of online and distributed learning. This speciication allows to formally represent the most important characteristics of a competency, and its main aim is to enable interoperability among learning systems that deal with competency information. The complete speciication consists of three documents: IMS-RDCEO information model (Cooper & Ostyn, 2002a), including the complete description of the main elements of the speciication: semantics, structure, data types, value spaces, multiplicity, and obligation. This information model is purposely extensible, minimalist, and model-neutral. IMS-RDCEO XML binding (Cooper & Ostyn, 2002b), constituting only one example of the possible bindings that might use the information model, is a binding of the Information Model to XML version 1.0. IMS-RDCEO best practices and implementation guide (Cooper & Ostyn, 2002c), a non-normative set of rules about the application of both the information model and the XML binding, as well as examples to, for example, illustrate how the conceptual framework maps to practical uses.
The information model deines a set of elements of information in ive different categories that can be used to deine a competency. Hence, competency data may include a deinition of the competency, evidences of the competency, information about its context, and the scale (i.e., proiciency on a predetermined scale). Following this schema, a competency can be described by stating information in the following ive main categories:
1. 2. 3. 4. 5.
Identiier, subdivided into catalog and entry Title Deinition Description, subdivided into model source and statement Metadata, subdivided into RDCEO schema, RDCEO schema version, and additional metadata
The deinition of a competency, according to this schema, is shown in the following example, a simpliication of a broader example taken from Cooper and Ostyn (2002c):
<?xml version=1.0 encoding=utf-8?> <rdceo xsi:schemaLocation=http://www.w3.org/XML/1998/namespace xml.xsd xmlns=http:// www.imsglobal.org/xsd/imsrdceo_rootv1p0 xmlns:xsi=http://www.w3.org/2001/XMLSchemainstance> <identiier>http://www.imsglobal.org/ictional/rdceo_cat1.xml#pass_eg </identiier> <title> <langstring xml:lang=en-US>Reading IMS speciications</langstring> </title> <description> <langstring xml:lang=en-US> Reads and understands IMS Global Learning speciications </langstring> </description> <deinition> <model>IMS Competency WG</model> <statement statementname=Performance> <statementtext> <langstring xml:lang=en-US> Reads and understands IMS Global Learning speciications </langstring> </statementtext> </statement> </deinition> <metadata> <rdceoschema> IMS RDCEO </rdceoschema> <rdceoschemaversion> 1.0 </rdceoschemaversion> </metadata> </rdceo>
However, although IMS-RDCEO is explicitly intended to be integrated in the description of learner proiles and learning objects (Polsani, 2003), its underlying model provides similar capabilities to that of HR-XML, a general-purpose competency schema that will be detailed in the next section.
HR-XML
The HR-XML (http://www.hr-xml.org/) is an independent, non-proit consortium whose main aim is to enforce e-commerce and inter-company exchange of human resources data within a variety of business contexts. Represented by its membership in 22 countries, the main effort supported by this consortium is the development of standardized XML vocabularies for human resources, as well as standards for stafing and recruiting, compensation and beneits, and training and work force management. Major companies such as Addeco, Cisco Systems, PeopleSoft GmbH, IBM, Microsoft, and many others are currently members of the HR-XML Consortium. Up to the present, the HR-XML Consortium has produced a library of more than 100 interdependent XML schemas that deine the data elements for particular HR transactions, as well as options and constraints governing the use of those elements. It has also produced schemas covering major processes, as well as component schemas used across multiple business processes. For example, the assessments standard, facilitates employers to leverage the assessment tests, tools, and expertise offered by assessment service providers. One of the schemas provided by the HR-XML Consortium is the competencies recommendation. This set of recommendations about competencies allows the capture of information about evidence used to substantiate a competency and ratings and weights that can be used to rank, compare, and otherwise evaluate of the suficiency or desirability of a competency (Allen, 2006). The competencies schema is particularly relevant to processes involving the rating, measuring, comparing, or matching an asserted competency (for example, a skill claimed in a resume) against one that is demanded (for example, a skill required in a job description). This fact, added to the fact that this schema is intended as a module that can be incorporated within broader process-speciic schemas, facilitates its use outside the HR domain as a general-purpose competency schema and makes it possible its integration in diverse frameworks. The only requirement for those frameworks is, of course, the use of some kind of competency management. Figure 2, taken from Allen (2006), depicts the components of a competency after what is stated in the HR-XML recommendation. This standard deines a number of elements of information for each competency, as well as the structure and information of the competency evidences and weights, among other information. The deinition of a competency, according to this schema, is shown in the following example, again taken from Allen (2006):
Figure 2. Components of a competency after the HR-XML recommendation
<Competency
name = Reading Comprehension
description = Understanding written sentences and paragraphs in work related documents> <CompetencyId id = 2.A.1.a/> <TaxonomyId id = O*NET idOwner = National O*Net Consortium description = Occupational Information Network/> <CompetencyWeight type = x:Importance> <NumericValue maxValue = 100 minValue=1>85</NumericValue> </CompetencyWeight> <CompetencyWeight type = x:Level> <NumericValue maxValue = 100 minValue=1> 57 </NumericValue> </CompetencyWeight> </Competency> HR-XML can also be used as a wrapper of an RDCEO record by using a URN, as shown in the following example taken from (2006): <Competency description=Can read and understand W3C Schema Language 1.0 name=Reads and Understands W3C Schema xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance xsi:noNamespaceSchemaLocation=Competencies-1_0.xsd> <CompetencyId description=IMS Global Example Competency Catalogue id=URN:X-IMS-PLIRID-V0::6ba7b8149dad11d180b400c04fd430c8/>  </Competency>
Integrating Competencies in the Knowledge Life Cycle (KLC)

In this section, the related concepts of competencies are described as the main elements to be integrated as resources in the KLC. Then, their integration inside the KLC model of the KMCI is described. The process of acquisition of a competency (or knowledge in a broader sense) usually starts from a business need originated in the context of the organization. This need triggers a process of assessing whether the organization can deal with the given need or not, which is commonly referred to as knowledge gap analysis (Sunassee & Sewry, 2002). This assessment process essentially consists of matching the competencies required for the newly appointed needs with the available ones. When the result of this process is not satisfactory, a process of acquiring the competencies identiied begins. After this process is considered inished, some kind of assessment would take place and, later on, an update of the registry of available competencies should be carried out. The newly acquired competencies might change the position of the organization to offer services or products, closing in this manner the so-called knowledge acquisition loop. As a knowledge acquisition endeavour, the just described cycle can be expressed in terms of knowledge management activities and products. According to the ontology of knowledge management by Holsapple and Joshi (2004), competences can be considered capabilities attributable to processors of knowledge representations (KR), and the inal learning activities carried out to obtain the competencies needed can be seen as speciic types of knowledge manipulation activities (KMA), consisting of knowledge acquisition or, eventually, transformation. Furthermore, processors are considered to have some capabilities as analysed by Sicilia (2005). This author identiies the terms (or, as it is called in the original work, abstract elements) related to competency management as a previous step to integrating them in the KLC: Competency registry: Not a term but instead a set of terms related to the description of competencies in detail, particularly those of the existing employees. Needs: an expression of the required competencies that can be represented in the form of triples (C: competency_description, L: level, I: intensity). According to Sicilia (2005), the level desired for the competency is expressed as an overall aggregate level that maps the levels of individuals inside the organization, whereas the intensity is an estimation of the part of the workforce that is needed to have the competency. Available competencies: A detailed record of employees competencies. Required competencies: A subset of the needs after matching them with the competency registry. Aimed at describing needs not covered by the existing competencies. Competency gap analysis: A process used to obtain the required competencies. This process has a collection of needs and a competency registry as inputs and the required competencies as outputs. Competencies update: The process of creation or update of competency instances, aimed at keeping the competency registry updated.
The main elements of the integration of the above listed terms to the KML model are depicted in Figure 3, which has been elaborated from the original KLC of the KMCI by including mappings to concrete competency usage points.
Figure 3. Mapping of the main terms of competency management to the KLC model
Mapping Competency-Related Concepts to Terms in Upper Ontologies

Competency management can be integrated in the broader framework of a knowledge management lifecycle to provide guidance for Information System development and insights into notions of organizational value of competencies, among others. However, even though current standards for the description of competencies are intended to provide data aimed at being interchanged by machines, the information they contain is currently intended for human interpretation. Present practices result in data lacking machine-understandable characteristics, which seriously hampers their use in semantic Web environments. Ontologies can be used to improve the quality of competency descriptions, but translating current competency descriptions that conform to a given standard (such as HR-XML) to an ontology language is not enough by itself to provide computational semantics to those descriptions. The right step in this direction is the integration of competency terms with high-level terms and deinitions in upper ontologies, as this constitutes an interesting direction for bringing explicit semantics to competency descriptions. An upper ontology is a large general knowledge base that includes deinitions of concepts, relations, properties, constraints, and instances, as well as reasoning capabilities on these elements. Limited to generic, high-level, abstract concepts, general enough to address a broad range of domains, upper ontologies do not include concepts speciic to given domains or do not focus on them. Opencyc (http://www.opencyc.org), an upper ontology for all of human consensus reality, includes more than 47,000 concepts, 306,000 assertions about them, an inference engine, a browser for the knowledge base, and other useful tools, which makes it one of the major efforts in the ield. It is the open source version of the larger Cyc knowledge base (Lenat, 1995), a huge representation of the fundamentals of human knowledge made up of facts, rules, and heuristics for reasoning about objects and events. The rest of this section sketches the main integration points of the KLC with competencies in the framework of existing work in formally conceptualizing KM. The direct mapping of the essential concepts described in this chapter and the terms in the Holsapple and Joshi (2004) ontology of KM, enables an effective integration of ontology-based KM and organizational competency management in existing upper ontologies such as OpenCyc. The ontology of knowledge management by Holsapple and Joshi (2004) describes fundamental KM concepts and axioms. In this ontology, the term KM is deined as an entitys systematic and deliberate efforts to expand, cultivate, and apply available knowledge in ways that add value to the entity [..]. This requires the early deinition of entities capable of engaging in KM, which include at least individuals, organizations, collaborating organizations, and nations. The term Organization in OpenCyc covers all such entities. Accordingly, the concept of knowledge processor as a member of an entity can be modeled by the concept of IntelligentAgent, which is by deinition capable of knowing and acting, and of employing its knowledge in its actions. Humans are by logical deinition intelligent agents, and certain software pieces may also be, since they are not restricted to not being able to know. The subtype MultiIndividualAgent its the deinition of collective agents. According to Cavaleri and Reed (2000), knowledge creation is the result of efforts by agents, acting
either as individuals, or collaboratively, as an element of a system, to make sense of their environment. This deinition focuses on the identity of the organization as a key driver of its learning behaviour and is complemented by a concrete view on creation as a process in which agents apply rules to perceived sets of circumstances to attain desired outcomes. The deinition of knowledge as that which is conveyed by usable representations can be integrated in OpenCyc by considering usable representations as information bearing things, that is, Each instance of InformationBearingThing (or IBT) is an item that contains information (for an agent who knows how to interpret it). This is appropriate at least for CKC that are tangible outcomes of the production process. Nevertheless, the KLC emphasizes the evaluation of information as tentative knowledge claims, so that terms subsumed by IBT are required to adequately it in the KLC, including the following:
EvaluatedKnowledgeClaim representing the surviving claims, which are required to have been subjectTo at least one KnowledgeClaimEvaluation process with a positive
outcome.
FalsiiedKnowledgeClaims, with the opposite deinition.
The rest of the KnowledgeClaim instances are subsumed by UndecidedKnowledgeClaim, representing different states before or after claim evaluation.
KnowledgeClaimEvaluation instances are a concrete kind of knowledge manipulation. The recognizable kinds of knowledge manipulation are referred to as KnowledgeManipulationActivities, and thus, CompetencyAssessment may be considered a subtype of KMA. In OpenCyc, activities are represented as Actions, collections of Events carried out (doneBy) a doer.
This generic concept of action can be specialized to represent KMA executions by restricting them to be carried out by intelligent agents. The predicate ibtUsed (subsuming the abovementioned subjectTo) can be used to represent the knowledge representations manipulated by KMAs. In addition, since KM activities are deliberate, it is preferable to use the subclass PurposefulAction. Each of the processes in Figure 3 can be considered KMAs. Competencies are represented in OpenCyc. However, the attribute Competence, subsumed by Quantity-ScriptPerformance (aimed at describing the manner in which an actor performs an action) and ScriptPerformanceAttributeType (aimed at describing the manner in which an action is performed), is deined as a general attribute to deine the level of skill with which an agent performs some task. For that reason, this notion of competency is considered too general and thus inadequate to deine the concept competency as it has been used in this work. The most accurate way to deine competencies is that of deining OpenCyc Actions. Accordingly, predicates related to the deinition, description, and use of competencies would be derived from the predicate SkillLevel. This OpenCyc predicate, as stated in the OpenCyc knowledge base, deines a relation between performers and types of actions in the follow-
ing manner: some performer (probably, but not necessarily, an Agent) has the ability to play a given role in a speciic type of Event with a certain level of PerformanceAttribute. For example: (skillLevel MagicJohnson PlayingBasket performedBy Creativity #$High) Meaning that, in general, Magic Johnson can play basket with great creativity. If this behaviour is translated to competency management, the knowledge about the fact that the employee Angela has a particular competence should be stated like this: (skillLevel Angela SpeakingInPublic performedBy Competence #$VeryHigh) In this example, the competency is represented by the action SpeakingInPublic, whereas the attribute Competence is just one qualiier to describe the manner in which the competency SpeakingInPublic is performed by the employee (others might be Charisma, Precision, Dexterity, or Gracefulness). This form of modeling competencies is similar to the manner in which competencies are deined in HR-XML and opens the door to a full description of other concepts related as the triples (competency, level, and intensity), easy to model in OpenCyc through a speciically-designed ternary predicate.
Conclusion
Competency management can be integrated in the broader framework of a knowledge management lifecycle to provide guidance for information system development and insights into notions of organizational value of competencies. Concretely, a feasible integration of such concepts into the KMCI KLC model has been described. Current standards for the deinition, sharing, and exchange of competencies, as well as the information about competencies that conform to this speciication included in the DOKB of the organizations, are intended for interchange by machines, but instead they are currently intended for human interpretation only. Their main aim is to enable interoperability among systems that deal with competency information by providing a means for them to refer to common deinitions with common meanings. However, these efforts insist in the construction of models of competencies but do not focus on semantic interoperability. The resulting ontological schemes shown in this chapter are intended as a foundation for further research and standardization activities. The authors consider that an additional effort of integrating current standards in commonsense knowledge bases, such as OpenCyc, through formalizing concepts in ontology languages, can be particularly rewarding as it would provide competency management with the beneits of the Semantic Web vision.
Acknowledgment
This work is funded by the LUISA EU project (FP6-027149).
References
Allen, C. (Ed.). (2003). HR-XML recommendation. Competencies (Measurable Characteristics). Recommendation, 2006 Feb 28. Retrieved March 1, 2006, from http://www. hr-xml.org/ Argyris, C., & Schon, D. A. (1996). Organizational learning II: Theory, method, and practice. Reading, MA: Addison Wesley. Baader, F., Calvanese, D., McGuinness, D., Nardi, D., & Patel-Schneider, P. (Eds.). (2003). The description logic handbook. Theory, implementation and applications. Cambridge. Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The semantic Web. Scientiic American, 284(5), 34-43. Cavaleri, S., & Reed, F. (2000). Designing knowledge creating processes. Knowledge and Innovation, 1(1). Cooper, A., & Ostyn, C. (Eds.). (2002a). IMS reusable deinition of competency or educational objective: Information model, (version 1.0), inal speciication. Retrieved March 1, 2006, from http://www.imsglobal.org/competencies/rdceov1p0/imsrdceo_infov1p0. html Cooper, A., & Ostyn, C. (Eds.). (2002b). IMS reusable deinition of competency or educational objective: XML binding, (version 1.0), inal speciication. Retrieved March 1, 2006, from http://www.imsglobal.org/competencies/rdceov1p0/imsrdceo_bindv1p0.html Cooper, A., & Ostyn, C. (Eds.). (2002c). IMS reusable deinition of competency or educational objective: Best practice and implementation guide, (version 1.0), inal speciication. Retrieved March 1, 2006, from http://www.imsglobal.org/competencies/rdceov1p0/imsrdceo_bestv1p0.html Gruber T. (1995). Towards principles for the design of ontologies used for knowledge sharing. International Journal of Human-Computer Studies, 43(5/6), 907-928. Hoffmann, T. (1999). The meanings of competency. Journal of European Industrial Training, 23(6), 275-286. Holsapple, C. W., & Joshi, K. D. (2004). A formal knowledge management ontology: Conduct, activities, resources, and inluences. Journal of the American Society for Information Science and Technology, 55(7), 593-612. Lenat, D. B. (1995). Cyc: A large-scale investment in knowledge infrastructure. Communications of the ACM, 38(11), 33-38. McElroy, M. W. (1999, October). The second generation of knowledge management. Knowledge Management Magazine.
McElroy, M. W. (2003). The new knowledge managementComplexity, learning, and sustainable innovation. Boston: KMCI Press/Butterworth-Heinemann. McElroy, M. W. (n.d.). The knowledge life cycle (KLC). Retrieved March 1, 2006, from http://www.kmci.org/media/Knowledge_Life_Cycle.pdf Polsani, P. R. (2003). Use and abuse of reusable learning objects. Journal of Digital Information, 3(4). Rothwell, W. J. (n.d.). A report on workplace learner competencies. Retrieved March 1, 2006, from http://www.ilpi.wayne.edu/iles/roth_present.pdf Snchez-Alonso, S., & Frosch-Wilke, D. (2005). An ontological representation of learning objects and learning designs as codiied knowledge. The Learning Organization, 12(5), 471-479. Sicilia, M. A. (2005). Ontology-based competency management: Infrastructures for the knowledge-intensive learning organization. In M. D. Lytras & A. Naeve (Eds.), Intelligent learning infrastructures in knowledge intensive organizations: A semantic Web perspective (pp. 302-324). Hershey, PA: Information Science Publishing. Sicilia, M.A., Garca, E., Snchez-Alonso, S., & Rodrguez, E. (2004). Describing learning object types in ontological structures: Towards specialized pedagogical selection. In Proceedings of ED-MEDIA 2004: World Conference on Educational Multimedia, Hypermedia and Telecommunications, 2093-2097. Sicilia, M. A., Lytras, M., Rodrguez, E., & Garca, E. (2006). Integrating descriptions of knowledge management learning activities into large ontological structures: A case study. Data and Knowledge Engineering. Sunassee, N., & Sewry, D. (2002). A theoretical framework for knowledge management implementation. In Proceedings of the 2002 Annual Research Conference of the South African Institute of Enablement Through Technology (pp. 235-245).
Ontology-Based Approach to Formalization of Competencies 185
Chapter VIII
Ontology-Based Approach to Formalization of Competencies

April Ng, Simon Fraser University Surrey, Canada Marek Hatala, Simon Fraser University Surrey, Canada
Abstract
Competency-based learning has been used in training employees to acquire the necessary skills for an organization to be successful in a dynamic and ever-changing environment. One of the core activities in competency-based training is learning material acquisition. Standardization efforts have made the retrieval of educational materials, also called learning objects, easier by describing them in pre-deined metadata schema. However, the existing standardized metadata schema and practices of learning object metadata annotation do not support automatic selection of resources by speciic competency requirements in the competency-based learning. We propose an ontology-based competency formalization approach as a way of representing competency-related information together with other metadata in ontology in order to enhance machine automation in resources retrieval. The approach represents competency with properties of deinition, knowledge reference, evidence of proiciency, and level of proiciency. The effectiveness of resource selection from each of the properties is evaluated.
186 Ng & Hatala
Introduction
Organizations are under increasing pressure from competitors, new products introduction, and new market entrants on a daily basis in the era of globalization. Successful businesses understand that strengthening employee attitudes and enhancing the capabilities of staff members is the key for gaining a competitive advantage (Sanchez & Heene, 2000). At the same time, individuals adjust to role changes during downsizing, merging, or organizational cultural changes (Dubois & Rothwell, 2004) in a dynamic and ever-changing world. Human resources and operation managers fully recognize that staff performance is highly correlated to the competencies they possess and the way they use competencies to achieve work outcomes (Dubois & Rothwell, 2004). Competency or competencies are the knowledge and skills that are required to perform a task successfully. In this environment, competency-based training offers a wide range of beneits. By focusing on building individual competence, it is highly customizable for the learners personal needs. Learners are more proactive and take initiative in inding out how to improve their existing competencies in order to reach the competencies deined by a career or by an organization. Competency-based training also helps deine and clarify the expected outcomes of any strategy implementation and therefore contributes to the organizations success. One of the cost-effective ways to provide personalized learning experience according to widely varied competency requirements is to use learning objects in multiple contexts. As a reusable resource that supports learning, a set of learning objects can be shared or reused to create different combination of instruction sequences in various circumstances itting personal needs. Examples of learning objects include a section of a book about intangible assets appearing in a four-month as well as a one-year accounting course. Finding learning objects for competency requirements starts from looking at the metadata of learning objects. Metadata describes information about a learning object such as title, author, created date, and purpose. However, businesses that apply learning objects on competency-based training are not able to take full advantage unless learning objects are described with the competencies that they are designed for, so as to allow a machine to automatically acquire and select those with the required competencies.
Personalization by Competency Requirements

A key commonality of competency-based training is to personalize learning experience with matching resources that it individual competency requirements of learners. One of the beneits of using learning objects in personalization is reduced cost by reusing and sharing them individually and widely from a large repository. Learning management systems add a higher level of personalization by discovering learning objects that it personal ability and requirements, including learners competency requirements (Blackmon & Rehak, 2003; Martinez, 2002; Karampiperis & Sampson, 2004; Schmidt, 2004; Shen & Shen, 2004; Woelk, 2002). Learners are able to tell what they need by performing competency gap analysis based on individual existing and desired levels of skills and knowledge (Schmidt, 2004; Shen & Shen, 2004; Woelk, 2002). The difference in each learners experience level, education level, and requirements ensures no single learning object its all. A number of appropriate
learning objects that help learners to acquire the personal desired level of knowledge and/or skills are suggested. In addition, they are presented in an optimised order considering learners background and requirements. Given a learning outcome with the desired competency requirement, such an order may be generated before presenting it to learners or on the ly, depending on learners performance during the course of reaching the target learning outcome (Karampiperis & Sampson, 2004). An example of personalization of learning approach is the system proposed by Schmidt (2004). He proposed a learning environment in corporate setting where modularised small learning objects, the work process, company characteristics, and individual proiles were undertaken to improve learning eficiency and enable self-paced learning. His proposed system develops an agent that recommends learning materials that enable workers to learn at individual pace and meet the demand of knowledge in corporate context. It performs skill gap analysis of learners, and if there is a discrepancy between the current and required competencies, it retrieves appropriate learning objects that help learners to bridge the gap. To enable this matching service, the author represents the roles and tasks of a company as learners current and desired competencies, and learning objects pre-requisites and objectives. If there is a match of a learning objects pre-requisites and objectives to learners current and desired ones respectively, the learning object is recommended. More complex learning programs are compiled based on the dependencies between the roles and tasks of the company. The evaluation of the system showed that it not only helped to establish quality-controlled training processes with short eficient courses, but it also increased the chance of reusing learning material by modularisation. While Schmidt (2004) focuses on the matching service, Shen and Shen (2004) concentrate on how to recommend learning objects in an appropriate sequence. They are motivated by the fact that most e-learning contents are linearly organised with pre-deined sequence (Shen & Shen, 2004). A system is proposed to lexibly organise learning contents in small chunks to meet individuals needs. In order to recommend appropriate content to learners, learning objects and learners competency proiles are linked to one or more concepts in a hierarchical knowledge base such as datalink layer and network layer under computer network. A number of sequencing rules are formed deining when learning objects are selected in the light of information from learners competency proile. Each concept in a knowledge base is a prospective target competency that may become the learning outcome in a learners competency proile. If a target competency has pre-required competencies that learners have not acquired yet, the learning objects tagged with the prerequired competencies are suggested before those tagged with the objective competency. The pre-required competencies become the intermediate objective. If the target competency or intermediate objective is a parent with child competencies in the tree-like structured knowledge base, the learning objects tagged with the child competencies are recommended sequentially from left to right as their complexity increases. The authors reported the feedback from the survey of around 200 students who had used the system; more than 70% of them gave positive remarks. Some of them found it was more eficient to ind learning contents and spend more time on learning.
188 Ng & Hatala
The higher the level of personalization using learning objects, the greater the cost that can be saved in meeting a growing demand for access to education by discovering and reusing learning objects to create different learning experience. Given a personal learning outcome, an intelligent learning management is able to offer customized learning experience according to unique learning outcome.
Challenges
Despite the promising results of reusing learning objects during personalization of learning experience according to competencies in previous application examples, there are a number of challenges in connecting and discovering learning objects for competencies.
Search Learning Objects by Competency

To accomplish personalization by competency requirements of a learning management system, it requires inding learning objects by competencies. Competencies are the abilities or characteristics that are necessary to perform a task successfully. Competencies of learning objects are one or more intended outcomes of pursuing the learning objects. Information about a learning object is typically separated from its contents and stored in metadata. Metadata enables users to locate items very quickly without investigating all individual items, similar to the label on soup can that lets people know its ingredients, name of soup, etc., without actually opening each can (Wiley, 2002). A widely adopted metadata schema by learning object repositories is the one proposed by Institute of Electrical and Electronics Engineer Learning Object Metadata (IEEE LOM) (Learning Technology Standards Committee of the IEEE, 2002). The schema are commonly used in the metadata in learning object repositories (9 out of 10) in the survey performed by Neven and Duval (2002). LOM speciications do not guarantee the quality of the metadata entered by the cataloguer of learning objects. Metadata records created by Merlot (http://www.merlot.org) were reported in a survey as fragmentary and unstructured (Snchez-Alonso & Sicilia, 2005). The actual use of metadata schema relected in the empirical testing on the learning object repository Ariadne (http://www.ariadne-eu.org/) (Najjar, Ternier, & Duval, 2003) showed that most elements were either never or rarely used by catalogers. The only mostly used data element was of a granular nature, which has ixed vocabulary values such as raw media, chapter, lesson, etc., which are obviously not learning outcome. The descriptive data element, for example, learning objectives or competency, was not provided in 88% of 3,700 records under analysis. The existing number of learning objects annotated with learning outcome is believed to be inadequate for a learning management system to support any learning experience depending on learning outcomes. Therefore, a learning management system not only requires creation of a set of competencies for learning outcomes of learning objects but also associates a large pool of learning objects to appropriate competencies.
Quality of Competency Deinitions

How is a competency deinition expressed in learning outcome of a learning object in such a way that learners know it is what they desired or have achieved? Mager (1984) argued that learning outcome must be speciic and clearly described. If learning outcome is not clear, there is no sound basis for selection learning resources. If one does not know the destination, it is not possible to ind the way to get there. If learners know the objective they are going to accomplish, they are able to decide on their part what kind of activity or materials help them to reach their objective and arrange their efforts to accomplish it. Without sharply deined learning outcomes, it is not possible to know if it is in fact accomplished. A useful learning outcome should have the following characteristics (Mager, 1984): 1. 2. 3. 4. the behavior the learners are expected to do the condition under which the performance of behavior is to occur the target audience, such as managers who need supervisory skills the level of performance that is considered acceptable
There is no reason why learning outcome cannot be a single sentence. However, for learning outcome useful in resources selection, it requires high precision and quality of expression as shown in the following examples: Be able to write a musical composition with a single tonal base within four hours. The composition must be at least sixteen bars long and must contain at least twenty-four notes. You must apply at least three rules of good composition in the development of your score. (Mager, 1984, p. 102)
Relationships between Competencies

Competency deinition alone does not help the learners to decide which one they should reach before the others. When it comes to suggesting a sequence of learning objects for competency requirements, the relationship among competencies may change the immediate competency objectives and thus the learning object selection. An objective is called the terminal objective when it is broken down into several enabling objectives that enable the reaching of the terminal one. Each enabling objective is assigned along with one or more learning objects that teach the objective (LAllier, 1997). Learners are recommended with the learning objects in each of the enabling objectives of the terminal one. The pre-required competencies of a target competency prevent learners from reaching the target one before satisfying the pre-requisites. Competencies are also of a diverse nature that may range from as common as acquisition of knowledge of certain subjects to complex development of mental
190 Ng & Hatala
structure (Snchez-Alonso & Sicilia, 2005). Various competencies can also be categorized into numerous types such as roles and jobs. More complex personalization can be achieved by recommending a set of learning resources for a particular type of competency. Such relationships among competencies certainly requires a representation model for machine processing during automatic course sequencing.
Consistency of Competency Deinitions

One or more learning objects designed for a competency are possibly annotated by different learning object cataloguers. If those learning objects are indexed with different expressions of the given competency, it prevents consistent performance of automatic learning object discovery, as todays technology offers limited natural language processing ability. Therefore, consistent usage of vocabulary in describing competency as learning outcome for learning object is required for effective computerized processing in personalization of learning experience. Although personalization of learning experience is pervasive, using current learning object metadata records without solving the above challenges hinders the opportunity of a higher level of automatic personalization.
States of Art
Some researchers implemented solutions to meet the above challenges. Some look for alternative sources of learning outcomes other than in the metadata and link them up with learning object manually (Shen & Shen, 2004) or automatically (LAllier, 1997; Woelk, 2002). The relationships among competencies are represented in ontology (Schmidt, 2004; Shen & Shen, 2004; Woelk, 2002). There have been some attempts to standardize the data model of competency to encourage reuse.
Location of Learning Outcome

A learning management system suggests appropriate learning objects based on individuals unique competency requirements. To enable it to do this, learning objects must be annotated with the competencies that they are designed for. The annotation may be part of metadata or external to metadata. Snchez-Alonso (Snchez-Alonso & Sicilia, 2005) and Mwanza (Mwanza & Engestrom, 2005) suggest that it should be included as an distinctive element in the educational category of LOM speciications. Shen and Shen (2004) implement learning objects as instances of the competency objectives that they belong to. LAllier (1997) does not use the LOM but composes a learning object as a structural component with objective
Figure 1. Concept tree adopted from Shen and Shen (2004)

Network Layer
IP Protocol
Routing Algorithm
Address Resolultion
ICMP
Concept Detail: Prerequisites: DataLink Layer, IPProtocol Part-of: Network Layer Previous: Routing Algorithm Subsequent: ICMP
as one of the component elements. Different ways of annotation of competencies of learning objects give rise to an interoperability problem that limits automatic discovery of learning objects by learning outcome.
Sources of Learning Outcome and Relationship Representation

One of the sources of learning outcomes is internal information of an enterprise. Concepts, relationships, and processes were used in Woelk (2002) while roles, tasks, knowledge area, and processes were used in Schmidt (2004). They did not explicitly show actual representations of the information, but it is represented in competency ontology. Ontology is a speciication of conceptualization with relationships, attributes, and property that helps to share information. Corporate information may include how to perform a task, regulations and rules, subject knowledge, and facts. Roles in a typical company like bank are teller, account manager, and credit oficer. The other source of learning outcomes is concepts in a domain. Shen and Shen (2004) deined the content of a knowledge base to which learning objects and learners objectives are linked. The knowledge base is composed of interrelated concepts forming a concept tree or ontology. An example of such a knowledge base represented in a concept tree adapted from Shen and Shen (2004) is shown in Figure 1. It is built based on the concepts used in a course related to computer networks. Learning objects are instances of concepts that are also their objectives. Learners objectives are also one or more of those concepts in the tree. Concepts may have sub-concepts implying a part-of relationship, sequencing information, and prerequisites. Sequence of learning outcomes one may
192 Ng & Hatala
take is determined by factors including whether the prerequisites of target learning outcome are satisied and whether the target learning outcome has children objectives. The relationship among concepts in the concept tree or ontology enables the recommendation of an ordering of learning outcomes that helps learners to achieve their target objectives. However, some authors (Schmidt, 2004; Shen & Shen, 2004; Woelk, 2002) did not seem to pay much attention to the quality of learning outcome representation. Learning outcome described as Internet control message protocol ICMP is not able to provide enough information about what learners are actually going to learn and how to tell if learners have achieved this objective. Consequently, learners are not able to clearly know if this is the objective they need and how to achieve this objective. A search on a university library (www. lib.sfu.ca) gives 16 results (as on June 11, 2005) for keyword search on ICMP. The results cover issues ranging from network security assessment using ICMP probing and programming tips on capturing ICMP message to ICMP as transmission control protocol/Internet protocol (TCP/IP). As ambiguous learning outcome does not provide sound basis for learning resources selection (Mager, 1984), all 16 results are candidates for satisfying ICMP objective as they all contain the necessary keyword. Hence, it is not clear if learners need to learn one or all of them in order to have ICMP objective satisied. However, effective learning outcome as stated previously is often composed of one or more sentences that may be structured in a way that promotes the usefulness of a learning outcome.
Pool of Learning Objects with Learning Outcomes

The annotation of learning objects with learning outcome is established manually (Schmidt, 2004; Shen & Shen, 2004) or automatically (Bejcek & Pezzuti, 2001; Woelk, 2002). Woelk did not describe how to connect them automatically while Bejcek (Bejcek & Pezzuti, 2001) applies patented matching and iltering algorithm using natural language algorithms in connecting training interventions (such as learning objects) and skill objects (such as skills and tasks). No information about the number of annotated learning objects and learning outcomes created is found in Schmidt (2004) or Shen and Shen (2004), but there are not expected to be many as annotation is done manually and learning outcomes are limited to the ones used by a course in a college (Shen & Shen, 2004) and two companies (Schmidt, 2004). The process of making appropriate annotation together with the creation of learning outcomes requires time and efforts of domain experts or machine processing. As both (Schmidt, 2004; Woelk, 2002) use information in corporate context, efforts of annotation and creation of learning outcomes are possibly duplicated. Duplicate efforts would otherwise allow more learning objects to be retrieved by learning outcomes. The opportunity of offering learning objects based on personal learning outcomes is not fully exploited as duplicate and limited efforts in making learning outcome-annotated learning objects available. In addition, automatic personalization by learning outcome is restricted only to those objectives that have linked to learning objects. However, learning management systems can be more supportive for individuals who are looking for learning objects by learning the outcome if they are allowed to discover learning objects themselves with public-accessible and welldeined learning outcomes. Therefore, a centralized location for easy access to learning
objects and learning outcome is required for maximizing opportunities of personalization by learning outcomes.
Consistency of Competency Deinition

The consistency of vocabulary of competency deinition is higher when the deinitions can be re-used over and over again to describe the learning outcome. Different description of a competency deinition deters the level of automatic discovery of learning objects because of the limited natural language processing of machines. Machines are not always able to tell if a teller of a bank is the same as customer representative in another bank. In this regard, Shen and Shen use single location in referencing competency objectives in their experiment so that objectives are described consistently (Shen & Shen, 2004). Snchez-Alonso (Snchez-Alonso & Sicilia, 2005) proposed a normative approach in describing learning outcome for automated processing by describing expected outcome using recognizable syntax. An example of the syntax given by the authors is lrn.knows(genitiveCaseEnglish)[80]. As it is not crystal clear what lrn and genitiveCaseEnglish mean, syntax description requires shared common understanding among the stakeholders of learning objects to enhance the consistency and reusability of learning outcome description. The higher the level of reusability and interoperability of learning outcome, the greater the number of learning objects with learning outcome with consistent vocabulary. The reusability issue fosters IEEE LTSC, which initiates a standard for deining reusable competency deinition (RDCEO) (Learning Technology Standards Committee of the IEEE, 2004) to facilitate reuse, exchange, and reference purposes. The deinition of competency covers general sense, including skills, knowledge, tasks, and learning outcomes. It is aimed to create a common deinition with common meaning by providing a data model for reference, exchange, and reuse of competency deinition in one or more learning systems. RDCEO contains the following elements with high level of extensibility: 1. 2. 3. 4. Identiier: a global unique identiier with catalog and entry elements Title: short but mandatory human-readable text label for the deinition Description: optional unstructured text of deinition description Deinition: optional structured text for more complete description with extensible elements of: i. ii. 5. Model source: the model upon which the deinition is based. Statement: single characteristic of deinition with Statement ID, Statement Text, Statement Name, and Statement Token.
Metadata: optional records of metadata that further describe RDCEO.
The standard should be able to help to improve understanding and make it easier to deine learning outcome clearly by improving expressiveness of learning outcome deinition. Therefore, it is a good candidate data model schema for learning outcome of learning objects.
194 Ng & Hatala
While this schema is now being developed and has not been widely used, the actual use of it may be subject to the drawbacks as in using LOM schema. The current LOM schema provides room for ambiguity and lack of precision due to inconsistent usage (Snchez-Alonso & Sicilia, 2005). The survey of metadata of the Ariadne (http://www.ariadne-eu.org/, Najjar et al., 2003) showed that there is a lack of precision as cataloguers chose different data elements during indexing learning objects and also different vocabulary values assigned to a data element according to personal template elements and values. The lack of quality control in the metadata makes most current records unusable for automation in learning object selection (Snchez-Alonso & Sicilia, 2005). Even if learning outcome uses RDCEO schema, the level of automation of learning object selection for personalization is again restricted by the unsolved ambiguity problem. Hence, the development of a solution is required that improves the understanding, reusability, accessibility, and interoperability of learning outcomes of learning objects so as to allow a higher level of personalization by learning outcome of learning.
Ontology and Competencies

Ontology has been used in competency representation. However, for learning objects being used effectively in both machine processing and manual selection, the following are the characteristics of ontological representation of competencies.
Consistent Competency Descriptions in Learning Object

One of the ways to minimize duplicate efforts in creating competencies is to describe competencies of learning objects consistently. Descriptions of competencies vary as the same competency can be represented using different vocabularies. These variations can prevent automatic learning object discovery because machine processing is not always able to tell that they describe the same competency. Clearly, it is a huge task to standardize all vocabulary used to describe competencies. Consistent descriptions require a shared common understanding of the meaning of competencies. The community that needs to reach the shared common understanding is not limited to educators but also to biologists who want to communicate with genes found in an organism as well as its attributes (http://www.geneontology.org/) and accountants who want to support inter-organizational use and statutory reporting requirement (Geerts & McCarthy, 1999). Ontology is in the right position to address this issue. They try to reach consensus and understanding by making the concepts or objects that are of interest to their communities explicitly speciied with necessary attributes and relationships in ontology open to public access and reference. Ontology captures the conceptualisation and clariies the structure of the potential terms used to describe and represent an area of knowledge (Chandrasekaran, Josephson, & Benjamins, 1999; W3C, 2003). Ontologies are used by people, databases, and applications that need to share domain information. A domain is a speciic subject area or area of knowledge, like medicine, tool manufacturing, real estate, automobile repair, inancial management, and so forth. Ontology encodes knowledge in a
domain and also knowledge that spans across multiple domains. Public ontology allows different data sources to commit to the same ontology for shared meaning (Chandrasekaran et al., 1999), making knowledge reusable and sharable. A competency ontology publicly accessible by learning object cataloguers not only minimizes duplicate efforts in creating and describing same competency in different learning management systems but also encourages their reuse in the ontology in annotating learning objects.
Annotating Learning Objects with Competencies

It is an enormous task for learning object cataloguers to provide consistent and complete information about learning outcomes as well as their relationship and attributes in metadata, considering there are more than an estimated 15 terabytes (Wiley, 2002) of information available on the publicly accessible Internet. The dificulty of this annotation task limits the number of learning objects annotated with competencies. As ontological representations of learning outcome become ready for object-oriented paradigm, instructional designers develop learning objects with an instance of competencies deined in their domain ontology, and all a cataloguer needs to do is to include the reference number of the instances of the competencies as the learning outcomes in the metadata of learning objects. Once learning objects developed for the same competency are annotated with the same reference to an instance, consensual competency deinitions of learning objects make automatic learning object discovery by learning outcome a reality.
Consistent Concepts by Extending Ontology

Ontology enhances a common understanding of competencies among learners, instructional designers, and learning object cataloguers not only by its ability to express controlled deinitions of competencies but also by its ability to expand to other ontologies that offer the same beneits. Gagn (Gagn, Briggs, & Wager, 1992) suggests that learned capability verbs used in conjunction with concrete concepts and deined concepts are one of the ive components that enable instructional designers to write unambiguous objectives. Concepts are possible sources of lack of clarity and ambiguity widely varied in terminology commonly used at a given time (http://www.geneontology.org/GO.doc.shtml#control). The clarity of meaning of concepts can be improved by referencing a concept in an existing ontology such as gene ontology (http://www.geneontology.org/index.shtml), which provides controlled vocabularies in classes and attributes.
Help to Retrieve Learning Objects

Intelligent learning management systems can provide learners with recommendations of learning objects according to personal learning outcomes. As individuals have unique personal learning outcomes, the number of people who are able to beneit from custom-
196 Ng & Hatala
ized learning by competencies is limited unless a large number of learning objects are annotated with competencies. However, publicly accessible competency ontology allows for searching of learning objects based on the description of desired competencies available in the ontology. Nevertheless, competency deined as ICMP or routing algorithm as in the experiment of Shen and Shen (2004), is an effective learning outcomes according to Mager (1984). According to Mager (1984), three characteristics help make an objective communicate an intent: 1. 2. 3. What should the learner be able to do? Under what conditions do you want the learner to be able to do it? How well must it be done?
When it comes to encouraging the above characteristics clearly described in learning outcome descriptions in the ontology, the ontology allows description in classes, relations, and properties. Classes are general things in the domain of interests such as a learning outcome. Relations are the relationships among classes, for instance, prerequisites of a learning outcome. Properties are the attributes that characterise a class. The above characteristics are represented in the following properties in learning outcome class: 1. 2. Description for the condition and what learners should be able to do; Performance indicator for How well must it be done?
If a learning outcome is deined with the above properties, it is expected that a performance indicator as well as an externally referenced concept in the previous section enable it to retrieve the relevant learning objects that cannot be found with title and description only. On the other hand, relations among learning outcomes are also expected to help retrieve learning objects for a target learning outcome. For instance, if a target learning outcome ICMP is part-of the network layer, some learning objects related network layer also discuss ICMP
Public Accessible Ontology and Quality Assurance

One of the beneits of using publicly accessible ontological representations of competencies is that it helps to improve the quality of the competency deinitions. A suggestion and review system complements the creation of controlled competency deinitions. If learners do not know if they understand the principle of relativity being a competency deined in the commonly accessible ontology, they may suggest or ask for review and editing of it such that the ambiguous objective is open to fewer interpretations and becomes clear enough to tell, for instance, the situation under which the behaviour of the objective is to be performed and the level of cognitive behaviour. It allows addition, clariication, and high precision of deinition of competencies as a result of peer scrutiny. A similar review system has been
Table 1. Explanation of Elements in the Data Model of Competency Ontology
Name Competency Title Description Identiier Deinition Performance Indicator Proiciency level Knowledge reference
Class / Property Class Property Property Class Class Class Class
Explanation Describes the nature of the competency as well as relationships with other competencies. Relates the title of a class in a short sentence. Relates the details of explanation. Describes a globally unique identiier. Describes more complete description of observable behaviors in different proiciency levels. Describes observable behaviors of satisfying a competency at a proiciency level. Describes the degree of expertise level in fulilling the competency. For instance, novice and expert. Relates the concepts or facts that are crucial in demonstrating fulillment of competency. It may be a description of concept or a reference to the identiier of a concept in another ontology. Relates suggested next objective after inishing a competency. Relates suggested previous competency before working on a competency. Describes the prerequisite competencies of a competency. Relates competency (A) to others (B and C) if A is broken down into enabling objectives B and C.
Property
Next objective Previous objective Pre-required objective hasParts
Property Property Property Property
Table 2. Data Model of Competency Ontology to be represented in OWL Syntax
198 Ng & Hatala
Class Competency
Property of Class Identiier Title Description Deinition Knowledge reference Next objective Previous objective Pre-required objective hasParts
Domain Competency Competency Competency Competency Competency Competency Competency Competency Competency Deinition Deinition Deinition Performance indicator Performance indicator Performance indicator Proiciency level Proiciency level Proiciency level
Range String String String Instance of deinition String Instance of competency Instance of competency Instance of competency Instance of competency String String Performance indicator String String Proiciency Level String String String
Min Cardinality 1 1 1 1 0 1 1 1 0 1 1 1 1 1 0 1 1 1
Max Cardinality 1 1 1 1 -
Deinition
Identiier Description Performance indicator
Performance indicator
Identiier Description Proiciency level
Proiciency level
Identiier Title Description
used by gene ontology (http://www.geneontology.org/) handling new terms suggestions and requests of changes. Constructing ontology of a wealth of well-deined controlled vocabulary is an ongoing process that requires signiicant efforts and commitment from people of a community. Therefore, efforts at developing one are devoted to those that are of interest and importance to the community, rather than everything in the universe. An ontological representation of competencies attracts the attention of all the stakeholders developing and using learning objects and encourages them to think of competencies as learning outcomes as well as their importance during designing, annotating, and using learning objects.
Complementary to HR-XML
HR-XML (HR-XML Consortium, 2004) is a reusable XML schema that allows exchange of information about competencies within business context. It could be potentially useful in recording, comparison, matching, and rating of competencies in human resources processes. An ontological representation of competencies with detailed description, deinition, and relationships among others together with HR-XML forms a complete package for competency management.
Ontological Representation and Recommendation

A data representation model for competencies captures the characteristics of the nature and relationships of competencies that encourage creation of competencies as effective learning outcomes (title, description, knowledge reference, proiciency level, and performance indicator) and establishment of rules for sequencing (pre-required objective, next objective, previous objective). It also provides the ability for extension to other ontology and referencing by other data models (knowledge reference). Table 1 explains all the elements in the data model to be represented either in class or property in Web Ontology Language (OWL) syntax. Table 2 shows the properties of classes and cardinality of properties. Property identiier is for both minimum and maximum cardinality to one, as it is a mandatory item uniquely identifying an individual that is not allowed to have more than one identiier. Some properties have one minimum cardinality as they are regarded as the necessary opponents in promoting effectiveness of learning outcome. The details of classiication taxonomy are extracted from Anderson, Krathwohl, and Bloom (2001). When it comes to recommending learning objects according to competency deined in the ontology, competency deinitions may be available for selection or they may be looked up from text search. Once one or more competencies are selected, learning objects are recommended according to the selected competencies. Recommendation based on free text search: The text in search determines the competencies that will be selected for recommendation. Once competencies are selected, learning objects are recommended based on competency selection in the
200 Ng & Hatala
Table 3. Application example of data model of learning outcome

Object Name Learning outcome Object Attribute Identiier Title Description Deinition Next objective Pre-required objective Cognitive state Classiication hasParts Previous objective Knowledge reference Deinition Identiier Performance indicator Performance indicator (multiple cardinality) Performance indicator Identiier Proiciency level Description 1. Attribute details CA-CompMap-V-1.3 Develops or evaluates systems for performance measurement Designs and develops or evaluates a chart of accounts to suit inancial accounting and reporting needs DeinitionExample1, an instance of Deinition Object CA-CompMap-V-1.4 Class evaluate and class procedural knowledge CA-CompMap-V-1.2 Capital budgeting DeinitionExample1 PI-1, an instance of performance indicator PI-2, an instance of performance indicator First instance PI-1 PL-1 Users of inancial data conirm that the journals and ledgers, or equivalent computerized enquiry capabilities, meet their decision-making needs; Financial data are complete and accurate; Reporting requirements and stakeholder needs are satisied. Second instance Identiier Proiciency level Description PI-2 PL-2 Description/explanation of the scope and process in developing information systems to capture measurement information relects an understanding of the subject matter and is appropriate in the circumstances First instance Identiier Description Second instance Identiier Description PL-2 Comprehend PL-1 Perform
2. 3.
Proiciency level
Figure 2. Extracted knowledge reference
following. Each word in the text in search is matched against the title, description, description in deinition, knowledge reference, and description in performance indicator. A competency is selected when there is a match. Recommendation based on competency selection: Learning objects are recommended in the following order: 1. 2. 3. Learning objects annotated with the selected competency. Learning objects annotated with the knowledge reference of the selected competency. Learning objects annotated with the competencies if any in the hasParts property of the selected competency.
202 Ng & Hatala
Figure 3. Average Percentage of Relevant Results from different Sources of Keywords

Title / Description 31%
Competency Evidence 23%
Knowledge Reference 23%
Parent / Sibling 23%
4. 5. 6. 7.
Learning objects annotated with the knowledge reference of the competencies in the hasParts property of the selected competency. Learning objects annotated with the previous objective of the competencies in 4. Learning objects annotated with the knowledge reference of the competencies in 5. Learning objects annotated with the next objective of the competencies in 4.
An Application Example
To illustrate this competency formalization approach, Table 3 describes an example of competency instance extracted from CA candidates competency map (The Canadian Institute of Chartered Accountants, 2004) taken from Chartered Accountant of Canada1, which provides a list of competencies expected from a prospective chartered accountant. The ontology representation of knowledge reference list in the competency map is implemented in Protg 2000 in Figure 2. A learning object is annotated with one or more instances of competencies in terms of expected outcomes and prerequisites, which are instances of competencies stated under the educational category of IEEE LOM as suggested by Mwanza and Engestrom (Mwanza & Engestrom, 2005) based on the theory of expansive learning (Mwanza & Engestrom, 2005).
Evaluation
Currently most learning objects are not annotated with competencies as learning outcomes, and it takes time to develop competencies and annotate learning objects with them. However, before a large pool of annotated learning objects becomes reality, competencies deined in ontology with elements as speciied in a previous section are good sources of keywords for inding learning objects by competencies. An evaluation with the aim of revealing the difference in contributions of relevant learning objects from title/description, knowledge reference, related competencies, and performance indicators was performed. Sixty queries (15 competencies x 4) for 15 competencies selected from CA Competency Map were sent to the Elton B. Stephens Company (EBSCO) host with Business Source Premier, Academic Search Elite, AgeLine, PsycINFO, EBSCO Animal. Figure 2 shows that although title/description accounted for the largest portion (31%) of relevant results, others contributed around a quarter of relevant results (23%). Therefore, we concluded that when competencies are deined not only with title and description but also with knowledge reference, related competencies, and performance indicators, they help to ind relevant learning objects by competencies in keywords search.
Conclusion and Future Work

The dream to deliver personalized learning according to competency requirements using learning objects that it unique individuals requirements is about to become a reality. Today, along with many important developments in instructional design that help to dynamically design and develop instruction and understand how individuals want and intend to learn in whole-person perspective, the foundations of truly personalized online learning are laid down. Personalized learning is important because it supports the increasing demand for education by dynamically adapting learning resources to it unique requirements of learners with different competencies, age, educational background, and personal style. Among a variety of ways of personalizing learning, recommending learning resources according to unique personal competencies offers a customized learning experience. Given a set competency requirements, sequenced or non-sequenced learning resources are recommended to learners to meet the requirements given their current satisied learning outcomes, language platform, or learning style. Not all learners have to follow the same sequence of learning resources in order to reach the target learning outcome. Conditional rules specify the condition under which learners can skip or change to learn some resources because they have or have not completed a role or prerequisite learning outcome. The notion of a small unit of instructiona learning objectenables sharing resources from a large repository of content individually, widely, and more economically. Flexibly arranging and using learning objects as the building blocks of instruction certainly aligns with the objective of personalization. Learning objects are expanding their role in supporting customized learning, as there is a gradual realization of beneits of instructional factors included in the metadata. However, the opportunity of
204 Ng & Hatala
personalization by competencies is not fully exploited due to limited accessibility of learning objects according to competencies. Currently, there is limited with possibly duplicate efforts in creating and annotating competencies for learning objects. Automatic discovery of learning objects by competencies is hindered by the descriptive nature of competencies. Descriptions of a learning outcome are not always exactly the same in metadata in different learning objects. Existing machines are not always able to identify inconsistent descriptions for the same learning outcome. Syntax description of learning outcome and IEEE RDCEO competency deinition schema do not completely solve the problem of inconsistency of description of learning outcome because common understanding of the intended meaning of vocabularies and commitment to using them are required. The ontological model proposed in this chapter is one of the ways of accommodating competencies in personalization in such a way that the level of accessibility of learning objects itting individual competency requirements is improved. It is because controlled competency descriptions represented in shared ontology enables the delivery of intended meanings of the description, thus creating common understanding among the stakeholders using competencies. Consistency of competency descriptions is increased by reuse and reference to the objective deined in a competency ontology during deining competencies as personal learning outcomes, annotating learning objects, and designing and assembling learning objects. The descriptions of learning outcomes in the metadata of learning objects can be as simple as reference numbers to the competency instances deined in a shared ontology. Including reference numbers as competencies not only allows the cataloguer to annotate more learning objects without repeating competencies deined in ontology but also increases consistency in descriptions of competencies. Accessibility of learning objects by competencies is thus improved by consistent performance in automatic discovery of learning objects.
References
Anderson, L. W., Krathwohl, D. R., & Bloom, B. S. (2001). A taxonomy for learning, teaching, and assessing: A revision of Blooms taxonomy of educational objectives. New York: Longman. Bejcek, A., & Pezzuti, A. (2001). Precision skilling. Retrieved June 16, 2005, from http:// www.netg.com/DemoDownloads/Downloads/skilling.pdf Blackmon, W., & Rehak, D. (2003). Customized learning: A Web services approach. In Proceedings of World conference on educational multimedia, hypermedia and telecommunications (Vol. 1, pp. 6-9). Downloaded from http://www.lsal.cmu.edu/lsal/expertise/papers/conference/edmedia2003/customized20030625.pdf Chandrasekaran, B., Josephson, J. R., & Benjamins, V. R. (1999). What are ontologies, and why do we need them? Intelligent Systems and their Applications, 14(1), 20-26. Dubois, D., & Rothwell, W. (2004). Competency-based or a traditional approach to training? T + D, 58(4), 46.
Gagn, R. M., Briggs, L. J., & Wager, W. W. (1992). Principles of instructional design (4th ed.). Fort Worth: Harcourt Brace Jovanovich College Publishers. Geerts, G. L., & McCarthy, W. E. (1999). An accounting object infrastructure for knowledgebased enterprise models. IEEE Intelligent Systems, 14(4), 89-94. HR-XML Consortium. (2004). Competencies (measurable characteristics) recommendation. Karampiperis, P., & Sampson, D. (2004). Adaptive instructional planning using ontologies. Proceedings of the IEEE international conference on advanced learning technologies (ICALT04) (pp. 126-130). LAllier, J. J. (1997). Frame of reference: NETgs map to its products, their structures and core beliefs. Retrieved June 16, 2005, from http://www.netg.com/DemoDownloads/ Downloads/Frame_Reference.pdf Learning Technology Standards Committee of the IEEE. (2004). Draft standard for learning technology: Reusable competency deinitions Learning Technology Standards Committee of the IEEE. (2002). Draft standard for learning object metadata (No. IEEE 1484.12.1-2002) Mager, R. F. (1984). Preparing instructional objectives (rev. 2nd ed.). Belmont, CA: Lake Publishing. Martinez, M. (2002). Designing learning objects to personalize learning. In D. A. Wiley (Ed.), The instructional use of learning objects (p. 152). Bloomington, IN: Agency for Instructional Technology. Mwanza, D., & Engestrom, Y. (2005). Managing content in e-learning environments. British Journal of Educational Technology, 36(3), 453-453. Najjar, J., Ternier, S., & Duval, E. (2003). The actual use of metadata in ariadne: An empirical analysis. Proceedings of 3rd annual ariadne conference, Katholieke Universiteit Leuven, Belgium. Retrieved from http://www.cs.kuleuven.ac.be/~stefaan/papers/ActualUseOfMetadata.pdf Neven, F., & Duval, E. (2002). Reusable learning objects: A survey of LOM-based repositories.In Proceedings of the tenth ACM international conference on multimedia (.MULTIMEDIA02) (pp. 291-294). Retrieved from http://doi.acm.org.proxy.lib.sfu. ca/10.1145/641007.641067 Sanchez, R., & Heene, A. (2000). Theory development for competence-based management. Stanford, CT: JAI Press. Snchez-Alonso, S., & Sicilia, M. (2005, May). Normative speciications of learning objects and learning processes: Towards higher levels of automation in standardized e-learning. International Journal of Instructional Technology and Distance Learning, 36(3), 453-463, Schmidt, A. (2004). Context-steered learning: The learning in process approach. In Proceedings of IEEE International Conference on Advanced Learning Technologies, 2004 (pp. 684-686).
206 Ng & Hatala
Shen, L., & Shen, R. (2004). Learning content recommendation service based-on simple sequencing speciication. In Proceedings of Advances in Web-Based Learning: ICWL 2004: Third International Conference (pp. 363-370). The Canadian Institute of Chartered Accountants. (2004). The CA candidates competency map: Understanding the professional competencies of CAs. W3C. (2003). Web ontology language (OWL) use cases and requirements. Retrieved June 13, 2005 from http://www.w3.org/TR/2003/WD-webont-req-20030203/ Wiley, D. A. (2002). Connecting learning objects to instructional design theory: A deinition, a metaphor, and a taxonomy. In D. A. Wiley (Ed.), The instructional use of learning objects (p. 3). Bloomington, IN: Agency for Instructional Technology. Woelk, D. (2002). E-learning: Semantic Web services and competency ontologies. World conference on educational multimedia, hypermedia and telecommunications, 2002(1), 2077-2078. Retrieved from http://dl.aace.org/10523
Linking E-Government with Organizational Memory 207
Chapter IX
Linking E-Government with Organizational Memory through Individual Competencies

Juan G. Cegarra-Navarro, Universidad Politcnica de Cartagena, Spain
Abstract
Organizational memory (OM) is a concept that has been used to refer to stocks of data, information, and knowledge to which every member of the organization has access. In the following chapter we distinguish between what we refer to as hard OM and soft OM. The term hard OM is used to refer to a variety of technologies that support the storage of, access to, and modiication of OM. Soft OM refers to policies, rules, reporting structures, and practices that encourage the active use and updating of OM in addressing current organizational concerns. This chapter considers the role of OM (hard and soft) in relation to the adoption and use of e-government systems. We argue that for e-government to be successfully adopted by companies, some competences to manage and exploit knowledge at the organizational level need to be developed and assessed. In doing so, the research supports the view that the identiication of problems, the changing of cognitive patterns and the integration of new measures in the organization mediate the relationship between the OM and e-government.
208 Cegarra-Navarro
Introduction
The organizational memory (hereafter OM) literature provides many varying and, occasionally, competing deinitions, but the meaning of OM at the organizational level is the same. OMs are usually based on what has worked in the past, that is, what has proven to be successful (Bent, Van der Paauwe, & Williams, 1999). OM is comprised of a stock of data, information, and knowledge (memories)every member of the organization has access to it, has the possibility of interpreting it, and has the possibility of acting upon it. OM includes not only hard information, for example, numbers, facts, words, and igures, but also soft-information, that is, individual or social information with meaning, for example, expertise, experiences, anecdotes, critical incidents, stories, and details about strategic and operative decisions. E-government can be deined as a set of activities supported by information systems with the aim to improve the relationships between government institutions and citizens (Heichlinger, 2004). The aim of e-government is to enhance public participation in decision-making. Varieties of e-government Web sites have been set up worldwide, providing services and information at different levels (local, regional, or national). Important users of e-government are not only citizens but also businesses. E-services include social contribution for employees, corporate tax, registration of a new company, submission of data to the statistical ofice, custom declaration, public procurement, and so forth. OM is especially important in e-services. Many existing sources of knowledge, laws, comments to laws, speciic regulations, old similar cases, available cases, documents and information, and so forth, are prevalent at different places and in different forms and representations at several degrees of formality, and are related by manifold links. The growth of this new electronic world requires innovation and the generation of new knowledge, with not only staff but also public servants listening to information in new and unexpected ways. This often means letting go of our existing knowledge and recognizing that they may prevent us from learning new things. For example, before new routines can be implemented (e.g., pay corporate tax through e-government Web sites), the old routines must be challenged, and this requires a willingness to unlearn (i.e., accept that knowledge and experience can and should be challenged; Sherwood, 2000). This is a challenging and painful endeavor. Therefore, although e-government has to be redesigned around OM to attain any beneit from the information provided by governments, managers must be aware that just as OM provides stability, it can also serve to block e-government implementation. Part of this dificult alignment involves critical self-relection about ones own work and changing roles (Spender, 1998). Competencies represent the knowledge and skills required for performing and supporting the business processes (Woodruffe, 1993) and the basis for creating value in an organization (Hoffmann, 1999), and they involve competence factors that are observable and measurable (Sicilia & Lytras, 2005). Considering this, we suggest that in order to implement e-government and transform such procedures and processes, those companies using and accessing government services need to acquire the necessary competencies about them. The focus of this chapter is how can OM enable e-government implementation in companies? In order to answer this question, we use competence factors to distinguish among the various levels of work within the OM. The chapter is organized as follows: the conceptual
framework of OM is introduced in the next section. The process of exploiting the knowledge stored in the OM and implement e-government is presented in the third section. The following section considers the key factors to implement e-government among companies. The conclusions of this chapter are presented in the inal section.
Background
Megill (1997) deines OM as all active and historical information about an organisation that is worth sharing, managing, and preserving for later reuse. OM is also called corporate knowledge by Hamel and Prahalad (1994, pp. 25). It refers to the repository where knowledge that has been learnt is stored for future use. Woolcock deines social capital as the norms and network which facilitate the collective actions for mutual beneit (1998, p.155). Selnes and Sallis (2003) suggest that relationship memory includes idiosyncratic routines in the form of encoded formal and informal procedures and scripts, including how the parties have learned to do things. The goal of OM is that all members of the organisation are aware of where the useful complementary abilities (e.g., who knows what? Who can help with that? Who can exploit new information?) reside. As we have discussed above, the data, information, and knowledge that constitute an organizations memory may be hard or soft. While the hard OM, or semantic memory, is comprised of general, explicit, and articulated knowledge (e.g., the organizational iles, documentary records, transactional records, or annual reports), the soft-OM, or episodic memory, is comprised of context-speciic and situated knowledge (e.g., context-speciic details of speciic organizational decisions). These memories are stored in different retention structures within the organization. The most obvious structures for encoding hard OM include information systems such as corporate manuals, databases, iling systems, and so forth. These systems are continually being updated and analysed and are thus capable of generating new streams of information, thereby expanding knowledge (Zuboff, 1988). Cegarra and Sabater (2005) classify information systems into three categories: (1) technologies that help users to retrieve information; (2) technologies that allow users to analyse information and thus create new information; and (3) technologies that enable users to apply this information in decision-making, thus potentially creating new knowledge. As shown in Figure 1, our hard structures have been concerned about those three possibilities. Internet technology (IT) enables public servants, customers, and employees to have instant access to information about products and services across time, zones, and distance. Electronic distribution technologies (EDTs) provide collaborative groups formed by employees, managers, and sometimes public servants with the ability to link large numbers of information units in a dynamic manner, and basic PC technologies (PCT) that facilitate that low of information are controlled by users. On the other hand, Walsh and Ungson (1991) suggest that soft OM is represented by many diverse aspects of an organization, for example, the organizations culture, transformations (production processes and work procedures), structure (formal organizational roles), ecology (physical work settings), and information archives (both internal and external to the organization). As shown in Figure 1, Soft-OM may be augmented with new information
210 Cegarra-Navarro
Figure 1. Organizational memories

HardHard -OM
Counters and trackers Web page or Homepage Site map, Search Engine Banners or links with other sites Bulletin Board Systems Open discussion forums Repository of documents Open voting systems Newsgroup (USENET) Open distribution lists Access to share data base Tools to recommendations Creation of billing systems Tools to provide estimated costs Tools to provide timeframes Shopping cart solutions Payment and verification
SoftSoft -OM
Culture of the organization Employee sentiment No groupthink
Common language Lateral communication Mutual adjustment
Knowledge filtered Individual autonomy Experimentation
or knowledge in a variety of ways. These include increasing awareness of the knowledge embedded in client relationships and engagements, which, if shared, could enhance organizational performance; focusing on knowledge-related employee behavior, with, for example, contributions to the organizations structured knowledge base attracting signiicant rewards and bonuses; and implementing the decision audit programs in order to assess whether and how employees were applying knowledge in key decisions.
Design and Use of E-Government Services via OM

External relationships are necessary to receive timely and pertinent information from the business environment (Hauser, 1998); this includes government institutions. For Unpa and Aspa (2001) e-government simply consists of the creation of a Web site where information about political and governmental issues is presented. Tat-Kei Ho (2002) asserts that through e-government, businesses can access to public information and services related to their speciic needs, simply by clicking on different Web links within the government Web site. They can also contact government oficials directly via e-mail or online request forms to obtain feedback on speciic issues. Abramson and Means (2001) deine e-government as digital governmental information, or a way of engaging in digital transactions with the public (citizens and businesses) and employees. Fraga (2002) suggests that e-government is the transformation of internal and external relationships in the public sector through net-enabled operations. Tapscott (1996) asserts that e-government is an internetworked government that links new technologies with legal systems internally and, in turn, links such government
information infrastructure externally with everything that is digital and with everybody tax payers, suppliers, business customers, voters, and every other institution in the society. The above considerations lead us to suggest that the use of e-government should take the form of cost reduction more eficiently, carefully, and responsibly, and that more adequate and timely ways of communication between users (e.g., companies) and government institutions should be provided. According to Cegarra, Dewhurst, and Briones (2005), there are three interaction levels that characterize an e-government framework: Level (1), or external relationship: At this level, e-government opens up new possibilities for governments to be more transparent to businesses, giving access to a greater range of information collected and generated by governments. For instance, visiting government Web sites enables businesses to have instant access to information about services across time, zones, and distance. Level (2), or relational relationship: E-government adoption at this level may enable fundamental changes in the relationships between the businesses and the state. Vertical and horizontal integration of services can be realized, enabling the integration of information and services from various government agencies to help businesses and other stakeholders get seamless services. Level (3), or internal relationship: Refers to the degree at which the government service is reengineered by the company in the transformation from an off-line government service to an e-service. This e-government level enables the users to apply their knowledge in decision-making concerning social contribution activities of employees, corporate tax, or the registration of a new company.
In order to make good use of e-services with informed, transparent, and accountable decisions consistent with the past, compliant with the law, and coherent with similar decisions in other places, all the memories stored in the OM (hard and soft) should be taken into consideration. However, implementing e-government services using the OM has brought some dificulties. For example, Schachter (1995) suggests that many individuals do not understand why they might want to take the trouble to seek out information about government Web sites. In fact, e-services may require a time commitment that many employees are not willing to give regularly. Furthermore, e-government systems can ensure obedience to given rules, but they do not correct some errors generated by technologies (e.g., e-government systems cannot ind unhappy employees or take corrective actions immediately). Consequently, it seems logical to account for the human attention and creativity needed for the renewal of archived knowledge and innovative application of e-government in faster e-services (saving time and gaining lexibility). Green (2000) describes individual competencies as a written description of measurable work habits and personal skills used to achieve working objectives. He emphasizes that individual competencies contribute to achieve goals, but they are also part of a work system that leads to the results sought. These indings support the view of Malhotra (2000), that what is done with data and information depends on a subjective interpretation of individuals and groups who transform these inputs into actions and performance. Boyatzis (1982) maintains that effective performance in managerial jobs depends on personal qualities and deines these
212 Cegarra-Navarro
personal qualities as competencies. Under his competency assessment model, individual performance was no longer perceived as a single subjective process by the individual but rather an interaction between the environment, job requirements and competencies of the individual. It suggests that effective performance (e.g., implementing e-government) will occur when all three critical components are consistent or it. If any one or two components fail to correspond, then Boyatzis concludes that ineffective behaviour will ensue. Therefore, the successful use of e-government comes from making connections between the environment and the job demands, job requirements, and individual competencies, and even between the environment and the individual competencies. With regard to this, Cegarra et al. (2005) assert that as OM delivers the right information to the right person at the right time, considerable differences exist between the knowledge people have access to and the knowledge they require to improve the usability of e-government. Consequently, companies have to identify the knowledge required to redesign the job that exploits the available knowledge stock but also reduces the gap of existing e-government knowledge. On the other hand, people generally do not have the time to contemplate their own stock of knowledge and consequently do not appreciate what is important to improve the usability of e-government and pass successively from level (1) to levels (2) and (3) of e-government (Bontis, 2001). In other words, as each type of knowledge user has different requirements for knowledge utilization, considerable differences exist between what people know about e-government and what they need to know. The effective management of this alignment yields the reactivation and development of new information, which fosters changing cognitive patterns and integration of new knowledge in members of the organization. However, at this stage, the interpretations or new knowledge are individual knowledge and often tacit. Therefore, it is necessary that this updated knowledge become embedded within organizational memory-structures in order that it becomes a component of the dominant design. New knowledge may be further consolidated through the emergent understandings that are created by group members when they interact (Schein, 1992) or by new technological systems that may offer a better way to deliver information (Cross & Baird, 2000). In order to fulil the challenges above, the implementation of e-government could be a dynamic process, where moving from the environment stage to the job requirement stage requires the identiication of problems as there are differences between the necessary knowledge to improve the usability of e-government and the current knowledge (e.g., some technologies may not promote active external relationship level if they are not adequately designed or do not exist in an appropriate context). At the same time, moving from the job requirement stage to the competences stage requires a shift from the organizational to the individual concept, as each type of knowledge reuser has different requirements for knowledge utilization. However, moving from the competences stage to the environment stage entails taking personally constructed cognitive maps and integrating them in a way that develops a shared understanding among the group members. Taking into account this classiication, factors to implement e-government could be grouped into three blocks: 1. 2. Those inluential factors that facilitate the identiication of problems (i.e., accessibility to the requirements of jobs to use e-services); Those inluential factors that facilitate changing cognitive patterns (i.e., the ability to change the problem identiied in the former phase both internally or externally);
Figure 2. Implementation of e-government

EXTERNAL RELATIONAL INTERNAL
Identification of problems
HARD & SOFT MEMORIES
Environment
Job requirement Changing cognitive patterns
Incorporation of new measures Competencies
WORK HABITS & PERSONAL SKILLS
3.
Those factors that facilitate the integration of new measures in the organization.
Figure 2 provides a synopsis of a process that could be used to enable a better design and use of e-government services. In the following subsections the different stages of such a process are described.
Identiication of Problems
It can be said that any form of change begins with failures, which have generated events that do not conform to expectations in individuals or organizations (e.g., not being able to ind the right e-service and information; Schein, 1993). Failures result in processes in which old ways of thinking and behaving are discarded and new ways can be accommodated. Consequently, organizational culture should foster the belief that employees have different points of view, some of which might be more valid or accepted than others. However, group members can become narrowly focused and fail to adopt alternative perspectives. Janis (1982) calls it groupthink and relates it to the tendency for members of the highly cohesive groups to lose their critical abilities and adopt narrow deinitions of problems. To address this issue, Janis (1982) developed strategies that involve personal rotation among different functional areas. This aims at permitting an organization to have different members at each meeting, allowing a multidisciplinary perspective, a greater luency of knowledge, and facilitating the creation of conlicts. Problem identiication about e-government could make organizations aware of the existence of cultural features in organizations that enable or inhibit alternative perspectives on government services to existing ones. For both of them (the companies and the governments) to ind a possible way to identify problems is to enable people to participate in different organizational activities related to the provision and use of e-services and involve other
214 Cegarra-Navarro
stakeholders (i.e., technology vendors). The use of systemic and participative approaches that enable identiication and sharing of different values and viewpoints in relation to change could provide useful insights. In the realm of systems thinking, a variety of methodologies are now available for use in intervention (see, for example, Clarke, 2001; Midgley 2000). Using these methodologies, companies could be supported by enabling individuals to review their expectations and views about e-government.
Changing Cognitive Patterns

Organizations need to develop new knowledge with or about e-government. We argue that to do so, they would need to change their cognitive patterns, or, in simple terms, they need to incorporate new know-how to allow e-government to be part of their business. Pfeffer (1998) suggests that some key obstacles for developing new knowledge in organizations are related to the existence of authoritarian management styles and hierarchies. To overcome this, Nonaka and Konno (1998) suggest intensifying dialogue among individuals, groups, and organizations. Hedlund (1994) argues that three fundamental catalysts are required to facilitate the low of individual knowledge throughout an organization: common language, lateral communication, and mutual adjustment. This can result, for instance, in the design and sharing of boundary objects (Wenger, 1999) among individuals who see themselves as part of a learning community. Boundary objects help communities to co-ordinate different activities and interact with the rest of the world. In the case of e-government, these objects could be designed around the idea of having a learning community of irms. To design such objects, Brown and Duguid (1998) suggest that irst there should be a climate of mutual trust, which is in turn based on mutual adjustments between different communities or teams. Systems methodologies also have been used to design alternative scenarios for organizations and their environments (Ackoff, 1981; Checkland, 1981). This requires establishing dialogue processes with stakeholders on what constitutes improvement(s) for them (Midgley, 2000) and relecting potential consequences that decisions for change will have. For companies using e-government services, this would require conceiving of changes that are technologically feasible (Ackoff, 1981) as well as culturally desirable (Checkland, 1981). In this regard, the design of improvements on e-government design and use (including any boundary object) can also incorporate the use of information systems and technologies to support feasible and desirable improvements (Crdoba, 2005, pp. 32-34). In many cases, EDTs offer opportunities for public servants and employees to directly participate in decision-making by allowing them to provide their own ideas and suggestions in forums and online communities (Cegarra & Sabater, 2005). This suggests that if EDTs are designed carefully and openly, they can be valuable resources for transparency as customers, employees, and other stakeholders (e.g., government institutions) should be able to see political and governmental information, rules, and policies.
Incorporation of New Measures

At the highest level (internal stage), it is when employees interact directly with government institutions that several bureaucratic procedures can be eliminated by completing transaction
forms and sending them via the Web site or application programs, which means eficiency in money, time, and resources. This, in turn, presupposes that companies incorporate e-services as part of their business strategies. From an individuals point of view, adopting this level requires investing time and effort in learning. Nonaka and Konno (1998) assert that the implementation of new measures requires individual autonomy, but it will only occur if this is fostered by management. Schein (1993) suggests that employees should be motivated to take risks, innovate, and come up with creative solutions to problems to facilitate unlearning. PCTs are also especially important at the highest level, where government institutions can deliver product and service directly to companies. Many existing sources of knowledge, laws, comments to laws, speciic regulations, old similar cases, available case-speciic documents and information, and so forth are prevalent at different places and in different forms and representations at several degrees of formality, and are related by manifold links. As we have noted above, the purpose of e-government will be different for everyone, but inside an organization, e-government will be driven by institutional objectives (e.g., survival and growth). From an individuals point of view, e-government can be conceived as an investment (the time and effort to learn) in the e-service. Consequently, to foster the e-government it is necessary to extend the range of indicators and measures by identifying common measures for organizations, individuals, and government institutions (e.g., considering aspects such as the number of hours devoted to the utilization of e-government rather than the number of projects presented). Additionally, the provision of knowledge via e-government services should act as an up-to-date source of information, and with the technology that organizations can (and should) manage at a particular moment in time. With regard to this, Cegarra and Dewhurst (2006) suggest that technological gatekeepers act as a catalyst to guard against going backward; therefore, they play an important role in ensuring the successful incorporation of e-government.
Practical Implementation of E-Government

A way of assessing the appropriateness of e-government Web sites toward the exploitation of knowledge by using OM could be to consider these Web sites boundary objects (Wenger 1999). For instance, enabling asking questions about e-government through Web sites could provide an indication of what groups of individuals interact together (i.e., public servants, the elderly, unemployed, businessmen, etc). The existence of such groups could also enable irms to create virtual communities that bond participants to a particular e-service by enabling frequent interactions on a wide range of topics, and thereby creating a loyalty and enhancing transaction frequency. This perspective of assessing the existence and development of boundary objects supports the Hagel and Armstrong (1997) suggestion that implementing new and lexible technologies and systems must enable current competencies for communication building, people networks, and on-the-job learning. The adoption of e-government requires irst an understanding of the nature of services to be provided or currently online. The identiication of problems could lead to improving the design of the transactional aspects of services and providing structures that enable the appreciation of e-government services in relation to the provision of memory and learning. With these in mind, the (re)design of services will be informed by the perspective of enabling those providing and using services to communicate better, gain autonomy, and expand
216 Cegarra-Navarro
their learning capabilities. Therefore, in order to get level (1) of e-government, discussions on e-services should be fostered until a course of action can be agreed upon. In this aim, managers have to make the e-service available at multiple off-line and online access points into the company. PCT can generate internal eficiency and external co-ordination through changes in intra- or inter-organisational integrative processes (Kambil, Nunes, & Wilson, 1999). PCT managers make knowledge available wherever and whenever it is needed, so it is easier to ind the e-services and Web sites they are looking for. Similarly, the ability to successfully mobilise technology resources in conjunction with other resources has been judged critical to superior performance (Bharadwaj, 2000). Therefore, from this point of view, PCT should improve operational performance, which, in turn, may affect higher levels of e-government. On the other hand, online penetration of e-service (level 2) can be improved by establishing links (e.g. pop-up windows), thus making employees more aware of the existence of the e-service. Therefore, the middle stage of e-government activities are coordinated across the borders of different communities (e.g., public servants and employees) using EDTs to facilitate faster information communication. Enabling employees to have instant access to information about government Web sites, managers will foster the employees identiication of problems related to e-government through what has already been learned and stored in the OM. At the highest level (Level 3), an organizations government e-services have to be aligned with the strategies of the corporation and external organizations, that is, with public servants, customers, and suppliers. We suggest that through EDTs, organizations create valuable decisions by focusing e-government efforts on more important problems for the organization and its customers. Moreover, it can also provide a way of assessing how e-government services operate at the levels presented above. For example, for organizations to improve the design and use of e-government services from level (2) (relational) to level (3) (internal), managers should evaluate how online applications are used by employees as objects that communicate two different communities (e.g., employees and government oficers). Only in this way are there certain possibilities that government institutions can shift their relationship with companies from off-line or in-line to online service.
Conclusion
E-government is considered a method by which a government can transform the way it does business with users. However, citizens usually take advantage of e-governments when colleagues direct them to a speciic location in the net for lessons or tools (Cegarra et al., 2005). These indings support the views that technologies cannot predeine and predict who is the right person, what is the right time, or what constitutes the right information; only individuals can apply their own experience and contextual understanding to interpret the details and implications of a particular situation in order to determine what is the appropriate action or actions to take (Gold, Malhotra, & Segars, 2001). For example, employees, rather than engaging in an extensive search through an organizations information technologybased repository of knowledge (e.g., databases), turn irst to friends and peers to learn where
to ind relevant knowledge. This suggests that public institutions and governments might have overestimated the power of e-government to shape the users behaviour. Governments may be over-investing in the development of e-government Web sites and under-investing in mechanisms to facilitate the implementation of e-government among companies in a subsequent step. Many current generations of e-governments are ofline services converted to online use without further development. In practice, however, implementing e-government by brute force is a poor strategy. Firstly, mandating usage and coercing users are likely to create hostility and tension, and may stimulate resistance against e-government. Secondly, successful implementation of e-government requires that the intended users are motivated to share their knowledge with others and participate actively in building up the e-service. This indicates that a major reason for companies failing to beneit from e-government is that governments failed to integrate e-government Web sites into their businesses (i.e., basic usability problems still play a major role in the satisfaction and usage of e-services). Considering this, we suggest that for users of e-government to realize the full potential of the system, they must be willing to use the technology and become an effective user. This means working through a speciic problem scenario and encouraging an open sharing of information about navigation, contents, and tools of government Web sites. Therefore, governments should strive to create an environment in which organizations and employees can use e-government services easier. In order to fulil this objective, this chapter suggests that three different contexts need to be addressed by the organization: 1. The provision of a context that makes it possible for all individuals to access the information that is required to improve the usability of e-government. In this aim, the development of e-government should focus on making the service available on the Internet. The provision of a context that makes it possible for all individuals to change their cognitive patterns. In this aim, the development of e-government should not only focus on making the service available on the Internet but also examine the different delivery service platforms. A multi-channel access mixed with a range of different contact points (e.g., government ofice, help disk, telephones, shops, Internet, etc.) will help users to develop relationships more quickly. The encouragement of implementing of new routines and technological systems. In most cases, implementations success or failure will depend on the perceived beneits and costs to individual users (now and in the near future), and when network effects are signiicant, they will play a crucial role for e-government adoption. Encouragement also means that users feel that their contributions are worthwhile. Members may express much disappointment unless their contributions will be used to change rules and procedures.
2.
3.
The study has some limitations. Any managers wishing to implement e-government must understand what the sources of value are in their companies. Nevertheless, the response to this issue is complex, since the differential values that companies have are intangible and heterogeneous elements and, furthermore, the management of these elements will be dif-
218 Cegarra-Navarro
ferent, depending on the type of business, its structure, and the strategy of the organization. Therefore, other factors that have not been included in this study are also likely to affect e-government. Taking into account its limitations, a possible research direction could extend the range of indicators and measures by identifying common measures for organizations, governments, and individuals. Another possible research direction could examine how the private sector can contribute to e-government implementations. For instance, the private sector can provide information about technologies and standardization issues. Companies can disclose the problems that they have experienced by using technologies as well as their countermeasures. Finally, some e-business literature has also indicated differences between large organizations and small-medium enterprises in e-business systems adoption (Riquelme, 2002). Small irms typically have smaller chances to fully enjoy the beneits of e-government. Therefore, for small irms and governments, the challenge is to be able to incorporate e-government services use, and this suggests that e-government implementation can be seen as an opportunity to enhance technology use in small irms. Similar studies of e-government, therefore, should be conducted to examine these opportunities.
References
Abramson, A. M., & Means, E. G. (2001). E-government, Price Waterhouse Coopers endowment for the business of government. Lanham, MD: Rowman & Littleield Publishers Inc. Ackoff, R. (1981). Creating the corporate future: Plan or to be planned for. New York: John Wiley & Sons. Bent, J., Van der, Paauwe, J., & Williams, A. R. (1999). Organisational learning: An exploration of organizational memory and its role in organisational change processes. Journal of Organisational Change Management, 12(5), 377-404. Bharadwaj, L. (2000). A resource-based perspective on information technology capability and irm performance: An empirical investigation. MIS Quarterly, 24(1), 169-196. Bontis, N. (2001). CKO wanted: Evangelical skills necessary: A review of the chief knowledge oficer position. Knowledge and process Management, 8(1), 29-38. Brown, J., & Duguid, P. (1998). Organising knowledge. California Management Review, 40(3), 90-111. Cegarra, J. G., & Dewhurst, F. W. (2006). Linking shared organisational context and relational capital through unlearning: An initial empirical investigation in SMEs. The Learning Organization, 13(1), 49-62. Cegarra, J. G., Dewhurst, F. W., & Briones, A. J. (2005, December 8-10). Implementing egovernment through learning objects in the Spanish telecommunication industry.In Proceedings of 4th Iberoamerica Academy of Management, Lisbon, Portugal. Cegarra, J. G., & Sabater, R. (2005). E-learning: Organizational requirements for successful Feedback learning. Journal of Workplace Learning, 17(5/6), 276-290.
Checkland, P. (1981). Systems thinking, systems practice. London: John Wiley and Sons. Clarke, S. (2001). Information systems strategic management: An integrated approach. London: Routledge. Crdoba, J. (2005). Evaluating e-government services: A contribution from the theory of communities of practice. In E. Coakes & S. Clarke (Eds), Encyclopaedia of communities of practice in information and knowledge management. Hershey, PA: Idea Group Reference. Cross, R., & Baird, L. (2000). Technology is not enough: Improving performance by building organizational memory. Sloan Management Review, 41(3), 69-78. Fraga, E. (2002). Trends in e-government: How to plan, design, secure, and measure e-government. In Proceedings of Government Management Information Sciences (GMIS), Santa Fe, New Mexico. Gold, A., Malhotra, A., & Segars, A. (2001). Knowledge management: An organizational capabilities perspective. Journal of Management Information Systems, 18(1), 185214. Green, P.C. (2000). Robust competencies. Executive Excellence, 17(1), 17. Hagel, J. I., & Armstrong, A. G. (1997). Net gain: Expanding markets through virtual communities. Boston: Harvard Business School Press. Hamel, G., & Prahalad, C. K. (1994). Competing for the future. Cambridge, MA: Harvard Business School Press. Hauser, M. (1998). Organizational culture and innovativeness of irmsAn integrative view. International Journal of Technology Management, 16(1/2/3), 239-255. Hedlund, G. (1994). A model of knowledge management and the n-form corporation. Strategic Management Journal, 15(5), 73-90. Heichlinger, A. (2004). Egovernment in Europes regions: Approaches and progress in IST strategy. Organisation and services, and the role of regional actors. Maastricht (Netherlands): European Institute of Public Administration. Hoffmann, T. (1999). The meanings of competency. Journal of European Industrial Training, 23(6), 275-86. Janis, I. (1982). Victims of groupthink. Boston: Houghton-Miflin. Kambil, A., Nunes, P. F., & Wilson, D. (1999). Transforming the marketspace with all-in-one markets. International Journal of Electronic Commerce, 3(4), 11-28. Malhotra, Y. (2000). Knowledge management for e-business performance. Information Strategy: The Executives Journal, 16(4), 5-16. Megill, K. A. (1997). The corporate memory. Information management in the electronic age. London: Bowker & Saur. Midgley, G. (2000). Systemic intervention: Philosophy, methodology and practice. New York: Kluwer Academic/Plenum. Miles, G., Miles, R. E., Perrone, V., & Edvinsson, L. (1998). Some conceptual and research barriers to the utilization of knowledge. California Management Review, 40(3), 281288.
220 Cegarra-Navarro
Nonaka, I., & Konno, N. (1998). The concept of Ba: Building a foundation for knowledge creation. California Management Review, 40(3), 40-54. Pfeffer, J. (1998). The human equation: Building proits by putting people irst. Cambridge, MA: Harvard Business School Press. Riquelme, H. (2002). Commercial Internet adopting in China: Comparing the experience of small medium and large businesses. Internet Research: Electronic Networking Applications and Policy, 12(3), 276-286. Schachter, H. L. (1995). Reinventing government or reinventing ourselves: Two models for improving government performance. Public Administration Review, 55(6), 530537. Schein, E. (1992). Organizational culture and leadership. San Francisco: Jossey-Bass. Schein, E. (1993). How can organisations learn faster? The challenge of entering the green room. Sloan Management Review, 34(2), 85-92. Selnes, F., & Sallis, J. (2003). Promoting relationship learning. Journal of Marketing, 67(3), 80-95. Sherwood, D. (2000). The unlearning organisation. Business Strategy Review, 11(3), 3140. Sicilia, M. A., & Lytras, M. D. (2005). On the representation of change according to different ontologies of learning. International Journal of Learning and Change, 1(1), 66-79. Spender, J. C. (1998). The dynamics of individual and organisational knowledge. In Eden & Spender (Eds.), Managerial and organisational cognition. London: Sage. Tapscott, D. (1996). The digital economy. New York: McGraw Hill. Tat-Kei Ho, A. (2002). Reinventing local governments and the e-government initiative. Public Administration Review, 62(4), 434-444. Unpa, & Aspa. (2001). Benchmarking e-government: A global perspective. Retrieved from http://unpan1.un.org/intradoc/groups/public/documents/un/unpan003984.pdf. Walsh, J. P., & Ungson, G. R. (1991). Organizational memory. Academy of Management Review, 16(1), 57-91. Wenger, E. (1999). Communities of practice: Learning, meaning, and identity. Cambridge: Cambridge University Press. Woodruffe, C. (1993). What is meant by a competency. Leadership & Organization Development Journal, 14(1), 29-36. Woolcock, M. (1998). Social capital and economic development: Toward a theoretical synthesis and policy framework. Theory and Society, 27(2), 151-208. Zuboff, S. (1988). In the age of the smart machine. New York: Basic Books.
Designing and Implementing Views on Competencies 221
Chapter X
Designing and Implementing Views on Competencies

Marcel van der Klink, Open University of The Netherlands, The Netherlands Kathleen Schlusmans, Open University of The Netherlands, The Netherlands Jo Boon, Open University of The Netherlands, The Netherlands
Abstract
In education as well as in human resource management, the concept of competency is becoming more and more important. However, competency is a fuzzy concept that lacks coherent terminology and is used in many different meanings. In the irst part of this chapter, competency is deined as a multi-dimensional construct with ive dimensions: speciicity, coherency, durability, activity, and trainability. Then two approaches are presented to establish the competencies organisations or educational institutes should address: the function-based approach, which focuses requirements for good performance on a speciic function, and the employee-based approach, which focuses on individual competencies for excellent performers. Finally, it is argued that a competency-based curriculum should move toward self-directed competency-based learning (SDCBL), using personal learning plans and adopting the lexible learning opportunities of e-learning. However, SDCBL can only be successful when learners can be taught to be responsible managers and owners of their own learning process.
222 van der Klink, Schlusmans, & Boon
Introduction
One of the ideas that has gained much attention these days is the notion of competency. There are many scholars and practitioners claiming that competency serves as a strong impetus to improve human resource management and education, in particular vocational and professional education. In addition to this, there is a strong belief that e-learning will contribute to the fast and broad adoption of competency-based practices in organisations and education. In this chapter, we start exploring the notion of competency, and we argue that this concept is, in fact, too fuzzy. Competency is a rather controversial concept: It lacks a coherent terminology and therefore it is unclear what it actually means. To improve the application of the concept of competency, we present a framework to deine competency more precisely. After exploring the concept of competency, we provide two dominant perceptions of competency-based learning that are grounded in two different contexts. First, we will elaborate on competency-based learning in organisations, followed by a discussion about the application of the concept of competency in education. What becomes clear from these descriptions of applications of competency-based practices is that competency-based learning differs from more traditional practices. However, for the near future we expect that new types of competency-based learning will emerge. In the last part of this chapter, we claim that mainstream competency-based learning (MCBL) will move toward what we deine as self-directed competency-based learning. We will compare traditional education to MCBL and SDCBL. The chapter ends with some remarks for discussion.
The Triumph of a Fuzzy Concept

There are several motives and circumstances that contribute to the contemporary prominent status of the concept of competencies. This section lists the utmost important factors, followed by an overview of the various views on competencies. The concept of competency refers to personal competencies as well as corporate competencies. Personal competencies consist of experience, skills and abilities of an individual. Corporate competencies consist of skills and expertise that belong to the organisation itself; they are embedded in company processes, and they tend to remain in the organisation even when individuals leave. They consist basically of routines, knowledge, and behaviour (Nelson & Winter, 1982). Kessels (2004) describes the core competencies of organisations as the knowledge, skills, and attitudes that add the main value to a product or service, distinguish the organisation in competition with others, and can be applied by the organisation to new product ields and services. The focus in this chapter, however, is not on organisational but on personal competencies. Van der Klink and Boon (2003) and Delamare Le Deist and Winterton (2005), among others, investigated causes and motives underlying the adoption of the concept of competencies. Table 1 lists the most powerful factors. As well in organisations as in education (vocational
Table 1. Motives for adopting the concept of competencies

Organisations Changing environment Shift of focus Coping with unpredictable future Stronger focus on output and performance Education Responding to labour market developments Stronger focus on professional skills and employability New concept for communication with employers
Co-operation and communication
Competency as unifying and appealing concept
and higher professional education), there is strong recognition that it is quite impossible to predict the future. The pace of technological innovation in products and processes together with the impact of economical globalisation creates a climate wherein uncertainty about the future remains prominent. One of the interviewed experts (a professor in information technology education) in Van der Klink and Boons report stated: The description and analysis of tasks in order to deine which skills, knowledge and attitudes are required is becoming increasingly dificult. The content of tasks is changing so rapidly that requirements cannot be deined in detail. The shift to the concept of competencies is therefore not a fashionable hype but a necessity for organisations to cope with uncertainty. (Van der Klink & Boon, 2003, p. 129) That is why it is necessary to increase the lexibility of the workforce to remain in competition. Demand-driven models of education and human resource management replace supply-driven types of education. There is, however, especially in higher education, no agreement upon the necessity to alter contemporary practices. Barnett (1994), for example, argues that the shift toward competency-based education is in fact the inishing stroke for the education of traditional academic virtues. He warns against the rather exclusive focus on vocational valuable skills that will lead to an underestimation of relective, generic, and academic competencies in curricula. Nevertheless, in organisations and in education, too, there is vivid discussion going on regarding re-balancing speciic and generic competencies. The challenge is to ind a balance that will improve immediate application of what is learned and at the same time assuring that other competencies are acquired that are presumably required for employability in the long run. Facilitating the communication and collaboration between various interested parties is mentioned quite often as a motive for the contemporary attention for the concept of competency. Organisations strive for integrated HRM policies, and competency serves here as a unifying concept for redesigning selection, training, career development, performance assessment, and reward systems. The concept of competency is more tangible than, for example, other popular concepts such as the learning organisation (Senge, 1990). It is more appealing and its better into the attempts to redesign human resource policies. In vocational and professional education, the adoption of the concept of competency is often justiied by referring to the need to establish a common language to discuss the content of curricula with labour market representatives (e.g., employers, unions, and professional bodies).
Table 2. Three perspectives on competencies

Perspective Position U.S.A. Geographic U.K. Difference in deinition Competency refers to behaviour and personal traits that contribute to excellent performance (e.g., Spencer & Spencer, 1993). Competence refers to collectively agreed occupational standards such as national vocational qualiications (Mansield and Mitchell, 1996). Kompetenz refers to the capacity of a person to act. It is more holistic than competency or competence, comprising not only content or subject knowledge and ability but also core and generic abilities (Delamare Le Deist and Winterton, 2005) In education and training, competencies are deined as clusters of skills and knowledge that can be learned. In selection practices, competencies are perceived as partly trainable and as partly rather stable traits that are dificult to alter. In performance assessment practices, competencies are deined as output of a task. Emphasises to link competencies with observable and measurable performance. Stronger focus on top-down development of competency-based systems. Stresses values and beliefs as important components of competencies. Stronger focus on workers participation in development of competency-based systems.
Germany
Training and education Field of application Selection
Performance assessment Cognitivism Underlying learning theory Constructivism
As Table 1 shows, the concept of competency serves as a remedy for solving rather different problems. Its usage for different purposes clearly expresses that it is a rather fuzzy concept allowing ample interpretations. However, because of its vagueness, it is probably at the same time a rather attractive concept. The vagueness attached to competency makes it dificult to use it as a practical concept for designing learning practices in organisations and education. Or, as Delamare Le Deist and Winterton (2005) stated: There is such confusion and debate concerning the concept of competence that it is impossible to arrive at a deinition. Van der Klink & Boon (2003) illustrated the conceptual confusion surrounding the concept of competency. They described three broad perspectives on the concept (see Table 2 for a brief summary), and within each perspective they distinguished several positions. It should be stressed that the three perspectives do not automatically exclude each other. This also holds for the positions within a given perspective. In practice, deinitions combine various perspectives or several positions within a speciic perspective. It is obvious that this should cause extra confusion. Notwithstanding the different perspectives on competencies, several attempts are undertaken to achieve shared agreement concerning the basic features of competencies (see, for example, Eraut, 1994; Weinert, 2001). Here the framework of Merrienboer, Van der Klink, and Hendriks (2002) is adopted, which was approved by the Dutch Education Council. The
authors recognise that competency is a multi-dimensional construct, and they propose a set of ive dimensions to deine competency, which are briely described. 1. Speciity: Some competencies are rather generic (e.g., learning competencies), implying that they can be trained, developed, and applied in a wide range of settings. Other competencies, however, are more restricted to a particular context. But for every competency it is possible to determine the range of contexts. Coherency: Competencies encompass combinations of the following elements: skills, knowledge, attitudes, and sometimes personal characteristics. The relative importance of the above-mentioned elements may vary in different deinitions. Durability: Competencies possess a certain degree of durability. For example, the competency designing instructional materials has always been relevant for human resources development professionals, and in the future this competency will remain important. However, this does not imply that the content of this competency remains unaltered. New notions about learning theories and the availability of new training delivery modes (e.g., e-learning) require additional knowledge and skills. Activity: Competencies are connected with activities. For the development and the assessment of competencies, one must engage in activities. It is noteworthy to mention that the notion of activity encompasses also mental activities that are not observable immediately. Trainability: The trainability of competencies varies and depends on the relevancy of dificulty to alter personal traits for the performance of the competency.
2.
3.
4.
5.
According to Merrienboer, Van der Klink, and Hendriks these ive dimensions need to be addressed in every deinition of competency to enhance proper communication about this concept. In this chapter, we stress the notion that competency is a multi-dimensional construct and that striving for one deinition is not the right way. We therefore stress the importance of these ive dimensions and do not provide a single deinition because we feel this is not contributing to the sound adoption of the notion of competencies.
Deining Competencies in Organisations and Education

In this section, we describe how competencies can be deined in two different settings: in organisations to evaluate and improve organisational performance, and in education to improve the students capabilities for professional performance. In both cases, ICT can play a crucial role; we discuss this aspect in the next paragraph.
Organisations
From a management point of view, two different approaches can be used to deine competencies in organisations, leading to a different view on learning and training needs. Both approaches are based on management by training (see Van der Klink & Boon, 2003).
The irst approach is function-based and focuses on the requirements for good performance in a speciic function. Based on an overview of tasks, responsibilities, and expected output, an inventory is the starting point to describe the competencies needed to fulil the function. This inventory of competencies can be presented in competency maps that can be used to describe the gap between the knowledge, skills, and attitudes individual employees possess and those they still have to acquire. The use of this mapping method is associated with training plans, aimed at the elimination of the gaps. In practice organisations, work with competency panels consisting of managers discussing the requirements put onto the team or the division, the responsibilities of the different functions involved, and examples of the behaviour that fulils the requirements. Based on this information a set of competencies is formulated and associated with the corresponding behaviour. Training based on function analysis is directed toward the realisation of the minimum requirements employees have to meet for performance in the function. A second approach to deine competencies is to start from the viewpoint of the employee. In this approach excellent performance is the starting point for competency description. One method for this approach is developed in the U.S. (McClelland, 1998) and entails the analysis of predictors for success in the fulilment of functions by analysis of excellent performers. This method focuses strongly on individual characteristics that deine success. The analysis intends to give an answer to the question of how to make sure that employees perform on the highest level and how to organise development tracks that ensure the best possible value added to the organisation. In practice, the identiication of excellent performance is based on interviews with frontrunners in the organisation using the technique of the behavioural event interview. In this interview, employees are asked to describe in a very detailed- and labour intensive- way their functioning in speciic situations, stressing what they thought, found, felt, and did. The results of the interviews are aggregated into competency proiles. These can be used in the phase of selection and recruitment of personnel or in professional development. The surplus value of the use of the detailed data is that it offers insight into the factors hindering excellent performance; these factors can be related to actual knowledge and skills, but can also be related to factors such as the perception of the self, norms, and values or motives.1 Another method for analysis of competencies from the point of view of employees consists of the measurement of competencies by the deinition of indicators that relect its intermediate products or consequences. The argumentation to use this method is that skills and knowledge cannot be observed and that the only observable factors are the eficiency and the effectiveness manifested in the activities carried out. An example of this empirical method is developed and tested by Escrig-Tena and Bou-llusar (2005). They report on four indicators, describing behaviour, indicating a competency leadership. The attracting feature of this approach is the claim to measure and even to quantify skills, knowledge, and attitude necessary for effective job performance.
Education
In order to implement the competency approach in education, it is important to focus on the context of work, companies, and managers, and not on the educational context; this implies that the concepts and tools have to be transformed, translated, or reinvented to it in an educational context. Methods to analyse the professional context in order to integrate authentic professional characteristics into curricula can focus on the description of a direct relation between professional functioning and educational requirements, the distillation of a few crucial characteristic features that can be used in the educational context or the analysis of context variables, social relations, the complexities of organisational structures. Most methods are based on interviews with professionals. They are asked individually or as a group, to relect on the competencies needed in speciic situations, critical incidents, and so forth. In a few cases, professionals cooperate with educational specialists in the development of a curriculum.
Toward a Competency-Based Curriculum

A competency-based curriculum has some speciic features. We try to describe these features by comparing three differences between three educational models: traditional learning, mainstream competency-based learning, and self-directed competency-based learning. Traditional learning occurs mostly in regular schools and universities, the focus on subject matter, a central curriculum, certiicates, and degrees. The curriculum is mainly monodisciplinary and teacher-centred, and the locus of control is the school or university that decides what is to be learned, how, and when and how it is assessed. Assessment is mainly summative and focuses on knowledge and skills. Mainstream competency-based learning came into existence as a reaction to traditional learning, which focused too much on degrees and not enough on preparing learners for the labour market. It was embraced by schools of vocational training, institutes for higher professional education, and in-company training. The main purpose of competency-based learning is to achieve a better match between the competencies of the learners and the demands of the labour market or the organisation. The curriculum is mainly multi-disciplinary, learner centred, and based upon authentic situations. The locus of control is still the school or the company, but there is an attempt to adapt the learning activities to the demands of the individual learner. One of the pitfalls of mainstream competency-based teaching is that only the needs of the organisation or the labour market are taken into account, and an individual is only trained in these competencies that are relevant to his or her job or functional performance; the training is linked to the business or labour objectives. This approach handicaps the ability of organizations to rapidly innovate and adapt, and it handicaps the individuals employability (cfr. Parkin, 2005). Also, the approach of MCBL puts too much responsibility for learning on the educational provider and not on the individual learner. That is the reason why comCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
Table 3. Comparison of traditional learning, mainstream competency-based learning, and self-directed competency-based learning
Traditional learning Main purpose To give out certiicates and degrees
Mainstream competency-based learning To achieve a better match between the competencies of the learners and the demands of the organisation or the labour market Based upon authentic situations, cases, and virtual companies in which learners have to demonstrate broad competencies, mainly multidisciplinary Carrying out authentic task and solving problems The curriculum is adapted to the entry level and personal needs of the learner within the framework decided upon by the school or organisation Demonstration of competencies
Self-directed competency-based learning The empowerment of the individual learner
The curriculum
Knowledge- and skills based, subject-oriented and monodisciplinary
No central curriculum but a personal learning plan in which the learner decides which learning situations he or she will use to acquire competencies The learners themselves decide which activities they will undertake to acquire competency The curriculum is completely adapted to the personal preferences and needs of the learners Demonstration of competencies in a way that is decided on by the learner Part of the study process Portfolio assessment, performance assessment, peer assessment Learning is a combination of learner-centred and teachercentred activities, to be decided by the learner
Learner activity
Acquiring predeined content
Flexibility
The curriculum is the same for all learners
Assessment (what)
Knowledge and skills
Assessment (when) Assessment (how)
At end of the study process Traditional achievement tests
Part of the study process Performance assessment, peer assessment Learning consists mainly of learner-centred activities
Teacher role
Learning is mainly teachercentred
petency-based learning has moved on in the direction of self-directed competency-based learning. In SDCBL, a shift of focus takes place; neither the trainer nor the training institute is responsible for the learning that takes place but rather the learner him- or herself. Note the competency-based curriculum lies at the heart of the learning but the personal learning plans. Personal learning plans or personal development plans are a suitable method for implementing SDCBL. In a personal learning plan learning is tailored to the needs of each learner and not only based on the job needs and the organizations core business requirements. The concept of self-directed competency-based learning ties in best with the focus on individual competencies, which is argued previously in this chapter.
The Personal Learning Plan as a Vehicle for Self-Directed Competency-Based Learning

The main component of SDCBL is the personal learning plan. It gives the answer to ive main questions (based on Mosman, 1989): Where am I coming from? Where am I now? Where am I going to? How do I get there? How do I know I have arrived? Where am I coming from?This question addresses the issue of assessment of prior learning. What learning has occurred in the past? Which learning experiences were successful and which were not? SDCBL takes all prior learning experiences of the learner into account, not only formal learning but also non-formal and informal learning. Where am I now?Several assessment methods are used to establish which competencies a learner already possesses, such as self-assessment, peer-assessment, assessment centres, and portfolio-assessment. Where am I going?The learner decides which competencies are going to be developed. The decision in which direction the learner is going should be guided by clear descriptions of professional and educational competency proiles. However, the main focus is on the learner who is responsible for the decision of what he or she is going to learn, not the organisation or the school. How do I get there?A learner should be able to choose between a large variety of learning activities to achieve the competencies that are going to be developed. The learning activities can vary from attending lectures to being trained at the workplace. There is not one correct educational method to achieve competencies. How do I know I am going the right way and how do I know I have arrived?This question addresses the assessment. There are several ways in which a learner can
demonstrate that he or she has achieved the level of competency that was mentioned in the personal learning plan, such as the assessment methods mentioned above.
E-Learning and Self-Directed Competency-Based Learning

There is a logical link between self-directed competency based learning and e-learning. Elearning ties in with all the aspect of a personal learning plan and offers an opportunity to adjust learning to the individual learners need. Via e-learning, a huge range of learning activities can be made available, many of which offer the opportunity to learn in ones own time, pace, and place. One of the major challenges for educators will be to develop repositories of learning materials that can be linked to competencies and to provide adequate metadata that allow learners to choose those learning activities that match their learning needs. E-portfolios provide a vehicle students can use to document and demonstrate their own achievements. Also via e-learning, a whole variety of assessment instruments can be made available. E-learning tools that really it in the SDCBL philosophy are, for example, communities of practice and Wikipedia. What these tools have in common is that the (individual) learner controls pace and content of the learning process, they can vary their level of personal involvement, and at the same time these tools provide opportunities for interaction, discussion, and collaborative learning. Thus learning remains a highly social event. Communities of practice are groups of people who share a concern, a set of problems, or a passion about a topic, and who deepen their knowledge and expertise by interacting on an ongoing basis (Wenger, McDermott, & Snyder, 2002, p. 4). E-learning tools can support the viability of the community of practice. However, the many attempts to establish communities of practice only using e-learning tools have made clear that these tools are not suficient for successful practices. Communities of practice need, in fact, other forms of communication and interaction, too, to truly become a vivid heaven for innovative learning. Tools can facilitate the development of communities of practice but cannot replace all other kinds of requirements. Wikipedia is, in fact, a communal encyclopaedia with free content. Wikipedia allows everyone to contribute voluntarily in an ongoing process. All the information in Wikipedia is free for anyone to copy, modify, and redistribute. It allows various forms of involvement: visitor, contributing to existing articles, or edit new articles. Wikipedia has some strengths, but there are also some problems (see http://en.wikipedia.org). Its strengths are access to a wide range of subjects, its open approach, and fast development of articles about the latest trends and topics (e.g., popular culture). On the other hand, information may contain omissions, and Wikipedia is more subject to subtle viewpoint promotion than other reference sources.
Discussion
In the irst paragraph it is argued that the transformation in the labour market, stressing employability, lexibility, and life-long learning abilities, affects the requirements toward education. To respond to the above-mentioned developments, the concept of competence is gaining more and more importance. The current shift to competencies as a key feature of curricula is manifest in all sectors of education. Although change is in the air, till now the competency-based approach is developed more extensively in vocational education on all levels while academic education is lagging behind, mostly by reluctance to adhere to a delineated professional domain of practice. Next we presented the description of professional proiles as one of the central elements in the development of competency-based education and showed that it is methodologically poorly developed. Moreover, the labour markets change rapidly and are not transparent enough to allow for a clear translation of requirements into educational curricula. The mix between long-term and short-term requirements poses several complex problems. Although it is necessary that competencies can be applied immediately, it is essential at the same time that they lead to employability in the longer run. In the last paragraph, we discussed three learning models and argued the beneits of self-directed competency-based learning. However, the paradigm of SDCBL poses several questions. It goes beyond saying that learners should take responsibility for their own learning; however, it can be argued that not everyone is a self-directed learner. According to Abdullah (2001), self-directed learners are responsible owners and managers of their own learning process (p. 1). They have the skills to access and process the information they need for a speciic purpose. We have to teach learners to become self-directed learners; it is not something that just happens. Are students and employees capable of learning and performing self-directed learning behaviour? And if not, what happens then? Is SDCBL just wishful thinking, or can we really make it happen? Therefore, it is essential to deine carefully what lies at the heart of the kind of self-directed learning that we advocate. Traditional self-directed learning is not suficient for successful performance in the contemporary knowledge economy. In traditional views, self-directed learning is highly attached to all kinds of theories and concepts concerning meta-cognitive learning strategies. The concept of self-directed learning we advocate is more linked with current ideas about employees self management in the knowledge economy (Pedler & Boydell, 1985). Besides meta-cognitive abilities, other aspects are important, too: intrinsic motivation, ambition, and the ability to cope with contrast viewpoints and insecurity. Maybe we could even use traditional educational methods to teach people to become self-directed learners. But in any case, e-learning and content management systems that provide access to a wealth of learning activities are essential conditions for successful SDCBL.
References
Abdullah, M. H. (2001). Self-directed learning (ERIC digest No. 169). Bloomington, IN: (ERIC Clearinghouse No. ED459458). Barnett, R. (1994). The limits of competence. Knowledge, higher education and society. Buckingham: The Society for Research into Higher Education & Open University Press. Bastiaens, T., Boon, J., & Martens, R. (2004). Evaluating integrated e-learning. In W. Jochems, J. van Merrinboer, & R. Koper (Eds.), Integrated e-learning. Implications for pedagogy, technology and organization (pp. 187-198). London: Routledge Falmer. Belbin, M. (2004). Management teams. Why they succeed or fail. Butterworth Heineman. Bommer, W. H., Miles, E. D., & Grover, S. L. (2003). Does one good turn deserve another? Coworker inluences on employee citizenship. Journal of Organisational Behaviour, 24, 181-196 Boon, J., Rusman, E., van der Klink, M., & Tattersall, C. (2005). Developing a critical view on e-learning trend reports: Trend watching or trend setting? International Journal of Training & Development, 9(3), 205-211. Delamare Le Deist, F., & Winterton, J. (2005). What is competence? Human Resource Development International, 8(1), 27-46. Eraut, M. (1994). Developing professional knowledge and competence. London: The Falmer Press. Escrig-Tena, B., & Bou-llusar J. C. (2005). A model for evaluating organisational competencies; An application in the context of a quality management initiative. Decision Sciences, 36, 2, 221-257. Hiemstra, R. (1994). Self-directed learning. In T. Husen & T. N. Postlethwaite (Eds.), The International Encyclopedia of Education (2nd ed.). Oxford: Pergamon Press. Reprinted here by permission. Retrieved from http://home.twcny.rr.com/hiemstra/sdlhdbk. html Kessels, J. W. M. (2004). The knowledge revolution and the knowledge economy. The challenge for HRD. In J. Woodall, M. Lee, & J. Stewart (Eds.), New frontiers in HRD (pp. 165-179). London: Routledge. Mansield, B., & Mitchell, L. (1996). The competent workforce. London: Gower Press. McClelland, D. C. (1998). Identifying competencies with behavioral-event interviews. Psychological Science, 9, 331-339. Nelson, R., & Winter, S. G. (1982). An evolutionary theory of economic change. Cambridge, MA: Belknap Press. North Western Regional. (sd). Educational laboratory developing self-directed learners. Retrieved December 19, 2005, from http://www.nwrel.org/planning/reports/self-direct/self.pdf Parkin, G. (2005). Competency based learning management. Retrieved March 23, 2006, from http://parkinslot.blogspot.com/2005/04/competency-based-learning-management.html
Pedler, M., & Boydell, T. (1985). Manage yourself. London: Fontana. Schlusmans, K., Koper, R., & Giesbertz, W. (2004). Work processes for the development of integrated e-learning courses. In W. Jochems, J. van Merrinboer, & R. Koper (Eds.), Integrated e-learning. Implications for pedagogy, technology and organization (pp. 126-138). London: Routledge Falmer. Senge, P. M. (1990). The ifth discipline: The art and practice of the learning organization. New York: Doubleday Currency. Spencer, L. M., & Spencer, S. M. (1993). Competence at work. Models for superior performance. New York: Wiley & Sons. Stoof e.a. uit tekst verwijderen (zie p.4) van der Klink, M. R., & Boon, J. (2002).The investigation of competencies within professional domains. Human Resource Development International, 5(4), 411-424. van der Klink, M. R., & Boon, J. (2003). Competencies: The triumph of a fuzzy concept. International Journal of Human Resources, Development and Management, 3(2), 125-137. van der Klink, M. R., Boon, J., & K. Schlusmans. (in press). Competencies and higher education. European Journal of Vocational Training. van der Klink, M. R., & Jochems, W. (2004). Management and organization of integrated e-learning. In W. Jochems, J. van Merrinboer, & R. Koper (Eds.), Integrated e-learning. Implications for pedagogy, technology and organization (pp. 151-163). London: Routledge Falmer. Van Merrienboer, Van der Klink, & Hendriks, M. Competenties: van complicaties tot compromis [Competencies: from complications toward a shared vision]. The Hague: Dutch Education Council. Velenturf, M. J., & Lap, E. (1998). Competency based development. In H. Burgt, R. van de Loo, F. van de Meijer, & M. J. Velenturf, Bewegen tot competentie. Lemma: Utrecht. Weinert, F. E. (2001). Concept of competence: A conceptual clariication. In D.S. Rychen & H.L. Salganik (Eds.), Deining and selecting key competencies (pp. 45-64). Seattle: Hogrefe & Huber Publishers. Wenger, E., McDermott, R., & Snyder, W. (2002). Cultivating communities of practice. Boston: Harvard Business School Press.
234 Koper & Specht
Chapter XI
TENCompetence:
Life-Long Competence Development and Learning
Rob Koper, Open University of The Netherlands, The Netherlands Marcus Specht, Open University of The Netherlands, The Netherlands
Abstract
In most solutions for supporting learning today, one single approach is leading the selection, installation, and usage of information technology (IT)-based tools. Either content-based approaches lead to the creation of a content-based infrastructure with course management systems and content repositories, or a collaboration background leads to the usage of virtual classrooms and collaborative learning environments. The TENCompetence project aims forward for integration of the different tools, perspectives, and learning environments in a common open source infrastructure based on todays standards on the level of knowledge resources, learning activities, competence development programs, and learning networks. TENCompetence will integrate tools in a service-oriented architecture (SOA) and evaluate the approach in a variety of pilot applications for life-long competence development.
TENCompetence: Life-Long Competence Development and Learning 235
Introduction
In the last seven years, a variety of tools and learning environments have been created and installed in schools, universities, and cooperates supporting learning. Mostly those tools have been created around learning content and collaborative learning activities like virtual classrooms. In the last two years, the terms competence, competency, skills, and knowledge have seen a renaissance, and the e-learning communities are becoming aware of the importance of competences and competence models as the driver for life-long learning. Competences enable users to work in their job; they enable students to learn and achieve the curriculum goals. Competences can be deined in a manifold of ways. There have been functional, cognitive, behaviorist, and many other approaches (for a nice overview and integration, see Cheetham & Chivers, 2005). The TENCompetence consortium interprets competence as all the factors for an actor to perform in an ecological niche. Performance includes the speciic context that is necessary for the interpretation of competence. Off-course competences include competencies and knowledge that are necessary to put the competence into performance. An example could be running a small bakery shop with all the necessary skills and knowledge ranging from the recipes and skills for baking bread to the selling and booking in the shop. Besides the speciic knowledge, competencies, and skills, the context in the ecological niche has an important impact, and to enable people to act in an ecological niche effectively, meta competences are necessary. To develop competences over time, including the changes of the ecological niche, meta cognitive processes and relection also play an important role (Schn, 1983). Competency models in the upper sense as models of interrelated competencies already play an important role in todays educational systems. In school curricula, competences build the basic structure to connect the different school levels and class curricula as well as their content. In organizations, competence-based assessments build the basis for controlling and steering services in the human resources (HR) departments, like stafing, career planning, and personalized training. Often the personalized selection of contents is mentioned these days as one main application for competence-based education. Nevertheless, daily practice in organizations often shows a focus on learning tools on different levels that are rarely integrated and perceived from a life-long competence development perspective. Often learning is driven by contents, so the main tools in organizations focus on the management, creation, sharing, and use of content. Discussions are focused on reusability of learning objects, the cheap and rapid production of learning contents, and content production for speciic training needs. The produced contents are rarely integrated with units of learning or learning designs. The developments regarding learning activities and units of learning have gained much visibility through the IMS Global Learning Consortium, Inc. Learning Design Speciication (Koper & Tattersall, 2005) developed at the Open University of the Netherlands and have opened a path for a standardized way of describing and sharing learning processes and a new focus on pedagogical and social issues in learning. Furthermore, the social aspects of learning like communities of practice, social exchange of learning resources, or informal learning aspects have gained more and more importance in the last years. Learning in this sense is no longer perceived as a mere business process that can be administered and handled in Enterprise Resource Planning systems but as an individual process that needs resources,
236 Koper & Specht
process structures, competence development programs, an enabling social environment, and active people to take place. TENCompetence is a European Integrated Project set up with the target to integrate different levels and approaches of learning content tools, learning activity tools, competence development programs, and learning networks in a common open source infrastructure to enable and foster life-long competence development and learning. The main objectives are: 1. To research and develop innovative methods and technologies for the creation, storage, use, and exchange of knowledge resources related to life-long competence development. To research and develop innovative, standards-based methods and tools for the creation, storage, use, and exchange of formal and informal learning activities and units of learning. This includes tools for the assessment of the learning process and learning outcomes. To research and develop innovative methods and technologies for the creation, storage, use, and exchange of formal and informal competence development programs (including the assessment of previously required competence levels, navigation support, and the sharing of successful formal and informal learning tracks). To research and develop models, methods and technologies for the creation, storage, use, and exchange of networks of competence development programs from different sources around Europe to support life-long competence development.
2.
3.
4.
This chapter will give an introduction to the problems and shortcomings of todays support for life-long competence development and highlight two main issues. After that, two main approaches for solving those problems will be introduced. First, the approach for the integration of different levels of granularity will be described and important links between content, competences, learning activities, and communities will be highlighted. Second, the aspects of social software in TENCompetence and its foundation in social exchange theory (Thibaut, 1959) and self-organization theory (Varela, Thompson, & Rosch, 1991) will be described, and irst approaches are sketched.
Background: Todays Challenges and Research Issues

The central need addressed by TENCompetence is to provide ubiquitous and life-long adapted access to facilities that support the creation, storage, use, and exchange of formal and informal knowledge and learning resources. To this end: "TENCompetence supports individuals, groups and organizations in Europe in life-long competence development by establishing the most appropriate techCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
nical and organizational infrastructure, using open-source, standards-based, sustainable and innovative technology." (TENCompetence, 2005) Seven major problems underlay the present lack of such an infrastructure. The solutions to these problems form the core requirement for the development of the TENCompetence infrastructure. 1. The pedagogical models that are applied in training, schools, and universities do not meet the demands and possibilities of life-long competence development and the new learning technologies that are available. The provision of such new, promising, innovative pedagogical approaches for life-long competence development will be supported by the TENCompetence infrastructure. The new models will integrate individual learning, collaborative learning, organizational learning, and knowledge management. They will not simply mimic the face-to-face approaches (e.g., virtual classrooms) and will be usable for informal, self-directed learning in companies or at home. Furthermore, the approaches will take into account that it is not only humans who can store and apply knowledge but that technological artifacts (e.g., intelligent agents, ambient technologies) can also support humans in their tasks. 2. For individuals, groups, and organizations in Europe, it is still hard to get an overview of all the possible formal and informal knowledge resources, units of learning, programs, and learning networks that are available and to identify the most appropriate for their needs. To this end, TENCompetence will provide tools to support individuals, groups, and organizations in Europe to ind the best solution for their formal or informal learning problem. The tools will connect existing repositories with learning and knowledge resources, and address present technological, organizational, practical, language, Intellectual Property Rights, and business model problems. Furthermore, the tools will provide suficient support for users in inding the best solution to their learning needs, given their prior knowledge, preferences, and situational circumstances. 3. The pro-active sharing of knowledge and learning resources is a major problem. For a variety of reasons, people are not able to (or do not want to) share their knowledge and other resources. To this end, TENCompetence will provide policies and software agents that support the pro-active sharing of knowledge and learning resources. The application of the principles of social exchange theory to the sharing of knowledge objects and learning objects will result in policies and tools in conjunction with the development of new business models that suit the needs of the different service providers within the ield. 4. For an organization in Europe, it is still hard to assess the competencies of applicants, employees, and learners who have studied and worked in a variety of settings. To this end, TENCompetence will provide models and software tools to assess the competencies of individuals, groups, and organizations in an exchangeable way. An
238 Koper & Specht
interoperable mechanism to express and exchange learners competencies will be developed, especially for informally acquired knowledge and competencies. Core aspects of such a mechanism are methods and technologies for interoperable person and group portfolios and personal development plans, as well as interoperable competence deinitions. 5. The availability of support is crucial for effective task performance. Current e-learning and knowledge management environments provide too little effective support to the users in their various tasks. To this end, TENCompetence will provide software for the effective and eficient support of users who create, store, use, and exchange knowledge resources, learning activities, units of learning, competence development programs, and networks for life-long competence development. The software will support both a) learners and knowledge users involved in learning or applying a new skill or complex knowledge, and b) experts, trainers, and teachers who provide learning support services in order to increase their bandwidth. 6. Centralized models for the management of a network do not work in Europe because a). the market is not homogenous, being strongly competitive and culturally diverse; b) individuals and organizations who collaborate in life-long learning infrastructure want to maintain their autonomy and control as much as possible. To this end, TENCompetence will provide models and software solutions to establish a decentralized, self-organized, and empowered management model when using the TENCompetence infrastructure. The decentralized, self-organized, and empowered management model in the network for life-long competence development will be based on principles of self-organization and social exchange theory. 7. Although the three areas of knowledge management, human resource management, and e-learning share many common themes (e.g., the need for a holistic view of individuals formally and informally developed competencies, the beneits to be gained from social approaches to competence development), there has been little unifying work that integrates models and tools for competence development during learning and working and across a lifetime. To this end, TENCompetence will integrate isolated tools that are available in the ields of knowledge management, human resources management, and e-learning. The integrated tools will be lexible, support a range of pedagogies, incorporate the more innovative aspects of the Web, be open standards-compliant, and plug-and-play.
The TENCompetence Approach

In the following section, we will describe some of the steps toward a unifying infrastructure and the integration of different levels. The main challenges are the integration of different levels of knowledge resource, learning activities, competence development, and learning networks. The main goals of the project are: raising the awareness and enabling a life-long
Figure 1. Different levels of learning support and their relation in TENCompetence
learning perspective and the integration of formal programmes with social software, and informal learning and community building in learning networks.
Integrating Different Levels of Learning Support

TENCompetence is designed to solve the problems of life-long competence development described by providing an advanced infrastructure that integrates the different models and tools in the ield of knowledge resource sharing, learning activities, competence development programs and life-long learning networks. It selected the concept of competence development as its core, because it uniies all the approaches. In human resource management, knowledge management, training, and regular education the concept is used increasingly. An overview of the different levels of tools and communities can be seen in Figure 1. The relations between the different layers are explained in the following section. Knowledge resources are basically the containers that store the explicit knowledge for sharing purposes. Examples are learning objects, articles, books, software programs, informal messages, etc. On the level of knowledge resources, the project will integrate with other initiatives in the knowledge resource sharing and management area to circumvent cold-start
240 Koper & Specht
problems, using open standards (Advanced Distributed Learning, 2004; IMS Global Learning Consortium, 2004a) and protocols for federated search and access (GLOBE, 2005; Prolearn, 2005). The project develops models and methods to stimulate and organise pro-active creation, storage, search, retrieval, packaging, and quality rating of knowledge resources. Beside current ongoing efforts to manage and share knowledge resources (Littlejohn, 2005), the project especially also looks at the social aspects in the sense of social exchange theory and using a set of rating and recommending mechanisms allowing users of knowledge resources to provide feedback on their quality for the competence development network. Based on this, we also perceive a set of reward mechanisms and customized community policies as essential. These mechanisms will be customizable to stimulate the following of policies for speciic domains (i.e., in areas with signiicant knowledge resource coverage, the policies might encourage re-use whereas in areas needing investment, the policies would encourage new resource development). Recent research in technology-enhanced learning has been dominated by learning objects, and the shrink-wrapping of content for delivery in different contexts is becoming mainstream e-learning practice. However, there is a growing feeling that while reusable learning objects are valuable, they do not lead to learning, education, and training as such. A counter approach builds pedagogical processes on top of the learning and knowledge objects. This new learning activities-based approach does not oppose the learning objects approach but integrates it with a higher-level layer. Learning activities in this sense are the designed or performed activities of a person that are directed at the attainment of a (explicit or implicit) learning objective. Designed learning activities are called units of learning (UoLs), such as courses, workshops, lessons, and so forth. A unit of learning adds a learning design to the knowledge resources; they add pedagogical aids like study tasks, tutoring, mentoring, monitoring communication services, feedback, formative, and summative assessments. TENCompetence work in this area will integrate and extend several existing initiatives around Europe to ensure that the beneits of the learning activities approach are apparent and its adoption is eased. Fundamentally, the project will ensure that the focus of technology-enhanced learning falls upon innovative approaches to competence development (e.g., learning in communities) rather than underlying technological infrastructure (Koper & Tattersall, 2005). On the one hand, a TENCompetence extension is necessary that links learning activities and knowledge resource sharing by integrating the creation of learning designs and the access to shared repositories. On the other hand, learning activities have deined learning objectives that can be linked and classiied according to standardized ontology-based competence descriptions. The UNFOLD (UNFOLD, 2004) project has already grown an active community for using and developing IMS Learning Design which the project will build on. TENCompetence will create a series of components that are easy-to-use, standards-based and open source, enabling users to create, store, use, support and exchange learning activities and units of learning. Furthermore, the extension of existing assessment standards like IMS-QTI (IMS Global Learning Consortium, 2004c) and competence assessment models will be integrated with learning designs to steer the individual learning process and the dynamic selection of knowledge resources based on competence proiles and individual preferences considering learning paths. Competence development programs crucially depend on a number of services and components. First, the competence records of learners cannot be treated as clean slates. As they
possess prior competencies at certain levels, this requires a positioning service. Second, the learners personal competence development plan needs to be translated into programbound learning activities, identifying those that are relevant to the plan and those that are not (learning path service). It is likely that several routes lead to the competence development goal as speciied in the personal development plan. Therefore, third, a navigation service (Janssen et al., in press; Tattersall et al., 2005) is required that provides the learner with a personalized recommendation for a route. Personalized recommendations in the research on adaptive hypermedia systems is often based on individual user models, but more and more social navigation support mechanisms become prominent and commercially successful. Fourth, while carrying out a learning activity, the learner is likely to need human help at some point. In this case a learner support service will ind related peers and tutors. Finally, a learner who has acted upon the recommendations of the allotment and navigation services, who with the help of tutor or peer support has achieved the goals speciied in his or her personal development plan, in the end will want to know at what level of competence he or she performs. A performance assessment service (posterior assessment) should spring into action to accomplish this and close the cycle to a new turn around to the next competence target proile. Competence development programs are formal or informal collections of learning activities and units of learning that are used to build competence in a certain discipline or job. Depending on the competencies to be built, these programs can be small or quite extensive (e.g., a masters program). In addition to formal programs offered by institutions, it is also possible to store and share learning routes and paths that are the result of exploratory behavior and exchange them among the users. In order for competencies to be used effectively, Europe clearly needs to go beyond the current syntactic approach, providing not only a systemic but especially a meaningful semantic way to contextualize and match competence data and related learning material for individuals and teams alike. For networks to meaningfully use competence data, a dual articulation model is proposed. A reusable competence deinition captures the part of competence information that may be reused for more than one person in one or more contexts and possibly with different metrics into simple and existing standard formats such as IEEE Reusable Competency Deinitions (IEEE, 2004), IMS Reusable Deinition of Competency or Educational Objective (IMS Global Learning Consortium, 2004d), or HR-XML (HR-XML Consortium, 2006). This context neutral information can be published in public competence registers. To describe a learners existing competence repertoire (accredited or non-accredited), a common language needs to be used. This language should help, for any speciic competence, to establish at what level a learner possesses the competence in question, whether a learner strives to acquire it, and whether a particular competence program caters to its acquisition. Research needs to be done into the questions of how semantic Web tools may be used. Last but not least, the competencies that users have to acquire for different jobs and tasks change all the time. It is essential that these competencies are monitored in the ield and that a mapping mechanism is available to map older deinitions of competences to the changed new ones. Networks for life-long competence development are deined as collections of programs. Networks can be deined within a single country or institute (e.g., all the programs offered by a university or training company) or across countries and institutes. Networks are, however, deined in one (larger) domain of knowledge like economy, digital media, manage-
242 Koper & Specht
ment, music, and so forth. In this activity, TENCompetence develops tools to support the interoperability of formal and informal competence development programs from different providers and from different countries.
Life-long Competence Development: Todays Missing Links

Competence development is a hot topic in different research communities, and different motivations have been important to discuss the topic. In e-learning, often an individual perspective is taken on competencies, skills, and knowledge of the individual. In the area of knowledge management organizational backgrounds, return on investment (ROI) analysis and controlling issues were the main motivation to discuss about organizational competence development and strategic competence management. Speciic requirements that come from the ield of life-long learning are often not suficiently relected in todays approaches: In life-long learning, the learner is put centre-stage, meaning that the learner is selfdirected and can perform different formal and informal learning activities in different contexts at the same time. For instance, on the same day, a person can have a jobrelated training course at work, learn a new language after work from a teacher in the neighborhood, read texts and search the Internet for information, and do a masters degree in psychology at a university in the evening. In life-long learning, participants in any given domain of knowledge have different levels of competence, varying from novices to top experts, from practitioners to researchers and developers. Taking the long-term perspective, the learners will come from a variety of different backgrounds and have different knowledge. To support life-long learning adequately, it is necessary to maintain a record of an individuals growth in competency in a persistent and standard way to ensure that learners can search for new learning facilities that it and extend their current knowledge. Current developments of standards in the area of e-portfolio relect this need (IMS Global Learning Consortium, 2004b).
All three requirements put the learner and an informal and formal competence model at the centre of learning processes. In this sense, learners need competence as the enabling factor for performing tasks in everyday life (Cheetham & Chivers, 2005). E-learning platforms have been installed in a large number of schools, universities, and companies, but there are at least three important factors that are driving the need for a new generation of learning technologies that are not taken into account in the irst generation described above: 1. New Internet technologies can support completely different kinds of learning than was possible only a decade ago. Ubiquitous access to information spaces and phenomena
like podcasting, blogging, or wikis are just some examples. 2. Our current knowledge-based society demands a different attitude toward learning and the provision of learning: learning is no longer tied to the school and university context but is life-long, more integrated into work and other life contexts, and combines different formal and informal learning activities The introduction of social-constructivist principles of learning has brought about fundamental changes in the way learning and teaching is conceived. Examples of these new social approaches are collaborative learning, the use of learning communities, use of authentic tasks, scaffolding principles, social tools (friend of a friend, social presence, wikis, syndicated blogs), legitimate, supported peripheral participation in real practice, and new assessment methods (competency-based assessment, peer assessment, collaborative rating techniques, 360 degrees feedback, etc.) based on the principle that assessment should be authentic and integrated into the learning process. These new approaches focus more on the process of learning (the learning activities) than on the content.
3.
Those factors stress the need for an integration of formalized and structured learning support based on pedagogical models with the collaborative and socially driven approaches for supporting informal learning. Developing competences is taken as a unifying concept between formal and informal learning processes in TENCompetence. As can be seen in Figure 1, the competence layer builds the link between formalized learning processes and knowledge resources and a learning network in which mainly informal and social processes for competence development play an important role. TENCompetence is designed to solve the problems of life-long competence development described above by providing an advanced infrastructure that integrates the different models and tools in the ield of knowledge resource sharing, learning activities, competence development programs, and life-long learning networks. The main aims of the infrastructure are that it solves the problems with life-long learning mentioned above, and that it is sustainable, widely accessible (free of charge), and userfriendly. We envisage that such an integrated infrastructure can provide a tremendous push toward further European integration and collaboration and the support for the European knowledge society. It can be used at all levels of learning: primary education, secondary education, higher education, continuing education and training, company training, job accreditation, adult education, workshops, conferences, and all kinds of informal learning. The broad scope of the platform is needed to attain the right critical mass to support life-long learning in Europe. In the current approach, the project analyses the requirements for such a platform from different perspectives. First, the needs of users in special domains are analyzed for irst pilot applications in which focused aspects of the whole infrastructure will be evaluated. Second, future scenarios and use cases are developed to identify the gaps between todays practice and the actual future need for life-long learning. Third, existing open source tools on the different levels described above are analyzed and a common infrastructure is deined with interfaces that allow the technical integration; those will include: 1. An extension of IMS QTI (IMS Global Learning Consortium, 2005), integrating new
244 Koper & Specht
assessment methods suitable for life-long competence development. 2. 3. A learning path speciication to describe the structure of programs (curricula, training programs, personal development programs, etc.) in an interoperable way. A learning services connector speciication that enables the runtime connection between communication and collaborative services (forums, chats, shared whiteboard, etc.) to learning design engines.
Building Learning Networks

As described above, TENCompetence will deliver an integrated technical framework for enabling life-long learning; nevertheless core questions for working communities and learning networks have often been identiied in the social aspects of how to get technology in use and how to create a stimulating environment that challenges and fosters the growth of a learning network. TENCompetence in this sense focuses on questions of : How can a distributed network of actors and artifacts be created that optimizes the emergence of effective, eficient, and attractive life-long learning in its participants and the network as a whole? What principles, theories, models, methods, rules and technologies govern such a network? What are the beneits and its restrictions? What is the critical size in terms of actors/agents and artifacts for emergence to become possible? What are reasonable relationships between learning networks?
In the following sections we will focus on two important backgrounds we consider essential for answering those questions, that is, social exchange theory and self-organization theory.
Social Exchange Theory

A main critical point in building software that is used and communities that become active learning networks is the engagement in the sense of active participation and contribution of the individual. The individual satisfaction and perception of effectiveness in that sense is closely related to the commitment of the individual to contribute and actively participate. Social exchange theory provides a theoretical framework with guidelines to increase active participation and decrease lurking. This theory (Constant, Kiesler, & Sproull, 1994; Thibaut, 1959), derived from economics rational choice theory, suggests that there is a
relationship between a persons effect (satisfaction with a relationship) and his commitment to that relationship, that is, his willingness to contribute and share knowledge. Social exchange theory argues that individuals evaluate alternative courses of action so that they get the best value at lowest cost from any transaction completed. Hall (2001) suggests four main reasons that could underlie the motivation and commitment of community members to share knowledge: 1. 2. 3. Personal need, or anticipated reciprocity: Learner has a pre-existing expectation that he will receive actionable and useful information in return. Reputation: Learner feels he can improve his visibility and inluence to others in the community, for example, leading to more work or status in the future. Altruism, or the perception of eficacy of the community in sharing knowledge as a public good, especially when contributions are seen as important, relevant, and related to outcomes. (Tangible) reward: Learners negotiate to get some kind of more tangible asset (inancial reward, bond, etc.) in return.
4.
Likewise, other authors make distinctions between individual (personal need, reputation, reward) and interpersonal factors (altruism) (Deci & Ryan, 1985), hard (e.g., access) and soft (e.g., satisfaction) rewards (Hall, 2001), quantitative and qualitative gain, or intrinsic and extrinsic factors (Hemetsberger & Christian, 2004). Many researchers warn us against introducing (more tangible) extrinsic rewards as incentives, since they might destroy the public good factor. When knowledge is considered a public good (McLure-Wasko & Faraj, 2000), knowledge exchange is motivated by moral obligation and community interest (altruism) rather than by narrow self-interest (personal need, reputation). A public good is a commodity that can be provided only if group members contribute something toward its provision; however, all persons may use it (Komorita & Parks, 1995). Greater self-interest reduces knowledge sharing (Constant, Kiesler & Sproull, 1994), and people are less likely to use collaborative technologies to share information perceived to be owned by the organization (Jarvenpaa & Staples, 2000). Introducing tangible rewards in return for the provision of public goods promotes self-interested behavior, reduces intrinsic motivation, and destroys the public good. The danger is that individuals may appear to be contributing something, but what is not contributed is more signiicant. This would appear to be more signiicant when reward mechanisms are in operation (Von Krogh, 1998). Incentives for knowledge sharing should match the spirit of what is to be achieved (Sawyer, Eschenfelder, & Heckman, 2000). Besides intrinsic and extrinsic factors that motivate individual community members to share knowledge, Hemetsberger & Christian (2003) identiied other key elements of exchange processes and relations in a community: 1. Common goals and values: Communities gather around a common interest, a passion (Kozinets, 1999). Values are important moderators for solidarity in relationships (Anderson, Challagalla, & McFarland, 1999). Communal relationships: On a micro level, intimate communal ties can be distin-
2.
246 Koper & Specht
guished (Rheingold, 2000). On a meso level, there can be a web of personal relationships in cyberspace or trusting relationships between two or more partners (Grnroos, 1999). All these elements of social exchange play a complex and often intertwined role in communities. The right combination of those elements can constitute a powerful mechanism for a self-sustaining system for the exchange and reuse of learning objects. All of them deine the willingness (motivation) of the members to exchange information and support in a community.
Self-Organization Theory
The management of a large distributed network such as that proposed here can be very complex, and consequently a decentralized management approach making use of self-organization is desirable. Another reason for introducing self-organization in life-long competence development networks is to increase the eficiency of the support structure in the networks. Active learners produce work such as written contributions to discussions and research reports. These have to be read, reacted to, or reviewed. In a traditional setting, there is a danger that these tasks will be assigned solely to the teacher, trainer, or expert, whose workload will then increase considerably. The assumption of TENCompetence is that the application of self-organization theory can be a foundation for the establishment of eficient systems with a minimum of planning and control, while maintaining maximum lexibility to adapt to the users needs. This will reduce overhead costs for maintenance, planning, control, and quality. This assumption is based on research into self-organization theory (Maturana & Varela, 1992; Varela, Thompson, & Rosch, 1991), which is grounded in complexity theory (Waldrop, 1992) and studies the characteristics of the social organization of communities that emerge from the interactions of lower level users. The management and application of policies in a network or community is termed sociability (Preece, 2000). Sociability governs social interaction in a community. It cannot be controlled directly, but can be supported by carefully communicating the purpose and policies of the community. Preece (2000) identiies several policies in a community: joining or leaving requirements; by-laws; codes of practice for communication; rules for moderation; issues of privacy and trust; practices for distinguishing professionally contributed information; rules for copyright; and democracy and free speech in the community. We identify the policies in networks in terms of: objectives and values; terms of use; membership/role policies; standards and quality policies; and reward policies. Using this perspective, the organization of life-long competence development networks can be realized by installing technical facilities that enable distributed interactions among participants directed at a common purpose (e.g., competence development in a disciplinary ield), governed by policies that stimulate participants to learn, share knowledge and support each other.
Building Learning Networks in TENCompetence

TENCompetence explores different approaches to stimulate and integrate formal and informal learning processes: Develop and test methods and tools to set up acompetency observatory to monitor and capture the competencies that have to be acquired in different professional and academic ields. Allowing the learner to locate him/herself and other stakeholders to contribute learning services. Develop and test components to provide an integral overview of all the possible formal and informal programs that are available for the users to develop their competencies. This requires a consistent mapping of contents, competences, and learning objectives integrated with approaches from the semantic Web and competence ontologies. Develop and test models and software agents to stimulate the pro-active sharing of programs and the underlying learning activities and resources among users. Develop and test policies and software agents to stimulate the pro-active sharing of learning and knowledge resources among users. The software agents must be compliant with the technical standards and architectural constraints. The task will identify services and policies that foster social exchange in networks for life-long competence development. It will identify reward mechanisms, member participation, terms of use, quality standards, and procedures, including their ontological requirements. It will describe and prepare experimental test beds to test the effects of policies and social exchange services on sharing behavior amongst members and the management of the network. Develop and test models and tools to manage the network whilst maintaining a maximum of autonomy and control of participants. Self-control of participants in learning networks is essential also in the sense of active participation and the perception of self-eficiency.
Recent experimentation in learning networks has shown that incentive mechanisms like the adaptive introduction of extra (bonus) material based on contributions can increase both active and passive participation in learning networks (Hummel et al., 2005). Beside the individual activity, adaptive rewarding mechanisms could also take into account the current needs of the community (which contributions are currently needed, i.e., new items, rating of items, Cheng & Vassileva, 2005) and the style and quality of individual contributions in the past, that is, fewer high-quality contributions or more lower-quality contributions in a shorter time. We perceive adaptivity and personalization as key issues for implementing mechanisms to foster and increase activities in life-long learning networks. Currently, an integrated approach that allows rewarding and incentive mechanisms on different levels of sharing and exchange (knowledge resources, learning activities, competence development programs) is researched in the TENCompetence project. The activities in this sense will identify from a perspective of self-organization theory the policies that are known, on an empirical basis, to
248 Koper & Specht
be associated with successful communities. From a perspective of self-organization theory the project will identify policies that are associated with successful community building. These include reward mechanisms, member participation, terms of use, quality standards and procedures, social exchange, management, and policy change in communities in order to identify policies and policy changes. The participants will be provided with recommendations and speciications of the appropriate policies for the implementation of competence development networks. Empirical tests of the effects of introducing these policies to a network will be carried out using ield experiments and possibly simulations, and the results validated in ield trials. TENCompetence will pilot applications in different domains also to develop sustainable and validated approaches that hold in different application areas with different policies and support for life-long learning. Currently four basic pilots are foreseen in the project, which will be extended with pilots from associated partners. The four pilots are: training for digital movie production, training for continuous education of health workers, Unesco-IHEInstitute for Water Education, and The Antwerp Life-long Learning City. Based on the integration of current learning support tools, the pilots will identify domainspeciic customizations necessary and the validity of the general domain model developed in TENCompetence.
Conclusion and Outlook

The central need addressed by TENCompetence is to provide ubiquitous and life-long adapted access to facilities that support the creation, storage, use, and exchange of formal and informal knowledge and learning resources. The project started in December 2005 and will run for four years, where within the irst 18 month already pilots are set up and evaluation of the core elements and methods described above will be done. The integration of the different layers of granularity is an essential need to work toward a holistic view on life-long learning. The usage of knowledge resources and their integration in pedagogically structured learning activities has already been demonstrated in recent research projects. The integration of competence development programs and learning activities is mostly solved proprietarily in different solutions but rarely standardized descriptions for competence models are used to build and exchange those models from a life-long learner perspective. Furthermore, the integration of learning networks or communities of practice are mostly implemented separately from content-driven e-learning solutions and do not allow for the integration of formal and informal learning activities. By combining the described layers in TENCompetence, we therefore foresee giving the learner a new perspective on learning and competence development by reusing and integrating existing technologies and solutions in an open source life-long learning infrastructure.
To this end, TENCompetence (http://www.tencompetence.org) will support individuals, groups, and organizations in Europe in life-long competence development by establishing the most appropriate technical and organizational infrastructure, using open-source, standards-based, sustainable, and innovative technology. Additionally, questions of combining informal, non-formal, and formal learning processes, and the different perspectives become an essential issue in todays computer based learning support. Therefore, TENCompetence is exploring mechanisms for supporting users in positioning and navigation in learning networks on the one hand but also provides approaches to stimulate active participation of different stakeholders in learning networks.
Acknowledgments
The authors would like to thank the management and staff of the Schloss Dagstuhl International Conference and Research Center for Computer Science for providing a pleasant, stimulating, and well-organised environment in which to write this paper. The TENCompetence Project is funded by the European Commission under the 6th IST Framework Project Number IST027087.
References
The date of download for electronic references September 4, 2006. Advanced Distributed Learning. (2004). SCORM 2004. Retrieved from http://www.adlnet. gov/index.cfm Anderson, W., Challagalla, G., & McFarland, R. (1999). Anatomy of exchange. Journal of Marketing Theory and Practice, 7(4), 8-19. Cheetham, G., & Chivers, G. (2005). Professions, competence and informal learning. Northampton: Edward Elgar Publishing. Cheng, R., & Vassileva., J. (2005, July 18-22). Adaptive reward mechanism for sustainable online learning community. Paper presented at the AI in Education, Amsterdam. Constant, D., Kiesler, S., & Sproull, L. (1994). What is mine is ours, or is it? Information Systems Research, 5, 400-422. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York: PlenumPress. GLOBE. (2005). Global Learning Objects Brokered Exchange (GLOBE) alliance. Retrieved from http://globe.edna.edu.au/globe/go Grnroos, C. (1999). Relationship marketing: Challenges for the organization. Journal of Business Research, 46, 327-335.
250 Koper & Specht
Hall, H. (2001). Input Friendly Intranets: Motivating Knowledge Sharing Across Intranets. Journal of Information Science, 27(3), 139-146 Hemetsberger, A., & Christian, R. (2004). Sharing and creating knowledge in open-source communities: The case of KDE. Proceedings of the 5th European Conference on Organizational Knowledge, Learning and Capabilities (OKLC). Retrieved from http://dissertation.martinaspeli.net/papers/communities-of-practice/hemetsberger-andreinhardt-2004-sharing-and-creating-knowledge-in-open-source-communities HR-XML Consortium. (2006). Completed speciications: Competencies. Retrieved from http://ns.hr-xml.org/2_4/HR-XML-2_4/CPO/Competencies.html Hummel, H., Burgos, D., Tattersall, C., Brouns, F., Kurvers, R., & Koper, R. (2005). Encouraging contributions in learning networks using incentive mechanisms. Journal of Computer Assisted Learning, 21, 355-365. IEEE Institute of Electrical and Electronics Engineers. (2004). IEEE Reusable competence deinition. Retrieved from: http://ltsc.ieee.org/wg20/. IMS Global Learning Consortium. (2004a). Content packaging speciication. Retrieved from http://www.imsglobal.org/content/packaging/index.html IMS Global Learning Consortium. (2004b). IMS eportfolio speciication. Retrieved from http://www.imsglobal.org/ep/index.html IMS Global Learning Consortium. (2004c). IMS question & test interoperability speciication. Retrieved from http://www.imsglobal.org/question/ IMS Global Learning Consortium. (2004d). IMS reusable deinition of competency or educational objective speciication. Retrieved from http://www.imsglobal.org/competencies/index.html IMS Global Learning Consortium. (2005). IMS question & test interoperability speciication. 2006, from http://www.imsglobal.org/question/ Jannsen, J., et al., (in press) Self-organizing navigational support in lifelong learning: How predecessors can lead the way. Computers and Education. Jarvenpaa, S., & Staples, D. (2000). The use of collaborative electronic media for information sharing: An exploratory study of determinants. Journal of Strategic Information Systems, 9, 129-154. Komorita, S., & Parks, C. (1995). Interpersonal relations: Mixed-motive interaction. Annual Review of Psychology, 46, 183-207. Koper, R., & Tattersall, C. (2005). Learning design: A handbook on modelling and delivering networked education and training. Berlin/Heidelberg: Springer. Kozinets, R. (1999). E-tribalized marketing? The strategic implications of virtual communities on consumption. European Management Journal, 17(3), 252-264. Littlejohn, A. (2005). Reusing online resources: A sustainable approach to elearning. London: Kogan Page. Maturana, H., & Varela, F. J. (1992). The tree of knowledge: The biological roots of human understanding. Boston: New Science Press. McLure-Wasko, M., & Faraj, S. (2000). It is what one does: Why people participate and
help others in electronic communities of practice. Journal of Strategic Information Systems, 9, 155-173. Preece, J. (2000). Online communities: Designing usability, supporting sociability. Chichester, UK: Wiley. Prolearn. (2005). Prolearn project Web site. Retrieved from http://www.prolearn-project. org Rheingold, H. (2000). The virtual community: Homesteading on the electronic frontier. Cambirdge, MA: MIT Press. Sawyer, S., Eschenfelder, K., & Hexkman, R. (2000). Knowledge markets: Cooperation among distributed technical specialist. In Srikantaiah, K., & Koenig, M. (Eds.), Knowledge management for the information professional. (pp. 181-204) Medford, NJ: Information Today. Schn, D. A. (1983). The relective practitioner: How professionals think in action. London: Maurice Temple Smith. Tattersall, C., et al., (2005)Self Organising wayinding support for lifelong learners. Education and Information Technologies. 10(1-2), 111-123 TENCompetence Consortium. (2005). TENCompetence project Website. Retrieved from http://www.tencompetence.org/ Thibaut J., K. H. (1959). The social psychology of groups. New York: Wiley. UNFOLD Project (2004). UNFOLD project Web site. 2004. Retrieved from http://www. unfold-project.net:8085/UNFOLD Varela, F. J., Thompson, E., & Rosch, E. (1991). The embodied mind: Cognitive science and human experience. Cambridge, MA: MIT Press. Von Krogh, G. (1998). Care in knowledge creation. California Management Review, 40(3), 13-153. Waldrop, M. (1992). Complexity: The emerging science at the edge of order and chaos. New York: Simon & Schuster.
252 Koper & Specht
Section III: Competencies and the Semantic Web
An Ontology-Based Competence Management Model 253
Chapter XII
An Ontology-Based Competence Management Model to Support Collaborative Working and Organisational Learning
Jos Braga de Vasconcelos, Universidade Fernando Pessoa, Portugal Chris Kimble, University of York, UK
Abstract
This chapter describes an ontology-based competence management model that can be used to support distributed collaborative working and facilitate organisational leaning. The chapter begins by examining the concepts of collaborative working and organisational learning in the context of knowledge management and competence management systems. It describes the different forms of knowledge that can be represented in such systems and uses this to describe an ontology-driven approach to competence management. An outline design for an ontology-driven competence management system and the related content management approach is then outlined using material from earlier work on a group memory system; the competence ontology and expert annotation technique used in group memory system is described in detail. The chapter concludes with some discussion concerning the role of such systems in knowledge-intensive industries and some directions for future research.
254 Braga de Vasconcelos & Kimble
Introduction
Besides labour, capital, and land, knowledge has been recognised as an important factor for productivity in organisations (Abecker et al., 1999). Knowledge-intensive companies employ highly skilled people, knowledge workers, who constantly face problem-solving tasks that require the eficient utilisation of existing sources of knowledge and information (Alvesson, 1995). In such an environment, people, the knowledge they hold, and the interactions between themin short, organisational knowledgeare seen as crucial for effective business performance (Brown & Duguid, 2000). However, since the 1980s, many organisations have taken steps to outsource and downsize in an effort to remain competitive. Outsourcing, downsizing, and programmes of planned redundancy all mean that, as people leave, they take with them a valuable stock of knowledge about the way an organisation works on a day-to-day level (DeLong, 2004). This loss of knowledge inevitably limits the long-term effectiveness of an organisation, as those who remain may be unaware of key resources that lie hidden in various knowledge repositories within the organisation (Dzbor, Parlic, & Parlic, 2000). Knowledge-intensive organisations (KIOs), in particular, rely heavily on making effective use of such knowledge. By deinition, knowledge-based tasks such as recognising patterns in organisational behaviour and dealing with abstraction, ambiguity, and uncertainty, form a large part of their corporate activity (Alvesson, 1993). In practice, much of this work is done through exploiting a constantly changing and evolving network of relationships between people, sources of information, and organisational needs. Organisational groups in such organisations need to create mechanisms to elicit innovation, ind sources of information, manage skills eficiently, and gather ideas and suggestions in order to do their work. In other words, to work effectively in a KIO, groups need to be able to work collaboratively. Collaborative work is work undertaken as part of a group activity that is directed toward a shared goal or common purpose. In recent years, collaborative working has tended to become synonymous with the use of technology to support the operation of virtual teams (Kimble, Li, & Barlow, 2000), communities of practice (Kimble & Hildreth, 2005), or networks of practice (Teigland & Wasko, 2004). Much of this literature, however, focuses on simple task groups whose role is to fulil a particular task; collaborative work in a KIO is more than a group of individuals working in isolation. For collaborative work to be effective, there must be a degree of esprit de corps. A balance of dealing with factual content, relationships, and the coordination of a central process is required. Social aspects such as a shared social context, a feeling of trust, and a human interest in each other need to be balanced against the more process-orientated aspects such as the planning of work and the scheduling of activities (Sharratt & Usoro, 2003). Tuomi (2002) claims that modern knowledge management (KM) systems have now moved away from their roots as dumb databases and have begun to encompass broader concepts such as computer-supported collaborative work, computer-mediated communication, groupware, collaborative systems, and organisational learning. Following Heijst, Spek, and Kruizinga (1997), we view the concept of organisational learning as an organisational process where three forms of learning can occur in parallel: individual learning by reusing lessons learned from previous experiences, learning through communication (or group learning), and learning through the use of an information repository. The competence management (CM)
system we describe is one approach to facilitating this last form of organisational learning. It is designed to act as a single front end for the human user, providing a coherent conceptual view based on their changing information needs. The emphasis in this chapter is on the competence modeling task that creates consistent descriptions of corporate competencies. This is seen as part of a KM initiative within an organisation with the goal of the deinition of a consensual model for competencies. This remainder of this chapter is organized as follows: the next section briely describes the theoretical standpoint we take in this work with regard to KM and CM. In it, we describe the relationship between KM systems and CM systems and outline different forms of knowledge that can be represented in such systems. This section concludes with a theoretical knowledge taxonomy, which draws this information together. The following section presents an outline design for an ontology-driven CM system and describes the related content management approach. Earlier work on a group memory system (GMS) is briely described in order to illustrate certain aspects of the design. The CM system is described in outline; the problem of representing expert knowledge in annotations and the competence ontology itself are described in greater detail. Finally, the chapter concludes with an examination of some of the practical problems of the effective integration of people-based KM techniques with technologically based learning infrastructures.
Knowledge Management and Competence Management

Knowledge management focuses on techniques to manage a common base of organisational knowledge and allows heterogeneous organisational groups, functions, and communities to coordinate their efforts and share knowledge across time, function, discipline, and taskspeciic boundaries (DAdderio, 2003). As people transform data, information, and experiences into effective corporate knowledge, the management of individual competencies will become more important in KIOs (Michellone & Zollo, 2000). According to Bahrami (1996), developing versatile employees and leveraging their capabilities (competencies) in order to cope with different corporate needs is a matter of pivotal importance for KIOs. The employees competencies in the form of technical and cognitive capabilities mainly affect the companys knowledge creation ability. Competence management is also concerned with the better usage of human skills and knowledge. In knowledge-intensive companies, most daily work tasks require professional expertise and the management of a large body of knowledge. As a knowledge-intensive company is a competency-based company too, it is no longer possible to separate the two ields (Michellone & Zollo, 2000). A seminal reference in competence management is the People Capability Maturity Model released by the Software Engineering Institute (Curtis, Kellner, & Over, 1995). This model was initially designed to help software organisations to focus on the improvement of the capability of their workforces. Corporate maturity in the people capability maturity model represents an organisations ability to improve the knowledge, competencies, and related skills of its workforce and align performance with the organisations business objectives.
Consequently, this chapter uses a KM approach to deine a CM system based on a coherent and structured web content management infrastructure where the CM system is seen as a step toward a larger KM system. This information layer is designed using an ontology-driven annotation framework to incorporate additional semantics into the existing organizational Web elements. The aim is to integrate less-tangible knowledge, such as employees competencies, contextual information, and related competencies with more formalised and structured knowledge stored in databases, procedures, manuals, books, and reports.
Representations of Knowledge in Knowledge Management Systems

Knowledge in an organisation is the collection of expertise, experience, and information that individuals and workgroups use during the execution of their tasks. Organisational knowledge is stored in individual minds or implicitly encoded and documented in organisational processes, services, and systems. In a KM initiative, a irst step should be an effective identiication, categorisation, and classiication of organisational knowledge. In order to manage knowledge as a resource, it is irst necessary to understand the characteristics of organisational knowledge and its different theoretical classiications and related types. The aim of this framework is to provide a deinition of the vocabulary and the related semantics for the identiication, categorisation, and classiication of knowledge assets within an organisation. Two general and related forms of knowledge taxonomy can be found in much of the literature on KM. The irst is the classiication of knowledge into tacit or explicit; the second is the classiication of knowledge into declarative, procedural, or heuristic. In this taxonomy, a third category has been introduced to represent the location of knowledge in the organisation: knowledge may be classiied into either individual or group (collective) knowledge. Within these three categories, it is possible to it forms of domain-speciic knowledge. The three categories are described briely below.
Tacit Knowledge vs. Explicit Knowledge

Tacit knowledge has been systematically investigated in the literature (cf., Polanyi, 1966; Reuber, Dyke, & Fischer, 1990, Sveiby, 1997). The nature of tacit knowledge was explored by Polanyi (1966) who described it as knowledge that is experientially acquired. Polanyi used the phrase we know more than we can tell to describe what he meant by tacit knowledge. Tacit knowledge is a kind of knowledge that is very dificult to articulate because it is either too complex or simply because it is internalised in peoples minds. The internalised knowledge of an individual and the culture of an organisation can be seen as examples of tacit knowledge. Similarly, knowledge in the form of project experiences, task heuristics, and human competencies that are hard to capture and externalise are also examples of tacit knowledge within the work environment. Explicit knowledge, on the other hand, refers to knowledge that is transmittable in formal and systematic language. It is the kind of knowledge we can easily share and articulate
because it is relatively independent of any particular individual or organisational group. Examples of explicit knowledge are information about physical aspects of an organisation, such as documents, reports, articles, software code, pictures, and other kinds of reusable technologies. Nonaka and Takeuchi (1995) described speciic modes of knowledge transformation within an organisation using tacit and explicit knowledge; they are: (a) socialisation that involves conversion from tacit knowledge to tacit knowledge; (b) externalisation that involves conversion from tacit to explicit knowledge; (c) combination that involves conversion from explicit to explicit knowledge; and (d) internalisation that involves conversion from explicit to tacit knowledge. The knowledge externalisation process has a particularly important role in this research. This process concerns the elicitation, capture, and translation of tacit knowledge into a computer system in the form of an information repository. The effective management of elements of human competency is viewed as a form of knowledge creation within an organisation. Knowledge creation can be realised through the identiication, representation, capture, and externalisation of less-tangible corporate competencies within the organisation.
Declarative, Procedural, and Heuristic Knowledge

Declarative knowledge is related to the physical aspect of knowledge. It is the knowledge type required in order to know what, who, where, and when. It is essential in both interpreting and describing, from a certain viewpoint (conceptualisation), the physical features of the world. It is knowledge of objects (entities or events) and consensual facts about the world, for example, information about a given content area. The knowledge represented in a database conceptual schema is an example of declarative knowledge. Formal representations using irst-order logic predicates are also examples of declarative knowledge. Procedural knowledge is the knowledge required to accomplish a certain task: it provides a description of speciic actions required to complete a particular task. It derives from the intellectual skill of knowing how to do something. Conventionally, procedural knowledge uses declarative knowledge to describe actions in step-by-step sequences. Procedural knowledge allows the representation of the behaviour of a speciic domain. A description of actions/steps, business rules, constraints, and exceptions that constitute a particular operation or organisational task is an example of procedural knowledge. Finally, heuristic knowledge, which can be interpreted as a speciic type of tacit knowledge, as is hard to capture and externalise, describes the knowledge related with work experience and implicit reasoning. As meaning depends on the individuals experience, heuristic knowledge grows with work experience. Heuristic knowledge is generated by an internal process and uses both declarative and procedural knowledge to solve problems and consequently to answer the question why (Davis, Shrobe, & Szolovits, 1993). For example, heuristic knowledge can assist a particular troubleshooting task involving a speciic machine. In this context, heuristics gained during work experiences assist the resolution of future tasks.
Individual Knowledge vs. Group Knowledge

A third category of knowledge has been introduced to represent its location within an organisation. In the conceptualisation above, tacit knowledge is seen as a property of individuals. However, research has made it clear that a team of interacting individuals can have knowledge that transcends the knowledge of each of them individually. Walsh (1995) uses the term knowledge structure to describe a mental template that is used to give a complex information environment form and meaning. Knowledge structures are built on experience and are used to store data to allow for subsequent interpretation and action. Hence, individual knowledge is concerned with personal knowledge structures, while group knowledge is related to organisational knowledge structures. Similarly, Brown and Duguid (1991) use the example of photocopier repairmen and technical specialists to demonstrate that each without the knowledge of the other, each would have been unable to effect a repair. They argue that ... individual learning is inseparable from collective learning. The insight accumulated is not a private substance, but socially constructed and distributed. This third categorisation of group knowledge is an attempt to recognise this understanding of knowledge and incorporate both individual and group knowledge. We deine group knowledge as the knowledge and skills acquired collectively by individuals who have been exposed to similar job-related situations (Reuber et al., 1990). For example, organisational groups have part of their knowledge codiied in a form of speciic work practices or group metaphors that only the members of that group can understand. Such practices and metaphors are owned by a group of people rather than an individual; typically, they are the result of communication within a workgroup rather than imposed from outside.
Theoretical Knowledge Taxonomy

Figure 1 shows the different organisational knowledge types and their relationships. Based on this taxonomy, knowledge assets, such as undocumented corporate competencies and contextual information connected to working documents, are presented as less-tangible knowledge assets within the organisation. A goal of this research is to develop mechanisms to identify, represent, and distribute such assets among people and processes. The following knowledge assets are emphasised in this chapter: the notion of individual and group competencies and the tacit, contextual knowledge embedded in working information sources. The following sections discuss corporate competencies as a speciic example of organisational knowledge that needs to be managed and preserved.
An Ontology-Driven Approach to Competence Management

The term ontology has its origins in metaphysics and philosophical sciences. In this context, an ontology is used to explain the nature of the reality. There are at least a dozen of deinitions of ontologies in the computer science literature. One of the most recent ones says that an ontology provides the skeletal knowledge and an infrastructure for integrating knowledge
Figure 1. Theoretical knowledge taxonomy

Organisational Knowledge
Individual Knowledge Group Knowledge Tacit Knowledge Explicit Knowledge
Heuristic Knowledge
Cognitive Elements
Mental Models Viepoints Conceptualizations
Technical Elements
Know-How Skills
Procedural Knowledge
Declarative Knowledge
Structural knowledge about documents and explicit communication between individuals
Less Tangible Assets
Tangible Assets
bases at the knowledge level, independent of particular implementations (Devedzic, 1999). This deinition assumes a particular importance in this research in terms of the necessary mechanisms to represent, share, and reuse the existing corporate group memories. A more traditional deinition of an ontology states that it is a high-level formal speciication of certain knowledge domain: a formal and explicit speciication of a shared conceptualisation (Gruber, 1992). A domain conceptualisation is a particular and abstract view about real entities and events and their relationships. Formal refers to the fact that an ontology is a form of knowledge representation and has a formal software speciication to represent such conceptualisations, for example, an ontology has to be machine readable. Explicit means that all types of primitives, concepts, and constraints used in the ontology speciication are explicitly deined. Finally, shared means that the knowledge embedded in ontologies is a form of consensual knowledge, that is, it is not related to an individual, but accepted by a group (Benjamins, Fensel, & Perez, 1998). This approach is applied in the group memory system deinition (following section), where its design was based on consensual knowledge that belongs to a speciic organisational group. The construction of a reliable description of competencies, for example, a consensual model for corporate competencies within the workplace, is both an organisational and a design challenge. The approach addressed in this chapter is particularly focused on supporting the management of employees competencies in an eficient and structured way. Thus, the focus is on the management of corporate competencies in KIOs. To assist in the management of corporate competencies, experts annotations within an organisational group are represented (see next section) to enable the capture and management
of individual and group knowledge. This also aims to assist the management of repeatable patterns of action (Chiesa & Manzini, 1997), for example, the preservation and future reuse of knowledge embedded on daily workgroup activities. The following section describes how organisational knowledge in the form of competencies or other knowledge assets can be managed and preserved within a distributed collaborative work environment.
The Group Memory Approach

The proposed CM approach is based on a previous research work that has developed a GMS framework (Vasconcelos, Gouveia, & Kimble, 2002) for the management of corporate competencies. In this context, speciic modeling constructs, such as the expert annotation concept, are applied in this work. The effective management of the existing expert annotations within the workplace can enrich the related group memory and consequently the process of organisational learning. A group memory is considered a speciic example in a narrower scope of an organisational memory. To deine a group memory, we need to consider the work practices that include formal and informal communication that exist between people working together or geographically distributed. Group memory happens in the context of personal and managerial aspects that encourage people to share their work practices to improve organisational performance. The GMS described in Figure 2 focuses on corporate competencies, in particular human knowledge sources, their competencies, such as cognitive elements, and technical expertise, as well as project experiences and related heuristics. At a general level, it includes written and spoken communication, face-to-face meetings, shared information, and co-ordinated work.
Figure 2. An example of a group memory

Organisational group structure and behaviour
Workgroup and workflow characteristics Individual and group heuristics Levels of competence - Individual vs. Group - Skills and Areas of Expertise - Roles & Responsibilities
Factual Information
Issues Assertions Decisions Alternatives Viewpoints
Project Experiences
Past experiences Past critical decisions Lessons learned Best practices
Group Memory
The Competence Management System

The underlying case study to support this research approach is based on the business activities of a group of knowledge workers (project managers and requirements engineers) in a multinational telecommunications company. The company designs, produces and sells a variety of telecommunication products and network solutions worldwide. Speciically, the case study addresses the problem of managing the creation of competitive bids, known as the bid response process (BRP). This work requires the exchange of a large amount of knowledge between different groups, which are distributed across the whole of the company.
An Outline of the CM System

The competence model described by the competence ontology concerns the different levels of human knowledge that are acquired during process-oriented activities within the organisation. Competence elements such as speciic skills, areas of expertise, organisational roles and responsibilities, project experiences, and acquired heuristics during the execution of problem-solving tasks are addressed in this model. The long-term aim of this model is the design of a consensual competence taxonomy based on background knowledge and on the project experiences of the organisational group. As stated above, the competence model is represented in two design models: (1) a competence taxonomy that is a hierarchy of primitive competencies combined with the related hierarchy of application areas, and (2) a expert annotation model that aims at the representation of work routines, project experiences, and their context-based competencies.
Figure 3. CM system framework

1 - Conceptual layer (Ontologies at the knowledge level)
Competence type is_a part-of Competency part-of skill part-of Project experiences Tech compliance
Competence Information source Ontology

is-a Document
Information Ontology
Area of expertise instance-of Bid compliance instance-of is_a
is-a Free-text annotation part-of title part-of section has-a Storage format instance-of
PDF
Domain ontologies
has-a Document type is-a NTP is-a PSA
instance-of
System description X11
2 Formal layer
Formal representation - Frame-based orientation (Formal ontologies written in Ontolingua) Semantic-data model to support Group Web/Data services knowledge-intensive tasks Information navigation (process-oriented tasks) Information retrieval User interface
3- CMS application layer
As previously stated, the competence modeling task should be viewed as a irst step of a KM initiative within the organisation. A CM system is an important instance of a KM system for the organisation. In this context, the main organisational stakeholders should be included in this modeling task in order to create reliable descriptions of corporate competencies, such as the competence ontology, that will be deined, applied, reused, and developed during the execution of internal KM practices in the workplace. The main goal of a CM system is to enrich the related group memory. The main role of the group memory is to act as a shared conceptualisation to facilitate communication between group members and to act as a common schema for software applications. Thus, the group memory supports both structurally and dynamically, a shared representation of knowledge that allows a consensual understanding of shared purposes, roles, and competencies. Figure 3 illustrates the CM system framework. Effective capture and reuse of less-tangible knowledge assets within the organisation, such as contextual knowledge, may be achieved using a well-structured common and shared vocabulary. Such common and shared vocabulary can be represented with ontologies. Such high-level organisational knowledge description is seen as a set of deinitions of context-speciic knowledge representation primitives consisting of domain-dependent classes, relations, functions, and object constants. These primitive constructs can be applied and represented differently in several organisational domains but should have the same meaning for human users and system designers. The layers shown in Figure 3 are interpreted as follows. The conceptual layer (layer 1) represents the organisational knowledge in an informal way that can be interpreted by different CM developers, such as domain experts, knowledge engineers, and software engineers. This layer aims to create a shared understanding of the organisational knowledge in order to create a common platform language for the CM system. The creation of a common vocabulary facilitates communication in design and maintenance issues across people with different professional backgrounds. The formal layer (layer 2) enables the reuse of domain terms and constructs from other ontologies in order to facilitate future CM system maintenance tasks. The ontological formal descriptions provide a common vocabulary for knowledge engineers in order to develop further applications in this domain, such as an inference layer and the related reasoning mechanisms. This layer is essentially a format layer, where translators to multiple languages and environments can be hooked. The application layer (layer 3) uses the encoded domain knowledge. The knowledge encoded with ontologies can be used in different application systems within an organization. This layer is the interface with the users and can be tailored to different needs; as it is independent of a particular syntax and application model, changes at the previous layers do not have much impact at this layer.
Representing Expert Annotations

An expert annotation represents an individual knowledge asset that has been made explicit. The expert annotation is related to the competence taxonomy, and new competencies and ap-
Table 1. Sample of an annotation object instantiation

Factual information Author Date Subject Summary Product Technical category Technical speciication Standards compliance Nature Source Contextual information (task-based) Compliance task Annotation type Rationale Competence requirements (competence expressions) Edwards 22-Oct-2004 Compliance speciication //server/public-iles//bid-pt/summ21 Opt-telecom-DX Optical interfaces Input port signal ITU-ISO9901 ITU-ISO9905 Recommendation Previous bid experience Optical interface compliance statement Working annotation Identifying a problem Has-skill-of (design, optical interfaces) Practice-of (Integration, optical interfaces)
plication areas can arise during context-based expert annotation tasks. The expert annotation model was deined to facilitate expert knowledge sharing and reuse across the workgroup. CM systems and related expert annotation techniques is one of a number of approaches used in collaborative work. Expert annotations are simply seen as an instance of expert knowledge that has been externalised and recorded in a speciic context. Accordingly, an annotation object is a computer formalisation of an expert annotation. Annotations are not necessarily a formalised (design or programming) object, but for the CM system to work they must be capable of being embedded in an annotation object. Table 1 presents a practical example of an annotation object. When the work is co-located, the expert annotations that are passed between the group can be both formal and informal in nature; for example, they may be passed indirectly through simple observation of the behaviour of colleagues and co-workers. However, when the work of the collaborative group becomes geographically distributed or virtual, the expert annotations between group members must be passed through some electronic medium. This simple statement has two profound consequences. Firstly, it has been argued by Hildreth and Kimble (2002) that to be effective, a CM system must be able to deal with both hard (factual) and soft (contextual) knowledge simultaneously in a single object, that is, an expert annotation must be able to convey knowledge in an appropriate form for the work. Secondly, because all annotations must be electronic, it should be possible to know how these objects are distributed, and we should be able to map the relationships between annotation objects and recipients in a reasonably accurate manner. The main challenge to designers of such systems is, therefore, to ind a way to exploit the data that is available on the relationships between the work, the annotation objects, and the users of those objects (the context of the expert annotation) to design a system that is suficiently lexible and dynamic to encompass the range of potential uses that the actual (annotation)
objects may be put to. Ontologies provide one way of meeting this challenge. Ontologies provide a high-level form of clariication and representation, and can be portable across platforms and systems, and dynamic, evolving over time. Ontology design also promotes a knowledge-based system architecture by which stakeholders and developers can achieve a consensual model of the systems function (Vasconcelos, Kimble, & Rocha, 2003). The focus of this CM model is on the representation of less-tangible knowledge assets. These knowledge assets arise from expert annotations, such as undocumented competencies and contextual knowledge representing the context in which information sources are created and used within the organisation/workplace. The idea is to incorporate the expert annotations with the competence ontology and dynamically create and gather new knowledge assets that can be reused to assist upcoming problem-solving tasks.
The Competence Ontology

The competence ontology aims to create a consensual structure of competencies within the organisational group. At the corporate level, this task presents some dificulties concerning the creation of a speciic (and consensual) structure of competencies. For this reason, the proposed competence ontology should have a set of representative high-level terms and allow a dynamic evolution of its structure. As stated in the group memory view, the competence ontology represents a competence model where two design dimensions are addressed: a competence taxonomy consisting of primitive competencies and their application areas, and the design of expert annotations to represent work routines and their project experiences. Table 2 presents an overview of the competence ontology.
The Competence Taxonomy

The competence taxonomy is a particular view of the competence ontology. Personal and group competencies are formalised as special elements of the group memory, such as competencies, skills, roles, and project experiences. The competence ontology represents the knowledge and skills needed within the workplace to perform certain business functions of the organisation. A competency can be stated at a very abstract level. In this way, competencies can be decomposed to more granular competencies, such as competencies in designing network solutions or writing product technical documentation. A competency can also be decomposed in the skills required to perform the business processes underlying the business function for which the competency is maintained. The competence taxonomy includes two hierarchies (primitive competencies and application areas) illustrated in Figure 4. The main classes in this classiication are the competency and the entity. The class competency allows the representation of different levels of competency granularity through the creation of sub-classes of competencies. In the same way, the class entity represents the different application areas in which a speciic competency can be applied. These hierarchies are combined through a set of competence relations (e.g., has-experience-of) that allows the combination of terms between these hierarchies. Other competence elements,
Table 2. Competence ontology: Speciication overview

Title Domain Creation Date Modiied in Ontology Engineer Main goal Scope Knowledge sources Competence Ontology Workgroup competence taxonomy and a model for expert annotations February 2001 September 2005 JBV Represent a consensual workgroup structure of competencies to assist business process activities Portfolio Engineeringa process-oriented company workgroup Applications Users Domain experts Interviews results Questionnaire focusing on CM Active participation communication patterns wit h in the workgroup Technical documentation
GMS Prototype Corporate managers Industrial practitioners KM & OM researchers Proiling people Dynamic creation of competencies within a company workgroup Routing information needs to the right experts Project teambuilding analysis Expert annotation systems Enterprise Ontology (Uschold et. al, 1997) Agent Ontology O ( ntolingua Library) Ontology of capabilities (Stader & Macintosh, 1999)
Usage Scenarios
Reused ontologies Reusable ontologies Question
Product Ontology RelationConcepts (classes)
What competencies are needed to install a OPT-7X optical interface?
Has-skill-of(Install, Optical Interface) Competency, Install, Optical Interface

Relation Has-skill-of(Install, Optical Interface)
Domain questions (some examples) Classes speciied Relations Functions Instances -
Question What competencies are needed to install a OPT-7X optical interface?
Concepts (classes) Competency, Install, Optical Interface
Figure 4. Competence taxonomy (Adapted from Stader & Macintosh, 1999)

Hierarchy of Primitive Competencies Competency Technical Competency Analysis Requirements analysis Testing Assessment Review Engineering Design Development Integration Maintenance Cognitive competency Understanding Reasoning Creativity Physical competency Manual Competency Repair Removal Fitting Connection Install Replacement Project Competency Management Planning Organising Controlling Communication Business Competency Marketing ... Hierarchy of Application Areas Information Ontology Entity Process Activity reused from Task Document Procedure Product Ontology Standard Directive Recommendation ... reused from Physical device Product Technical category Electromagnetic Comparability Environmental conditions Electrical Interfaces Optical Interfaces Abstract device Software system Operating system Language Computer language Natural language Organisation National Multinational
Named competence relations Work experiences Has-experience-of Has-worked-on Participation-in Project-leader-of General Competencies Author-of Background-of Practice-of Has-role-of Has-skill-of
such as skills, roles, responsibilities, background knowledge, and project experiences are represented with these formal relations (competence relations) between the hierarchies. The hierarchy of primitive competencies is an agreed set of competence terms relating to a workgroup. The primitive competencies represent domain-independent knowledge. This means that this taxonomy should represent cognitive aspects related to a speciic knowledge process, such as a problem-solving task. These primitive competencies can be used in any problem domain. On the other hand, the hierarchy of application areas represents domaindependent knowledge. This taxonomy of competencies needs to be dynamic considering organisational changes and technical developments. New terms in both hierarchies and new relations must be created during the execution of work-related tasks. Some terms and their instances are reused from the previous ontologies. The idea is to create and reason with competence expressions, for example, to be able to represent that a speciic domain expert has-skill-of (competence relation) in design (primitive competence) of optical interfaces (application area). This expression example written in a formal notation is has-skill-of (design, optical interfaces). An expert annotation is described with factual and contextual information. Some information is reused from the information and product ontologies, such as the values for the nature and technical speciication terms. Contextual information allows the representation of work-related tasks and their competencies. The expert annotations are created to be shared by other members of the workgroup in other process tasks. This approach aims to enhance the group memory focusing on the capture, sharing, and reuse of individual knowledge
assets. Moreover, this approach can assist the development of a consensual taxonomy of competencies. The formal representation of competencies and the related expert annotations enable the deinition of an inference layer (set of inference rules and logic procedures) to deine reasoning mechanisms at the competence management level. This inference layer (unimplemented) would be a value added for the GMS framework. For example, giving a compliance question set as input, reasoning mechanisms could be developed to infer technical competencies (at different levels of granularity) to assist speciic compliance tasks. This way, the explicit and formal representation of competencies and expert annotations will act as a competence knowledge base within the group memory.
Final Remarks and Future Research

DAdderio (2003) makes a number of pertinent observations about the type of system we describe that need to be considered in any practical implementation. She showed that delegating memory to software can fundamentally reconigure the mechanisms by which organisational knowledge is structured, stored, retrieved, and reused and that attempts to impose a common language can generate a push toward greater standardisation and lead to the reduction of organisational heterogeneity. While this might be acceptable (even desirable) in some circumstances, others (e.g., Brown & Duguid, 1991; Wenger, 1996) argue that it is this very heterogeneity that provides fertile ground for innovation. Even in a more mundane sense, Hutchins (1993) has shown that the ability to be able to view a problem through the lens of different forms of representation can make an intractable problem seem less intractable. We believe that by using a combination of an ontology-based competence model and the system of expert annotations described above, we may be able to avoid this problem. Similarly, we hope to avoid the problem of the continuously emergent nature of meaning in organisational settings. In an organisation shared meanings are not a given but are continually negotiated and reconstituted; the everyday activity of work both generates new knowledge and provides a framework through which existing knowledge can be viewed. We believe that our approach should help us to avoid the problem of focusing on abstract representations to the detriment, if not exclusion, of actual practice.
References
Alvesson, M. (1993). Organisations as rhetoric: Knowledge-intensive irms and the struggle with ambiguity. Journal of Management Studies, 30(6), 997-1015. Alvesson, M. (1995). The management of knowledge-intensive companies. Berlin: Walter de Gruyter.
Bahrami, H. (1996). The emerging lexible organisation, perspectives from Silicon Valley. The California Management Review, 34(4), 55-75. Benjamins, V., Fensel, D., & Perez, A. (1998, October). Knowledge management through ontologies. In Proceedings of the 2nd Conference on Practical Aspects of Knowledge Management (PAKM98). Brown, J. S., & Duguid, P. (1991). Organizational learning and communities of practice: Toward a uniied view of working, learning, and innovation. Organization Science, 2(1), 40-57. Brown, J. S., & Duguid, P. (2000). Balancing act: How to capture knowledge without killing it. Harvard Business Review, 78(3), 73-80. Chiesa, V., & Manzini, R. (1997). Competence levels within irms: A static and dynamic analysis. In Sanchez & Heene (Eds.), Competence-based strategic management (pp. 195-214). John Wile & Sons. Curtis, B., Kellner, M., & Over, J. (1992). Process modelling. Communications of the ACM, 35(9), 75-90. DAdderio, L. (2003). Coniguring software, reconiguring memories: The inluence of integrated systems on the reproduction of knowledge and routines. Industrial and Corporate Change, 12(2), 321-350. Davis, R., Shrobe, H., & Szolovits, P. (1993). What is a knowledge representation? Artiicial Intelligence Magazine, 14(1), 17-33. DeLong, D. W. (2004). Lost Knowledge: Confronting the threat of an aging workforce. Oxford University Press Devedzic, V. (1999). Ontologies: Borrowing from software patterns. ACM Intelligence, 10(3), 14-24. Dzbor, M., Parlic, J., & Parlic, M. (2000). Knowledge management in a distributed organization (Tech. Rep. No. KMI-TR-94). The Open University. Gruber, T. (1992). Ontolingua: A mechanism to support portable ontologies (Tech. Rep. No. KSL-91-66). Knowledge Systems Laboratory. Heijst, G., Spek, R., & Kruizinga, E. (1997). Corporate memories as a tool for knowledge management. Expert Systems with Applications, 13(1), 41-54. Hildreth, P., & Kimble, C. (2002). The duality of knowledge. Information Research, 8(1). Retrieved from http://InformationR.net/ir/8-1/paper142.html Hutchins, E. (1993). Learning to navigate. In S. Chaiklin & J. Lave (Eds.), Understanding practice (pp. 35-63). Cambridge University Press. Kimble, C., & Hildreth, P. (2005). Dualities, distributed communities of practice and knowledge management. Journal of Knowledge Management, 9(4), 102-113. Kimble, C., Li, F., & Barlow, A. (2000). Effective virtual teams through communities of practice (No. 00/9). Glasgow: University of Strathclyde. Michellone, G., & Zollo, G. (2000). Competencies management in knowledge-based irms. Journal of Manufacturing Technology and Management, 1(1), 20-41.
Nonaka, I. (1994). A dynamic theory of organisational knowledge creation. Organisation Science, 5(1), 14-37. Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company. Oxford University Press. Polanyi, M. (1966). The tacit dimension. Garden City, NY: Doubleday. Reuber, A., Dyke, L., & Fischer, E. (1990). Using tacit knowledge methodology to deine expertise. In Proceedings of The 1990 ACM SIGBDP Conference on Trends and Directions in Expert Systems (pp. 262-274). Sharratt, M., & Usoro, A. (2003). Understanding knowledge-sharing in online communities of practice. Electronic Journal of Knowledge Management, 1(2), 187-196. Stader, J., & Macintosh, A. (1999). Capability modelling and knowledge management. Applications and Innovations in Expert Systems VII. In Proceedings of ES99: 19th International Conference of the BCS Specialist Group on Knowledge-Based Systems and Applied Artiicial Intelligence (pp. 33-50). Sveiby, K. (1997). The new organisational wealth: Managing and measuring knowledgebased assets. Berrett Koehler Publisher. Teigland, R., & Wasko, M. M. (2004). Extending richness with reach: Participation and knowledge exchange in electronic networks of practice. In P. Hildreth & C. Kimble (Eds.), Knowledge networks: Innovation through communities of practice (pp. 230242). Hershey, PA: Idea Group Publishing. Tuomi, I. (2002). The future of knowledge management. In Lifelong Learning in Europe (pp. 69-79). Luxembourg: Ofice for Oficial Publications of the European Communities. Uschold, M., King M., Moralee S., & Zorgios Y. (1997). Enterprise Ontology. Artiicial Intelligence Applications Institute (AIAI), University of Edinburgh, Technical Report AIAI-TR-195. Vasconcelos, J., Gouveia, F., & Kimble, C. (2002, November 20-22). An organizational memory information system using ontologies. In Proceedings of the 3rd Conference of the Associao Portuguesa de Sistemas de Informao, Coimbra, Portugal. Vasconcelos, J., Kimble, C., & Rocha, A. (2003, July 2-4). Ontologies and the dynamics of Organisational environments: An example of a group memory system for the management of group competencies. In Proceedings of I-KNOW03, Graz, Austria. Walsh, J. (1995). Managerial and organisational cognition. Organisation Science, 6(3), 280-321. Wenger, E. (1996). Communities of practice: The social fabric of a learning organization. Healthcare Forum Journal, 39(4), 20-24.
270 Colucci, Di Noia, Di Sciascio, Donini, & Ragone
Chapter XIII
Integrated Semantic-Based Composition of Skills and Learning Needs in Knowledge-Intensive Organizations

Simona Colucci, SisInfLab, Politecnico di Bari, Italy Tommaso Di Noia, SisInfLab, Politecnico di Bari, Italy Eugenio Di Sciascio, SisInfLab, Politecnico di Bari, Italy Francesco Maria Donini, Universit della Tuscia, Italy Azzurra Ragone, SisInfLab, Politecnico di Bari, Italy
Abstract
Holding and creating competencies is one of the most strategic activities in organizations, especially in knowledge intensive ones. An organization dealing with a task to perform will irst check for the required skills among the available personnel. If such a search process leads to discovering lacking competencies, organizations may hire external personnel or encourage internal personnel to learn new competencies on the unavailable skills. The
Intergrated Semantic-Based Composition of Skills and Learning Needs 271
current availability of several well-organized e-learning modules makes such a possibility appealing and economically advantageous. A skill management system performing both the processes of searching among available skills and facilitating the creation of missing ones can hence be a noteworthy source of competitive advantage for a knowledge-intensive organization. We present here an approach and a system for such purpose, which exploits recent advances in semantic-based inference services and technologies. The proposed approach employs Description logics formalism and reasoning services and is fully in the Semantic Web initiative mainstream.
Introduction
Creation and maintenance of competencies is one of the most strategic activities in organizations, especially in knowledge-intensive ones, as is the case of consulting companies. The role of core competencies in making a company achieve competitive advantage has been widely investigated in Hamel and Prahalad (1990). Other studies (Gronau & Uslar, 2004) show that the return on investment is signiicantly impacted by enriching knowledge management systems companies use, with components for the speciic management of skills (Skill Management Systems, SMS). One of the services SMS should provide is the search for competencies inside the organization; an organization dealing with a task to perform will irst check for the required skills among the available personnel, an activity whose complexity grows up with the size of the company. If such a search process leads to discovering a lack of needed competencies, organizations may hire external personnel or encourage internal personnel to learn new competencies on the unavailable skills. The current availability of several well-organized e-learning modules makes such a possibility appealing and economically advantageous. A skill management system performing both the process of searching among available skills and facilitating the creation of missing ones can hence be a noteworthy source of competitive advantage for a knowledge-intensive organization. We present here an approach and a system for supporting the whole process of skills retrieval and creation inside a company. The proposed framework exploits recent advances in semantic-based inference services and technologies. It is noteworthy that the terms skill and competence are not kept distinct in the rest of the chapter. They both are meant to describe any ability or sort of knowledge held by individuals. The approach exploits the formalism and the reasoning services provided by description logics (DLs) and is fully in the Semantic Web initiative mainstream. It is well known that standard reasoning services from DLs can be used to evaluate if the individual proile and the taskboth described in DLs formalismcompletely match, for example, the proile is classiied by the task. Yet usually one is not only interested in perfect matches, which can be rare when complex and expressive descriptions are used. If we revert to classical unstructured-text information retrieval systems, we may obtain a similarity-based match, but such matches are only probabilistic and, for example, two fragCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
ments of descriptions such as C++ programmer and experienced analyst and developer using Object Oriented languages would simply get no similarity. Nevertheless, it is obvious, from the above example, that a human user would infer that the two descriptions are at least a potential match and would ask which are the object-oriented languages known to the individual. Hence, also potential matches have to be taken into account; we still care about the reasons why the descriptions do not exactly match with each other in order to rank all the individuals potentially matching the task to perform and individuate further competencies needed. We employ two non-standard reasoning services proposed and studied for description logics (Di Noia, Di Sciascio, Donini, & Mongiello, 2003) for evaluating such reasons: concept contraction for discovering the characteristics in the individual proile that are in conlict with the ones speciied in the task description (this process is performed in order to suggest a belief revision); and concept abduction for explaining the characteristics required for carrying out the task and missing in the individuals proile. Let experienced analyst and developer using object-oriented languages and C++ novel programmer be respectively a required task and an individual proile. A contraction process will return the information that, in order to gain a potential match, the requirement about experience has to be given up and the rest of the demand may be kept. If we consider instead the two fragments before C++ programmer and experienced analyst and developer using object-oriented languages, respectively, as task and individual proile, the abduction process would return the concept C++ because it is not explicitly said in the proile that the individual knows about C. Meanwhile, we can state that the individual knows object-oriented programming, which is part of the background of a C++ programmer. The choice of management in the assignment process may revert to a single individual (one-to-one skill matching) or on an ad-hoc composed team (many-to-one skill matching). When a team has to be composed, a concept covering problem (Colucci, Di Noia, Di Sciascio, Donini, Piscitelli, & Coppi, 2005a) is solved. The problem can be seen as analogous, yet with noteworthy peculiarities, to a classical set covering problem in a semantic-based framework because of its deinition in terms of concept abduction. The approach so far outlined performs the process of searching among available skills; when the organization personnel is not endowed with all the skills required for needed tasks, missing competencies are individuated by exploiting the abduction service and can become part of the learning process employees may be asked to start. The nice property of the proposed approach is that the e-learning modules discovery and composition process can be basically carried out iterating the formerly outlined process, as the unavailable competencies become a speciication of required courseware and are used to discover and compose, using semantics of descriptions, learning resources covering as much as possible the learning need and orchestrating resources according to speciied prerequisites. The current implementation of our approach extends to this aim learning object metadata (LOM) standard header (IEEE, 2002), embedding a structured module description expressed in an Ontology Web Language - Description Logics (OWL-DL) (OWL, 2004) subset. The rest of this work is organized as follows: in the background section, a literature review on skill management and e-learning systems is provided; an overview on description logics and their relationship with Semantic Web languages is also given to make the proposed semantic-based approach understandable. The issues, problems, and controversies the chapter
aims to solve are presented in the section on integrated semantic-based composition of skills and learning needs to motivate our solution proposal. The system is then outlined with the help of an example. Conclusions and future trends close the chapter.
Background
Skill Management Systems
Systems and techniques for skills management have recently become the object of growing interest, as knowledge and expertise of individuals have been acknowledged strategic assets of knowledge-intensive companies (Hamel & Prahalad, 1990; OLeary, 1998a). In Gronau and Uslar (2004), the role of skill management systems in organizational activities such as expert inding, personnel recruitment, personnel development, and project management has been underlined. All these activities need the deinition of the skills individuals are endowed with. Such an activity traditionally involves human judgment in individuating and classifying skills held by individuals in evaluating the degree of competence and in keeping up-to-date individuals proiles. Information technology (IT)-supported system use is then suggested in managing companies competencies in order to downsize the subjectivity of human evaluation. Matching personal proiles is an activity required in a rising number of scenarios, ranging from recruitment agencies and human resource organizational units to dating services. Those contexts, even in their deep dissimilarity, share the need to satisfy a list of individual requests by matching them with available individual proile offers. The match between individual proiles we are interested in is, obviously, not an exact one, which is both quite simple and rare. Given a task description and individual proile descriptions, the matchmaking process has to return one or more best-possible matches among the available ones. The problem of expert inding is also modeled in Yimam (2000), in which the requirements of a skill-inding approach are formally outlined in a domain analysis. A server architecture for expert inding, based on the previously mentioned domain analysis, is also proposed. The use of ontologies as knowledge repositories has now become almost common in novel knowledge management architectures in order to give a common vocabulary and to use inference services on elicited knowledge (OLeary, 1998a; OLeary, 1998b). Skill management systems presented in literature, almost all embedding skill searching facilities, may be classiied in two categories: non-ontology-based and ontology-based systems. Among non-ontology-based approaches, database querying and similarity between weighted vectors of stemmed terms, typical of text-based information retrieval, have been used to evaluate possible matches (Veit, Muller, Schneider, & Fiehn, 2001). Obviously, forcing proiles to be expressed by data structures or vectors of terms does not allow dealing with incomplete information, always present in matchmaking context in the form of either unavailable or irrelevant information.
Skill matching also has been modeled as a bipartite graph in which the irst set of vertices includes assignees and the second one includes tasks to be performed (Saip & Lucchesi, 1993). Edges belonging to this graph link people to tasks. By determining a cost function that associates each edge with a real value, a weighted bipartite graph emerges, which results in a well-known problem in operational research, the assignment problem (Galil, 1986, Hillier & Lieberman, 1995; Kennington & Wang, 1991). Among proposals on the subject, in Sure, Maedche, and Staab (2000), two skill-matching systems, ProPer and OntoProper, were presented, both storing in a database skill proiles represented as vectors and using approaches from decision theory to allow for approximate match not obtainable with plain database queries. OntoProper embeds also an ontology, reducing skill database maintenance effort by enriching the database with ground and inferred facts from secondary information, such as project documents. But both systems lack an ontology as skill repository, allowing them to infer on previously introduced proiles. In Becerra-Fernandez (2000), two people inder knowledge management systems are proposed: the searchable answer generating environment(SAGE) and the expert seeker. Both systems use DataBase Management Systems (DBMS) as skill repositories and query engines performing a keyword search for expertise, even if the second one provides more search options. Even though proposing a database approach, the paper underlines the need to employ artiicial intelligence technologies in people inder knowledge management systems in order to infer new knowledge from elicited skills and to keep automatically up-to-date proiles employing data mining techniques. Also, agent technologies have been employed to support the search for the right expert. In Garro and Palopoli (2003), an XML multi-agent system providing support to management in searching the most suitable employee for a speciic job, together with many other facilities is proposed. In Sugawara (2003), an agent-based application for supporting job matchmaking is proposed, focusing on the telework scenario. In Lau and Sure (2002), an ontology-based skill management system is proposed, allowing employees to elicit their skills and providing an advanced expert search within a company intranet. In Hefke and Stojanovic (2004), a semantic-based portal is proposed. The portal answers users queries about tasks to perform by providing ad-hoc organizational teams. The user request is formalized as a query searching the competences required for the task in the ontology used as skill repository. The system returns a set of one or more workers able to cover all the competences required for the task. All the available sets are ranked on the basis of the ontological closeness of query concepts to concepts formalizing skills held by proposed people. In Liu and Dew (2004), a system integrating the accuracy of concept search with the lexibility of keyword search is proposed to match expertise within academia. The system is based on the use of Semantic Web technologies, and in particular on Resource Description Framework (RDF) and extensible markup language (XML) in order to extract expertiseintegrated proiles from heterogeneous information sources. Our approach takes full advantage of structured, ontology-based descriptions and adopts an open world assumption wherein the absence of a characteristic in a description is not interpreted as a constraint of absence but instead considered a characteristic that could be either reined later or left open if it is irrelevant, typical of knowledge representation. It obCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
viously allows users to ind a set of one or more individuals who, based on provided skills descriptions, cover the requested task but also, when a completely satisfactory set cannot be retrieved due to lack of requested skills, provides a logic-based answer to what is missing or what should be revised in the task request to cover it based on available skills. Skill management can be characterized in terms of multiplicity relationships between individuals skills and tasks to be accomplished (Colucci, Di Noia, Di Sciascio, Donini, & Mongiello, 2003c): One to one: one task or job proile has to be matched with one individual; Many to one: one task has to be assigned to several individuals who, together, are endowed of all skills requested for task realization; One to many: several tasks have to be matched with the skills of an individual able to accomplish them; Many to many: several tasks have to be assigned to several available individuals.
In Colucci, Di Noia, Di Sciascio, Donini, and Mongiello (2003b) a semantic-based approach was presented to the problem of skills inding in an ontology-supported framework. The framework is devoted to one-to-one skill matching and considers skill management an electronic marketplace of knowledge in which skills are a peculiar kind of goods that have distinguishing characteristics with respect to traditional assets; buyers are entities who need the skills of people, such as projects, departments, and organizations. On the other hand, knowledge sellers are individuals who offer their own skills. Obviously, descriptions of proiles share a common skills ontology. Although semantic facilitators have been proposed in the literature for several scenarios (Staab, Studer, Schnurr & Sure 2001; Sure et al., 2000; Trastour, Bartolini, & Priest, 2002), they do not take full advantage of the ontological structure and limit their search to simple subsumption/classiication matching. Our approach, based on description logics formalization and reasoning, is oriented to inding the best individual for a given task or project, based on proile descriptions sharing a common ontology. The approach copes with cases in which no perfect matches exist, that is, inding those available proiles that for a given skill request best match also, if not identical, and vice versa. In particular, we logically distinguish cases in which some skills in a request proile are not speciied in the offered one, yet there is no contradiction and, for example, further inquiries can be done (what is called a potential match); and cases in which some skills in the request are in contrasted with the given proile (what is called a partial match). In this case, the one who is carrying out the search may check for unsatisiable requests and eventually retract them if no better choice is at hand. It is noteworthy that our approach allows not only a logical categorization but also a ranking of matches within each category. Notice that a full match is hence just a special case of a potential match. In Colucci Di Noia, Di Sciascio, Donini, Mongiello, and Piscitelli (2004), an approach was presented to the contemporary optimisation of several one-to-one skill matches, endowing with semantics the process of searching solutions to task assignment. The problems classical application is to assign jobs to employees minimizing an objective function measuring
the total cost of assignment. We may think of the cost function used for weighting arcs in terms of suitability of persons to tasks. This assumption causes the objective function to measure quantitatively the effectiveness of performing all the tasks instead of the total cost of the assignment. Evaluating the suitability of an individual to a job is a task traditionally performed by companies management on the basis of personal knowledge of workers. As a result, knowledge about coeficients measuring suitability of different matches is subjective and implicit, not allowing end users to clearly determine the reasons for match suggestions and to eventually revise them. Such an approach makes this process objective and explicit by using algorithms (Colucci et al., 2003c) exploiting reasoning services from DLs. All the systems and approaches so far outlined deal with the search for skills among the internal personnel. Earlier in the chapter, the importance of creating new competencies when the available ones are not enough to perform all the needed tasks was underlined. In order to achieve such knowledge creation, SMS may integrate components supporting the training process of employees, exploiting e-learning technologies. The term e-learning has become common, describing several concepts, from complete Web-based courses to distance learning and tutoring. Recently, also thanks to various standardization efforts (IEEE, 2003), emphasis has been placed on the concept of learning object, that is, small and easily reusable educational resources to be composed to allow personalized instruction and courseware creation (Ajami, 2004; Cabezuelo & Beardo, 2004; Ip, Young, & Morrison, 2002; Vossen & Jaeschke, 2003). Obviously, discovery and composition of such learning objects in an automated way requires the association of unambiguous and semantically rich metadata, deined in accordance with shared ontologies. The LOM (IEEE, 2002) standard, though limited in the basic annotation items, allows freely deined annotated metadata describing a learning resource. The semantic-based annotation of educational resources is hence fully in the stream of the Semantic Web initiative (Berners-Lee, Hendler, & Lassila, 2001), and it can share with it both techniques and approaches (Bennacer, Bourda, & Doan, 2004; Gasevic, Jovanovic, & Devedzic, 2004; Sanchez & Sicilia, 2004). In particular, as more and more learning objects become available on the Web as services with well-deined machine interpretable interfaces as described, for example, in OWL-S (OWL-S, 2004, Sycara, Paolucci, Ankolekur, & Srinvivasan, 2003), personalized learning units can be built from scratch by retrieving learning resources. Automated composition of learning resources, exposed as Web services, for example, can then match a personalized learning need.
Description Logics and OWL Basics

Description logics (Baader, Calvanese, McGuinness, Nardi, & Patel-Schneider, 2002) are a family of logic formalisms for knowledge representation whose basic syntax elements are concept names and role names. Concepts stand for sets of objects, for example, ProcessEngineer, Graduate, BusinessApplication, while roles, for example, hasAbility or specialized, link objects in different concepts.
Basic elements can be combined using constructors to form concept and role expressions. Based on the set of constructors adopted, different DLs can be deined. Every DL allows one to form a conjunction of concepts denoted as ; some DLs include also disjunction and complement to close concept expressions under boolean operations. Roles can be combined with concepts using existential role quantiication () , e.g., GraduatehasAbility. NegotiationSkills, which describes the set of graduated people with negotiation skills, and universal role quantiication (), for example, Programmer hasMasterDegree. Engineering, which describes programmers having only an engineering degree. Other constructs may involve counting, as number restrictions: Graduate(3hasAbility) expresses graduates having at least three abilities, and AccountManager(2 hasTechnicalSkills) describes account managers endowed with at least two skills belonging to the technical area. The representation of knowledge is achieved in DLs formalism by using concept expressions to structure inclusion assertions and deinitions. For example, we could impose that programming may be partitioned into structural and object oriented using the two inclusions ProgrammingStructuralProgrammingObjectProgramming and StructuralProgramming ObjectProgramming. We can state also that working teams have to be composed of at least two members as Team (2hasMembers). Historically, sets of such inclusions are called TBox (Terminological Box). The basic reasoning problems for concepts in a DL are satisiability, which accounts for the internal coherency of the description of a concept (no contradictory properties are present), and subsumption, which accounts for the more general/more speciic relation among concepts that forms the basis of a taxonomy. More formally: A concept C is satisiable if there exists an interpretation in which C is mapped into a nonempty set, unsatisiable otherwise. If a TBox T is present, satisiability is relative to the models of T; that is, the interpretation assigning C to a nonempty set must be a model of the inclusions in T. A concept C subsumes a concept D if every interpretation assigns to C a subset of the set assigned to D. Also, subsumption is usually established relative to a TBox, a relation that we denote T C D.
Also, a TBox can be categorized satisiable if there exists at least one model (i.e., an interpretation fulilling all its inclusions in a nontrivial way). It is easy to see that T in DLs represents what is called an ontology in a knowledge representation system. In the rest of the paper we refer to the ALN (Attributive Language with unqualiied Number restrictions) DL. The choice of such a DL is due to a trade off between language expressiveness and computational complexity of inference services (Brachman & Levesque, 1984). Constructs allowed in an ALN DL are:
universal concept. All the objects in the domain. bottom concept. The empty set. A atomic concepts. All the objects belonging to the set represented by A. A atomic negation. All the objects not belonging in the set represented by A. CD intersection. The objects belonging both to C and D. R.C universal restriction. All the objects participating to the R relation whose range are all the objects belonging to C. R.C unqualiied existential restriction. There exists at least one object participating in the relation R.
Table 1. Correspondence between OWL-DL and ALN DL syntax
OWL syntax <owl : Thing/ > <owl : Nothing/ > < owl : Classrdf : ID ="C"/ > < owl : ObjectPropertyrdf : ID ="R"/ > < rdfs : subClassOf/ > < owl : equivalentClass/ > < owl : disjointWith/ > < owl : intersectionOf/ > < owl : allValuesFrom/ > < owl : someValuesFrom/ > < owl : maxCardinality/ > < owl : minCardinality/ > < owl : cardinality/ >
DL syntax C R = =
( n R), ( n R), (= n R). Respectively the minimum, the maximum and the exact number of objects participating in the relation R.
We use a simple-TBox in order to express the relations among objects in the domain. With a simple-TBox in all the axioms (for both inclusion and deinition), the left side is represented by a concept name. Ontologies using the above logic can be easily modeled using languages for the Semantic Web (DAML+OIL, 2001; McGuinness, Fikes, Hendler, & Stein, 2002; OWL, 2004). These languages have been conceived to allow for representation of machine understandable, unambiguous, description of Web content through the creation of domain ontologies and aim at increasing openness and interoperability in the Web environment. The strong relations between DLs and the above-introduced languages for the Semantic Web (Baader, Horrocks, & Sattler, 2003) is also evident in the deinition of the three OWL sub-languages: OWL-Lite: Allows class hierarchy and simple constraints on relation between classes; OWL-DL: Is based on description logics theoretical studies; it allows a great expressiveness keeping computational completeness and decidability; OWL-Full: Using such a language, there is a huge syntactic lexibility and expressiveness. This freedom is paid in terms of no computational guarantee.
The subset of OWL-DL tags allowing to express an ALN DL is presented in Table 1. In the rest of the chapter we will use DL syntax instead of OWL-DL syntax, to make expressions much more compact.
Integrated Semantic-Based Composition of Skills and Learning Needs

The skill management approach presented here is aimed at supporting decisions of company management in the whole process of inding competencies required for tasks to be performed. When a task has to be performed, it is reasonable that company management tries as a irst attempt to ind the skills needed among the internal personnel. It is up to the management either to assign the task to a single person (one-to-one matching) or to an ad-hoc created team (many-to-one matching). In both cases, the match we are interested in providing is not just a perfect one. In case neither an employee nor a team is able to cover all the skills required for the task, we still search for the best available match, as humans would do. The proposed approach allows management to know the reasons why the match is not full, that is, which skills either the employee or the team is missing to perfectly cover the task. In other words, the approach provides an explanation hypothesis, exploiting the semantics of both the task and individuals proile descriptions. On the basis of such explanation, a personnel
training process on lacking skills may take place. In the case of one-to-one matching, the employee assigned to the task has to make an effort in the learning process. In the case of a team ad-hoc created for performing the task, each member of the team has to exhibit a different learning process dependent on the background knowledge expressed by his or her proile. The courseware employees are asked to attend is automatically composed on the basis of the learning need resulting by the previous searching process.
The Semantic Base of the Composition Process

Standard inference services in DLs include subsumption and satisiability. These basically return a yes/no answer. The scenario we outlined requires instead both explanation and belief revision in order to cope with cases in which no perfect march exists. Hereafter, we recall basic deinitions of two non-standard inference services that will be used to overcome highlighted limitations of classical ones. For the readability of the following deinitions, it can be helpful to think of the concept S (supply) in terms of provided individual proile and D (demand) in terms of required task.
Non-Standard Inferences for Semantic-Based Belief Revision and Explanation

Let us consider two concepts S and D. If their conjunction S D is unsatisiable in the TBox T representing the ontology, that is, they are not compatible with each other, we may want, as in a belief revision process, to retract requirements in D, G (for Give up), to obtain a concept K (for Keep) such that K S is satisiable in T. Deinition 1 Let L be a DL, S, D, be two concepts in L, and T be a set of axioms in L, where both S and D are satisiable in T. A concept contraction problem (CCP), identiied by L, D, S, T, is inding a pair of concepts G,KL L such that T D G K, and K S is satisiable in T. We call K a contraction of D according to S and T. Obviously, there is always the trivial solution G,K = D, to a CCP, that is, give up everything of D. In our skill-matching framework, it models the (infrequent) situation in which, in front of some very appealing proile S, incompatible with the requested task, a recruiter just gives up completely her speciications D in order to meet S. On the other hand, when SD is satisiable in T, the best possible solution is , D, that is, give up nothing, if possible. Hence, a concept contraction problem is an extension of a satisiability one. Since usually one wants to give up as few things as possible, some minimality in the contraction must be deined (Grdenfors, 1988). In most cases, a pure logic-based approach could be not suficient to decide between which beliefs to give up and which to keep. There is the need of modeling and deining some extra-logical information. One approach is to give up minimal information (Colucci et al., 2003a). Another one considers some pieces of information more important than others, and the information that should be retracted is
the least-important one that is negotiable and strict constraints are introduced (Di Noia, Di Sciascio, & Donini, 2004a). When subsumption does not hold, that is, a full match is unavailable, one may want to hypothesize some explanation of the causes of this result. In Di Noia et al. (2003), the concept abduction problem (CAP) was introduced and deined as a non-standard inference problem for DLs to provide a logic-based answer to such question. Deinition 2 Let S, D be two concepts in a description logic L, and T be a set of axioms, where both S and D are satisiable in T. A concept abduction problem (CAP), denoted as L, C,D, T, is inding a concept H such that T SH , and T S HD . P is a symbol for a CAP, and SOL(P) denotes the set of all solutions to a CAP P. Di Noia et al. (2003) also propose a minimality criteria for H and a polynomial algorithm to ind solutions that are irreducible for an ALN DL. Given a CAP, if H is a conjunction of concepts and no sub-conjunction of concepts in H is a solution to the CAP, then H is an irreducible solution. The rankPotential algorithm (Di Noia, Di Sciascio, Donini, & Mogiello, 2004b) allows one to numerically compute the length of H. Obviously such length, for a complex ontology, is much more than a trivial depth in a hierarchy. The solution to a CAP can be interpreted as what has to be hypothesized in S and in a second step added to make S more speciic than D, which would make subsumption result true. So, as concept contraction extends satisiability, concept abduction extends subsumption. The two inference services introduced above are the basis of the logic-based approach we propose and detail in the following.
One-to-One Skill Matching

The process of assigning one employee to one task we propose here involves both belief revision and explanation. The request for assignment is supposed to be revisable if no individual proile potentially matching it is at hand. At the same time, we are still interested in an explanation of the skills uncovered by each proile, either in case the request is contracted or not. Such information is used to rank all the possible candidates with regard to the job description on the basis of both conlicting and hypothesized skills in proiles. In particular, we deine here an algorithm Assign(TK,P,T) evaluating the space of possible matches between one needed task TK and a set of available proiles P= {Pi}, i =1...n , with regard to a TBox T deined in ALN. 1: Algorithm Assign(TK,P,T) 2: input P, T K G concepts in L such that both T P / and T TK / 3: output as Passigned, H, G,
4: begin algorithm 5: 6: 7: 8: 9: 10: 11: 12: 13: 14: 15: 16: 17: 18: 19: 20: 21: 22: 23: 24: 25: 26: 27: 28: Passigned = ; Umin = ; N = rankPotential(;TK;T); for each Pi P if T TK Pi then Gi,Ki = contract(Pi,TK,T); else Ki = TK; end if Hi = abduce(Pi, Ki, T); ki = rankPotential(,Ki,T); hi = rankPotential(Pi,Ki,T); gi = rankPotential(Ki,TK,T); Ui= u(N, ki, hi, gi); if Ui<Umin; then Umin = Ui ; Passigned = Pi; H =Hi; G =Gi; end if end for each return Passigned, H, G;
29: end algorithm Notice that Hi = abduce(Pi, Ki, T) determines that Hi is a solution for the CAPL, T,Pi, T while Gi,Ki = contract(Pi,T,T) determines Gi,Ki is a solution for the CCPL, T,Pi, T. The algorithm returns an assignee Passigned among the elements of P, together with an explanation H of skills to be hypothesized in Passigned to completely cover T and the concept G in case a preliminary contraction process on T is needed. The choice among P is made by minimizing the function U proposed in Colucci, Di Noia, Di Sciascio, Donini, and Mongiello (2005c) computing a measure, indicating how good a match is, according to the simple closed form:

N h *(1 Ng k
U ( N , k , h, g ) = 1
with the following meaning for parameters: k: length of K that belongs to the solution of a concept contraction problem between Pi and T h: length of H solution of a concept abduction problem between K (Pi if no contraction is needed) and T g: length of G that belongs to the solution of a concept contraction problem between Pi and T N: length of T.
Such lengths are computed by applying the algorithm rankPotential as follows: k = rankPotential(,K,T) h = rankPotential(P,K,T) g = rankPotential(TK,K,T) N = rankPotential(,TK,T)
The rational of the closed form is given in Colucci et al. (2005c). By choosing the candidate minimizing U, the algorithm takes into account both g and h, that is, a numerical measure of how much it has to be given up in the request T and how much to hypothesize in the proile analysed at each stage.
Many-to-One Skill Matching

When an ad-hoc team has to be created for performing task T, the process of team composition can be carried out by solving an extended concept covering problem deined in terms of abduction and contraction. We recall the deinition of a concept covering problem in terms of concept abduction, independently on the employed DL L originally given in Colucci et al. (2005a): Deinition 3. Let D be a concept, R={S1, S2,...,Sk} be a set of concepts in a Description Logic L, and T be a set of axioms, where D and S1, i =1.....k are satisiable in T. 1. A concept covering problem (CCoP), denoted as CCoP(L, R,D, T ), is inding, if it exists, a set Rc R, such that both for each SjRc, TSj, and H SOL( L, Sj,D, T ) is such that T HD.
2.
We call Rc,H a solution for the CCoP L, R,D, T.
In the same paper, the algorithm GREEDYsolveCCoP exploited concept abduction to extend a tractable greedy set-covering algorithm (Cormen, Leiserson, & Rivest, 1990). We do not delve into details but notice that a concept covering problem has noteworthy peculiarities with regard to a set covering one (Colucci et al., 2005a). Here we propose the algorithm TeamComposer(P,T,T), where T is the needed task and P= {Pi}, i=1...n is a set of individual proiles both described with regard to a TBox T to take also concept contraction into account during the process of selection of team members. 1: Algorithm TeamComposer(P,T,T) 2: input concepts T, Pi P, where T and Pi are satisiable in T 3: output Pc, Tuncovered, Gcontraction 4: begin algorithm 5: 6: 7: 8: 9: 10: 11: 12: 13: 14: 15: 16: 17: 18: 19: 20: 21: 22: 23: 24: 25: 26: 27: Pc=; Tuncovered= T; Gcontraction= ; Hin= ; Gin= ; Umin = ; N = rankPotential(, T,T); do Pmax= ; for each Pi P such that Pc {Pi} covers Tuncovered if T TuncoveredPi then G,K = contract(Pi,Tuncovered,T); else K= Tuncovered; G = ; end if H = abduce(Pi, K, T); k = rankPotential(, K,T) h = rankPotential(Pi, K, T); g = rankPotential(K,TK, T); U= u(N, k, h, g); if U<Umin Pmax= Pi; ;
28: 29: 30: 31: 32: 33: 34: 35: 36: 37: 38:
Hin= H; Gin= G; end if end for each if (Pmax / and K / ) then P =P\{Pi}; Pc = Pc {Pi}; TKuncovered= Hin; end if while (Pmax / and K / ); Gcontraction = Gcontraction Gin;
39: return Pc, TKuncovered, Gcontraction; 40: end algorithm The algorithm tries to cover T as much as possible, using the concepts Pi P. If a new individual proile Pi can be added to the already-composed team Pc, i.e., T(Pk Pc Pk) Pi, then an extended matchmaking process is performed (rows 14-32). If Pi is not consistent with the uncovered part of the task Tuncovered, the latter is contracted and subsequently an abduction process is performed between the contracted uncovered task and Pi (rows 15-17). If Pi is consistent with Tuncovered, only a concept abduction problem is solved (rows 19-21). Based on the previously computed concepts G, K, and H, a global score is used as a metric to evaluate how good Pi is with respect to the covering set (rows 23-26). We do not need to change the deinition of concept covering problem provided in Colucci et al. (2005a); the problem we solve is still the one detailed in Deinition 3. The distinguishing element between the two solving algorithms GREEDYsolveCCoP and TeamComposer is in the choice criterion among the Pi greedy selected to compose the team. In GREEDYsolveCCoP such choice was made on the basis of a minimality criterion on H, the solution of an abduction problem on Tuncoveredthe part of the needed task yet to cover at each stage of the algorithmand the selected proile Pi. In TeamComposer, instead the choice criteria is the minimization of the function U explained in Section 3.1.2, which takes into account also a measure of how much the task request should be contracted before the members selection, thus improving the selection process. The outputs of TeamComposer are: Pc: the set of employees composing the team Tuncovered: the part of the task description not covered by the ad-hoc created team Gcontraction: the part of the task description given up to at the end of the whole team composition process
Composition of the Training Process1

The matching process so far outlined returns an explanation of missing skills both in the case of assignment to one employee and of team composition. The formal representation of this explanation is given by H returned by Assign in one-to-one skill matching and by Tuncovered returned by TeamComposer in many-to-one skill matching. H or Tuncovered represent the learning need to be covered by a courseware to be automatically generated by composing learning objects through a process exploiting the DL standard and non-standard inference services. In Colucci, Di Noia, Di Sciascio, Donin, and Ragone (2005b), we proposed a general framework based on OWL technologies for composition; such framework can be easily integrated in existing metadata speciications, such as SCORM (shared content object reference model, SCORM, 2004), LOM (IEEE, 2002), IMS (IMS, 2001), and DublinCore (DublinCore, 1999), although we currently use a LOM extended header. The courseware composition was considered a learning objects ( ) retrieval problem. In this perspective, if there is a *learning need* and a repository of learning objects potentially satisfying the learner speciications, a solution to a -retrieval problem is: retrieve (a sequence of) some s from the repository such that their composition satisies the learning need as far as possible. In case a perfect covering of the learning need is not found, an approximate solution has to be taken into account, together with explanation hypothesis of what remains missing. Formally learning objects and learning needs are deined as follows: Learning Object1 : = D, BK. D describes the knowledge the user will acquire after she/he uses . Using a language endowed with a well-deined syntax and semantics, it models the offered knowledge. BK is a representation of prerequisites in order to beneit from . Learning Need: = D, BK, where D is the description of the requested learning need and represents the background knowledge owned by the requester before looking for the BK courseware. According to the framework we propose in this chapter, the learning need is the one returned by the previous searching process: D is exactly the explanation H or Tuncovered returned in a case of one-to-one or many-to-one skill matching, respectively. As concerns the background knowledge BK held by the assignees when the training process starts, we have to distinguish the multiplicity cases: one-to-one skill matching : BK is represented by the proile Pi returned by Assign, that is, the proile of the individual selected for performing the task
many-to-one skill matching : each member Pi of the ad-hoc created team Pc returned by TeamComposer has to start a training process in which BK is represented by her proile Pi.
Notice that we need to search for one courseware for each member of the ad-hoc created group because the background knowledge is typical of each individual and the learning process to initiate depends on the skills initially held by the members. The algorithm teacher presented in Colucci et al. (2005b) automatically computes a composite courseware. The algorithm takes as input a set of learning objects R = { i = iD, iBK}, the learning need = D, BK, and an ontology T and returns the composite courseware ( ,R) and the uncovered part, Duncovered , of the request description D . A composite courseware is a sequence of learning objects such that both the following conditions hold: it can be started using some background knowledge the requester owns ( BK) and the provided composite courseware covers the user request description ( D). More formally, we need the following deinition of courseware low to deine a composite courseware. A courseware low with respect to some initial background knowledge BK, denoted as ( BK), is a inite sequence of learning objects ( 1, 2,...., i,.... n), where for each learning object i belonging to the courseware, all the following conditions hold: 1. The initial background knowledge, BK, is at least 1BK, that is, the background knowledge required by 1, the irst learning object of the sequence. In order to learn from a sequence of learning objects (LOs), the user must have at least the prerequisites needed to learn from the starting LOs. After using i-1, the user has a background knowledge that is at least iBK, that is, the one required by the i-th Learning Object. While beneiting from the composite LOs, the user acquires new knowledge that becomes part of his or her background. Such an *updated* background knowledge must satisfy the i requirements.
2.
The background knowledge of the learner before the fruition of i is the conjunction of all the knowledge provided by jD, with j<i, and the initial background knowledge BK.
Indicating with BKi the background knowledge before using i, using DL syntax, the following relation ensues:
BKi =
BK
1 D
2 D
.....
i-1 D
We can now deine formally a courseware low. Deinition 4. A courseware low with respect to some initial background knowledge BK is a inite sequence of learning objects ( BK) =( 1, 2,...., i,.... n) with i=1,...n, where for each i ( BK) all the following conditions hold:
1. 2.
BK
BKi
1 BK i BK
We indicate with D the set of learning objects descriptions in ( ( BK)}.
BK
). D = {
i D
Based on the previous deinition of courseware low, it possible to deine a composite courseware with respect to a request . Deinition 5. Let R = { iD, iBK}, with i=1..k, be a set of learning objects i, and D, , be a request for a courseware, such that iD, iBK, D and BK are modeled as concept BK descriptions in a DL with regard to an ontology T. A composite courseware for = D, BK, with respect to R, denoted ( ,R), is a courseware low such that for each j in the courseware low, D = { j D| j ( ,R)}, covers D. At each stage of the algorithm teacher, the learning objects to be added to the lesson low are searched for only within the current usable learning objects. In particular, we deine a usable learning object as follows: Deinition 6. Given a courseware low ( BK) =( 1, 2,..., n), we say that a learning object is a usable learning object u for ( BK) if and only if
Figure 1. System architecture
1. 2.
(
BK
BK
)
u
) =( 1, 2,..., n,
) is a courseware low.
A usable learning object u for ( BK) is a learning object that can be used after the user beneits from ( BK), that is, its required background knowledge is provided by ( BK). Actually, given a courseware low, several usable learning objects exist. Deinition 7. Given a courseware low ( BK) and a set of learning objects R = { i } we call usable set for ( BK), the set of all the iR such that i is a usable learning object for ( BK). U ( BK) ={ iu | iu is a usable learning object for ( BK)}. The usable set is hence the set of all the learning objects that can be used after the user beneits from a courseware low.
Figure 2. An excerpt of the skills ontology
Figure 3. An excerpt of the learning ontology
Figure 4. Learning objects descriptions
A System for Integrated Semantic-Based Composition

The described framework has been integrated in a prototype system supporting the whole process, sketched in Figure 1. A skills ontology is used as vocabulary through which internal personnel proiles and needed tasks are formally described. A different Learning Ontology is instead used to describe the domain of learning and learning objects and the requests for courseware descriptions. Both of the ontologies are about skills domain, but they look at it from different perspectives: the skills ontology stores information about what an employee knows; the learning ontology describes what learning objects teach. After the explanation process, it is possible to know which knowledge the assigned employees lack. Such information represents what they should learn through the learning process. Consider, for example, the need for a Process Engineer endowed with communication skills and the available employee John: Engineer, specialized in Plant Management and inclined to communication, both stored in the skills ontology according to the following formalization:
Task T = ProcessEngineer6hasAbilityhasAbility.CommunicationSkills John P = EngineerhasAbilityhasAbility.CommunicationSkills specializedspecialized.PlantManagement
T and P refer to the example skills ontology in Figure 2. By applying the algorithm Assign(P,T, T) we obtain the explanation of lacking skills: H = specializedspecialized.ProcessControl. Such learning need has to be mapped, according to the vocabulary of the example learning ontology in Figure 3, in a learning request for an automatic courseware composition: = givesSpecialization givesSpecialization.ProcessConD trol. In Figure 4, examples of the learning objects stored in the learning ontology are shown. According to such formalization, the description component of each learning object (the iller of the role hasDescription) is evaluated for the creation of the courseware satisfying the learning request D. The background knowledge component (the iller of the role requiresBackgroundKnowledge) is instead used to evaluate whether the individual proile at hand suficient prerequisites to beneit from that module. By looking at the given example, it is easy to see that we need some mapping rules to convert the properties expressing knowledge needs in the skills ontology in the properties expressing teaching actions in the learning ontology. In particular, we need mapping rules, which map patterns of expressions written according to the skills ontology into corresponding patterns in the learning ontology. Those mapping rules could be modeled as axioms in a mapping ontology, taking our approach out of the ALN OWL-DL subset in which the algorithms we presented so far work. Such language, though limited in expressiveness, has on the other side the advantage to keep acceptable the computational complexity of the proposed algorithms. In order not to lose this advantage, we preferred to keep both the skills and the learning ontology in an ALN DL and to write the rules out of the ontologies. When the need for performing a task arises, the management starts either the process of assignment to one individual or of creation of an ah-hoc team. The system makes calls to MatchMaking Service (MAMAS) (http://dee227.poliba.it:8080/ MAMAS-tng) in order to exploit the inference services. MAMAS provides an assignment (either employee or group) together with a formal explanation of the reasons why the match is not perfect. Such explanation represents the lacking skills to be covered by a learning process the assignees (one or more) have to start; this explanation is then translated into a learning request according to the mapping rules. On the basis of the formalized request, MAMAS is called again to automatically compose the personalized courseware endowing the assignees
with the skills needed for the task. If a courseware completely fulilling the learning need is not found, MAMAS will return an explanation about the skills still missing.
System Behaviour
In this section, the whole process of composition of skills and learning objects is detailed with the help of an example. Let us suppose a company has the problem to assign the realization of a task according to the following simple speciications: engineers, with lead-role, negotiation and communication skills and at least two years experience. The task requires process control, Web technology, process plan, and Enterprise Resource Planning (ERP) systems knowledge and orientation to team work. Such a request can be formalized in DL as: T= Engineerspecializedspecialized. (ProcessControl WebBasedTechnology ERPsystem ProcessPlanning) hasAbility hasAbility.(NagotiationSkills CommunicationSkills TeamCoordinator Lead-Role) hasExperiencehasExperience.( 2 years) Suppose now the four individuals described in the following are available: Julia: Julia is a process engineer specializing in business application, able to lead a team and coordinate meetings. She got a masters degree recently. Tom: Tom is a Web-oriented programmer. His favourite programming languages are Java and C++, used for Web-based applications. Richard: Richard is an engineer specializing in ERP systems. He has leading ability, particularly for working teams. He has three years work experience. Alison: Alison is an account manager specializing in IT-consulting and writing technical documents. She has a good knowledge of both English and German. The above employees are candidates to solve the task, and their proile can be formalized as DL concepts: P1 (Julia) = ProcessEngineerspecialized.BusinessApplication
hasAbility.(MeetingsCoordinationTeamCordinator) NewGraduate
P2 (Tom) = WebProgrammerspecialized.WebBasedTechnology programmLanguageKnowledge.(JavaC++)
P3 (Richard)= Engineerspecialized.ERPsystemhasAbility.(Lead-roleTeamCordinator) hasExperience.( 3 years)

P4 (Alison) = AccountManagerspecialized.(ITconsultingTechnicalWriting) hasLanguageKnowledge.(EnglishGerman)
In case management decide to assign the task T only to one employee, the algorithm Assign(P,T,T) is applied, with P={P1, P2, P3, P4}. Assign chooses the candidate minimizing U: Richard. Richard is characterized by the following solution to the contraction and the abduction problem, respectively: G P3 = K P3 = Engineerspecialized.ERPsystemhasAbility.(Lead-roleTeamCordinator) hasExperience.( 3 years) H P3 = specialized.ProcessPlanning hasAbility.(NegotiationSkills CommunicationSkills) Such results show that Richards proile is consistent with the required task, so no contraction is needed and the concept K P3 still equals P3. HP3 shows instead that, though Richard represents the best individual able to cover the task, he lacks negotiation and communication skills and a specialization in process planning. According to the deinitions given earlier in the computed utility for such an assignment is UP3=0.23076922. The analogously computed utilities of the other proiles are: UP1=0.25, UP2=0.9230769, UP4=0.84615386. HP3 represents the learning need to be covered by an automatically composed courseware, but it needs then to be rewritten in the learning ontology vocabulary, according to the rules contained in the mapping ontology in the following learning request: = givesSpecialization.ProcessPlanning teachAbility.(NegotiationSkillsCommunicationSkills) As a result of the automated courseware composition process, Richard is asked to learn MeetingCoordinationCourse and ProcessPlanningCourse, to be learned in the speciied order. Richard holds the background knowledge needed to attend MeetingCoordinationCourse and, thanks to such fruition, acquires the background knowledge needed to attend ProcessPlanningCourse. After the fruition of both courses, Richard covers all the skills he missed for performing the task he was selected for.
Future Trends
The Semantic Web initiative (Berners-Lee et al., 2001) put the tricky challenge of endowing the World Wide Web with a shared interpretation of data. Many steps have been made in this direction and, even if the complete realization of such an ambitious project is a long way off, the possibilities opened by the objectives achieved so far are noteworthy. In particular, the vision of knowledge-intensive companies is strongly affected by the chance to exploit novel techniques for the management of their most strategic asset: competencies. The standardization process in knowledge representation carried out on the Semantic Web mainstream puts the basis for collaborative environments representing a new frontier for global economy. By sharing a common knowledge on competencies and exploiting the Semantic Web-enabled technologies to exchange or produce such an economically appealing good, companies could give up to their traditional borders and take advantage of the resulting global knowledge market.
Conclusion and Future Work

In this chapter, a system and an approach to integrated semantic-based composition of skills and learning needs have been proposed to help knowledge-intensive companies in holding, retrieving, and creating competencies. When a task has to be assigned to one or more individuals, the system supports both the search for skills among internal personnel and the composition of personalized training processes about the skills unavailable in the company. The approach exploits non-standard reasoning services from description logics in order to explain the skills still uncovered after the search process (concept abduction) and to revise the request for skills in case all the available employees have proiles conlicting with the required one (concept contraction). A minimality criterion on a function taking both belief revision and explanation into account is proposed to carry out the selection among internal personnel. On the basis of the explanation of lacking skills, a courseware is automatically generated by retrieving a sequence of learning objects such that their composition satisies the learning need as far as possible. The improvement of the learning component of the proposed system is the object of our current studies. In particular, we are evaluating methods for taking into account belief revision in the automatic composition of courseware and for making the learning process personalized inside the selected team, in case an ad-hoc team is created for performing the needed task. Finally, the proposed function is continuously being tested with real data from human resource department of knowledge-intensive companies.
References
Ajami, K. (2004). Specifying and implementing interoperable and reusable learning objects: One step beyond. In Proceedings of International Conference on Information and Communication Technologies: From Theory to Applications (pp. 111-112). Baader, F., Calvanese, D., McGuinness, D., Nardi, D., & Patel-Schneider, P. (Eds.). (2002). The description logic handbook. Cambridge, MA: Cambridge University Press. Baader, F., Horrocks, I., & Sattler, U. (2003). Description logics as ontology languages for the Semantic Web. In D. Hutter & W. Stephan (Eds.), Festschrift in Honor of Jrg Siekmann, Lecture Notes in Artiicial Intelligence. SV. Becerra-Fernandez, I.(2000). The role of artiicial intelligence technologies in the implementation of people-inder knowledge management systems. Knowledge-Based Systems, 13(5), 315-320. Bennacer, N., Bourda, Y., & Doan, B. (2004). Formalizing for querying learning objects using OWL. In Proceedings of International Conference on Advanced Learning Technologies (pp. 321-325). Berners-Lee, T., Hendler, J., & Lassila, O. (2001).The Semantic Web. Scientiic American, 248(4), 34-43. Brachman, R., & Levesque, H. (1984). The tractability of subsumption in frame-based description languages. In Proceedings of the Fourth National Conference on Artiicial Intelligence (AAAI-84) (pp. 34-37). Cabezuelo, A., & Beardo, J. (2004). Towards a model of quality for learning objects. In Proceedings of International Conference on Advanced Learning Technologies (pp. 822-825). Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., & Mongiello, M. (2003a). Concept abduction and contraction in description logics. In Proceedings of the 16th International Workshop on Description Logics (DL03), vol. 81 of CEUR Workshop Proceedings, 46-55. Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., Mongiello, M., & Mottola, M. (2003b). Finding skills through ranked semantic match of descriptions [Special issue]. Journal of Universal Computer Science, 168-173. Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., Mongiello, M., & Mottola, M. (2003c). A formal approach to ontology-based semantic match of skills descriptions. Journal of Universal Computer Science, 9(12), 1437-1454. Colucci, S., Di Noia, T., Di Sciascio, E. Donini, F., & Mongiello, M. (2005c). Concept abduction and contraction for semantic-based discovery of matches and negotiation spaces in an E-marketplace. Electronic Commerce Research and Applications, 4(4), 345-361. Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., Mongiello, M., & Piscitelli, G. (2004). Semantic-based approach to task assignment of individual proiles. Journal of Universal Computer Science, 10(6), 723-730.
Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., Piscitelli, G., & Coppi, S. (2005a). Knowledge based approach to semantic composition of teams in an organization. In Proceedings of the 20th Annual ACM (SIGAPP) Symposium on Applied Computing SAC-05 (pp. 1314-1319). Colucci, S., Di Noia, T., Di Sciascio, E., Donini, F., & Ragone, A. (2005b). Semantic-based automated composition of distributed learning objects for personalized e-learning. The Semantic Web: Research and applications. 2nd European Semantic Web Conference, (Vol. 3532, pp. 633-648). Cormen, T. H., Leiserson, C. E., & Rivest, R. L. (1990). Introduction to Algorithms. The Massachusetts Institute of Technology. DAML+OIL (2001). DAML+OIL speciications. Retrieved from ww.daml.org/2001/03/ daml+oil-index.html. Di Noia, T., Di Sciascio, E., & Donini, F. (2004a). Extending semantic-based matchmaking via concept abduction and contraction. In Proceedings of the 14th International Conference on Knowledge Engineering and Knowledge Management (EKAW 2004), 307-320. Di Noia, T., Di Sciascio, E., Donini, F., & Mongiello, M. (2003). Abductive matchmaking using description logics. In Proceedings of the 18th Intl. Joint Conference on Artiicial Intelligence (IJCAI 2003) (pp. 337-342). Di Noia, T., Di Sciascio, E., Donini, F., & Mongiello, M. (2004b). A system for principled matchmaking in an electronic marketplace. International Journal of Electronic Commerce, 8(4), 9-37. DublinCore (1999). DublinCore metadata element set, (version 1.1): Reference description. Retrieved from http://dublincore.org/documents/1999/07/02/dces/. Galil, Z. (1986). Eficient algorithms for inding maximum matching in graphs. ACM Computer Survey, 18(1), 23-38. Grdenfors, P. (1988). Knowledge in lux: Modeling the dynamics of epistemic states. Cambridge, MA: Bradford Books, MIT Press. Garro, A., & Palopoli, L. (2003). An XML multi-agent system for e-learning and skill management. In Agent Technologies, Infrastructures, Tools, and Applications for EServices, LCNS (Vol. 2592, pp. 283-294). Gasevic, D., Jovanovic, J., & Devedzic, V. (2004). Enhancing learning object content on the Semantic Web. In Proceedings of International Conference on Advanced Learning Technologies (pp. 714-716). Gronau, N., & Uslar, M. (2004). Integrating knowledge management and human resources via skill management. In Proceedings of I-KNOW 04 (pp. 135-142). Hamel, G., & Prahalad, C. K. (1990). The core competence of the corporation. Harvard Business Review, May-June, 79-91. Hefke, M., & Stojanovic, L. (2004). An ontology-based approach for competence bundling and composition of ad-hoc teams in an organization. In Proceedings of I-KNOW04 (pp. 126-134).
Hillier, F. S., & Lieberman, G. J. (1995). Introduction to operations research. MGH. IEEE (2002). IEEE standard for learning object metadata, std 1484.12.1-2002 edition. IEEE (2003). IEEE standard for learning technology-learning technology systems architecture (LTSA), std 1484.1-2003 edition. IMS. (2001). Ims, learning resource meta-data best practices and implementation guide (version 1.1)Final speciication. Retrieved from http://www.imsproject.org/metadata/mdbestv1p1.html Ip, A., Young, A., & Morrison, I. (2002). Learning objectsWhose are they? In Proceedings of 15th Conference of the National Advisory Committee on Computing Qualiications (pp. 315-320). Kennington, J., & Wang, Z. (1991). An empirical analysis of the dense assignment problem: Sequential and parallel implementations. ORSA Journal on Computing, 3(4), 299-306. Lau, T., & Sure, Y. (2002). Introducing ontology-based skills management at a large insurance company. In Modellierung 2002, Modellierung in der Praxis - Modellierung fr die Praxis (Proceedings), Tutzing, Germany, March 25-27, 2002, pp.123-134. Liu, P., & Dew, P. (2004). Using Semantic Web technologies to improve expertise matching within academia. In Proceedings of I-KNOW 04 (pp. 370-378). McGuinness, D., Fikes, R., Hendler, J., & Stein, L. (2002). DAML+OIL: An ontology language for the Semantic Web. IEEE Intelligent Systems, 17(5), 72-80. OLeary, D. (1998a). Enterprise knowledge management. IEEE Computer, 31(3), 54-61. OLeary, D. (1998b). Using AI in knowledge management: Knowledge bases and ontologies. IEEE Intelligent Systems, 13(3), 34-39. OWL. (2004). Retrieved from www.w3.org/TR/owl-features/ OWL-S. (2004). The owl services coalition. Retrieved from www.daml.org/services/owls/1.0/owl-s.html Saip, H. B., & Lucchesi, C. (1993). Matching algorithms for bipartite graph (Tech. Rep. No. CC -03/93). Universidade Estudal de Campinas, Departamento de Cincia da Computao. Sanchez, S., & Sicilia, M. (2004). On the semantics of aggregation and generalization in learning object contracts. In Proceedings of International Conference on Advanced Learning Technologies (pp. 425-429). EEE. SCORM. (2004). Advanced distributed systems (adl) lab, sharable content object reference model (SCORM). Retrieved from http://www.adlnet.org/index. cfm?fuseaction=scormabt Staab, S., Studer, R., Schnurr, H.P., & Sure, Y. (2001). Knowledge processes and ontologies. IEEE Intelligent Systems, 16(1), 26-34. Sugawara, K. (2003). Agent-based application for supporting job matchmaking for teleworkers. In D. Patel, S. Patel, & Y. Wang (Eds.), 2nd IEEE Intl. Conference on Cognitive Informatics (ICCI 2003) (pp. 137-142). London: IEEE Computer Society.
Sure, Y., Maedche, A., & Staab, S. (2000). Leveraging corporate skill knowledgeFrom proper to ontoproper. In Proceedings of the Third International Conference on Practical Aspects of Knowledge Management, October 30-31, 2000, Basel, Switzerland, pp 22:1-22:9. Sycara, K., Paolucci, M., Ankolekar, A., & Srinivasan, N. (2003). Automated discovery, interaction and composition of Semantic Web services. Journal of Web Semantics, 1(1), 27-46. Trastour, D., Bartolini, C., & Priest, C. (2002). Semantic Web support for the businessto-business e-commerce lifecycle. In Proceedings of International World Wide Web Conference (WWW) 02 (pp. 89-98). Veit, D., Muller, J., Schneider, M., & Fiehn, B. (2001). Matchmaking for autonomous agents in electronic marketplaces. In Proceedings of International Conference on Autonomous Agents 01 (pp. 65-66). Vossen, G., & Jaeschke, P. (2003). Learning objects as a uniform foundation for e-learning platforms. In Proceedings of International Symposium on Database Engineering and Applications Symposium (pp. 278-287). Yimam, D., & Kobsa, A. (2000). Expert inding systems for organizations: Domain analysis and the demoir approach. In M.S. Ackerman, V. Pipek, &V. Wulf (Eds.) Beyond knowledge management: Sharing expertise (pp. 327-358). Boston: MIT Press.
Endnote
1
Without loss of generality here we consider only the information needed for a semantic discovery and composition.
Semantic Web Technologies in the Recruitment Domain 299
Chapter XIV
Semantic Web Technologies in the Recruitment Domain

Ralf Heese, Humboldt-Universitt zu Berlin, Germany Malgorzata Mochol, Freie Universitt Berlin, Germany Radoslaw Oldakowski, Freie Universitt Berlin, Germany
Abstract
Due to the large number of job offers published online, it is almost impossible for job seekers and job portals to gain an overview of the entire employment market. Since job offers lack semantically meaningful annotations, their location and integration into databases is extremely dificult. In this chapter, we demonstrate how the application of Semantic Web technologies can enable unambiguous identiication of concepts and relationships between concepts and how the e-recruitment process provides advantages for all participants in the market. When comparing job and applicant proiles, this above mentioned identiication through the use of a dedicated matching function is a key element for increasing the precision of search results provided by search engines. Furthermore, it allows for automating and supporting recruitment processes. In this chapter, we present an application scenario and our prototypical implementation discussing the construction of a human resource ontology for annotating job offers and job applications and our matching function.
300 Heese, Mochol, & Oldakowski
Introduction
The cornerstone of an organizations success is the alignment of all corporate resources with business objectives and strategies (Wright & McMahan, 1992). The key asset of every organization is its employees. The crucial factor for businesses to compete effectively is inding the best person for a given task and developing and leveraging their skills and capabilities. Furthermore, the employer has to maximize the impact of training and educational efforts and to align the activities of the employees with corporate objectives as well as to retain top performers by an effective incentive management (Ferris, Hochwarter, Buckley, Harrell-Cook & Frink, 1999). Whereas most of the chapters in this book deal with employee development through organizational e-learning, this chapter focuses on human resource recruitment, in particular Web-based recruitment. We present an application scenario in which online recruitment processes are streamlined using Semantic Web technologies. Additionally, we describe our prototypical implementation of the technological infrastructure. Semantic Web technologies can support the unambiguous identiication of concepts and formally describe relationships between concepts thereby allowing representation of data in a more machine-understandable way. In our scenario, job position postings and job applicant proiles are annotated using controlled vocabularies combined with domain ontologies. This opens up new possibilities for better job posting discovery by search engines as well as more intelligent matching of open positions with candidate proiles, which no longer relies on the containment of keywords but exploits domain-speciic knowledge, leading to increased automation of the recruitment process. This chapter is structured as follows: In the second section we provide some background information concerning the current situation on the electronic job market and point to the shortcomings of todays online recruitment process. The third section contains a brief introduction into the fundamentals of the Semantic Web. Subsequently, in the fourth section, we describe a typical recruitment process seen from the perspective of a company as well as an applicant. We outline what impact the application of the Semantic Web technologies would have on each phase of the process. The ifth section deals with practical aspects of ontology engineering. It begins with a comparison of two ways of building ontologies: from scratch and through ontology reuse. Next, we give some insight into the structure of the human resources (HR) ontology we have developed as the building block for annotating job postings and applicant proiles. The sixth section is devoted to the prototypical realization of the scenario. Here, we describe the general scenario architecture as well as illustrate how the implemented semantic matching functionality enables search and ranking possibilities far beyond simple keyword-based algorithms. Finally, the seventh section summarizes the impact of the application of Semantic Web technologies in the electronic recruitment domain.
E-Recruitment
Over the last few years, the Internet has evolved into the primary recruitment medium. As reported in Keim et al. (2005), 90% of human resource managers in Germany rated the InCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
ternet as an important recruitment channel. One reason for this high rate is that the Internet, in particular, reaches applicants who are young and highly qualiied. Despite the fact that companies use more than one channel to publish their job openings, over half of all personnel recruitment is the result of online job postings (Monster, 2003). Although a large number of online job portals have sprung up, dividing the online labor market into information islands, employers still publish their job postings on a rather small number of portals in order to keep costs and administrative effort down. The hiring organizations simply assume that a job seeker will visit multiple portals while searching for open positions. Alternatively, companies can publish job postings on their own Web site (Mlder, 2000). This way of publishing, however, makes it dificult for job portals to gather and integrate job postings into their database. Thus search engines such as Google or Yahoo! have become vital in the job search. Furthermore, dedicated search engines, such as worldwidejobs.de (http://www.worldwidejobs.de/), are entering into the market, allowing detailed queries as opposed to the keyword-based search of current search engines. The quality of search results depends not only on the search and index methods applied. Further inluential factors include the processability of the used Web technologies and the quality of the automatic interpretation of the company-speciic terms occurring in the job descriptions. The deiciencies of a websites machine processability result from the inability of current web technologies, such as HTML, to semantically annotate the content of a given Web site. Therefore, computers can easily display the content of an HTML site, but they lack the ability to interpret the content properly. Our conclusion is that the information low in the online labor market is far from optimal. The publishing behavior of companies not only makes it almost impossible for a job seeker to get an overview of all the appropriate openings but also complicates or even prevents job offers from reaching a greater range of potential employees. In this chapter, we show how the shortcomings of the current situation in the online recruitment market can be overcome by utilizing a new Internet vision called Semantic Web, leading to higher market transparency and enhanced optimization of the recruitment process.
Semantic Web
The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries (World Wide Web Consortium, 2001). The term Semantic Web comprises a family of technologies that enable applications to generate and automatically process metadata describing distributed Internet resources. The objective is to use the Web as a global distributed database that can be deployed by applications to perform certain tasks automatically (Berners-Lee, Hendler, & Lassila, 2001). The development of the Semantic Web is a joint effort of scientiic institutions (Massachusetts Institute of Technology, Stanford, The Institute for Learning and Research Technology (ILRT), etc.) together with top businesses (Hewlett-Packard, IBM, Nokia, etc.) (Quan & Karger, 2004) and is led by the World Wide Web Consortium (W3C). To realize the vision of the semantic Web networked resources, for example, Web sites, are annotated by structured and machine-understandable metadata, which are assigned a well-deined meaning and are interpreted by means of logic rules.
Figure 1. Technology stack of the Semantic Web

Semantic Web applications
(6) Trust
(5) Logic Framework (4) Ontology (3) Semantics (2) Data Encoding (1) Identification of Resources
Digital signatures
Inference algorithms DAML + OIL, OWL RDF, RDF Schema XML, XML Schema Unicode, URI
The fundamental technologies of the Semantic Web are presented in this section. For naming Web resources and terms uniform resource identiiers (URIs) are used as a global identiication mechanism (layer 1). At the syntactical layer, the exchangeability of semantic information is guaranteed by the use of the XML mark-up language (layer 2). The resource description framework (RDF) (Beckett & McBride, 2004) speciies a data model for publishing metadata on the web (layer 3) and utilizes XML as serialization syntax for data transmission. With RDF schema (RDF/S), the user can develop his own metadata vocabularies. The Web ontology language (OWL) (Dean & Schreiber, 2004) at layer 4 of the technology stack extends RDFS with terms and concepts for an even more expressive knowledge representation in the form of ontologies. Ontologies formally name and describe the central concepts of an application domain and relationships between them. Since ontologies conceptualize a part of the real world, they can serve as means for communication between the users of an application domain. Generic ontologies (upper ontologies) deine general concepts and relationships between them and are currently being standardized by international standardization committees, for example, SUMO by the IEEE standard upper ontology working group (Suggested Upper Merged Ontology (Sumo), 2005), while domain-speciic ontologies are being developed by communities of domain experts, for example, the gene ontology by the Gene Ontology Consortium (The Gene Ontology, 2005) Layer 5 provides the means for specifying sets of deductions that can be made from a collection of data, together with formalisms for describing steps taken to reach conclusions from given facts. Finally, the trust layer deines mechanisms upon which applications decide whether or not to trust the given information. These two topmost technology layers are currently being researched and simple application demonstrations are being constructed, e.g. (Bizer, Cyganiak, Gauss & Maresch, 2005). These new Internet technologies (Semantic Web technologies) are maturing and moving out from academic applications into the wider industry. This is demonstrated on one hand by the strong and growing interest in these topics shown by various commercial sectors
and on the other hand by public bodies like the European Commission, which supports the distribution and transfer of these technologies to the business world. One such activity is the Knowledge Web EU Network of Excellence, which has formed an industry board (http://knowledgeweb.semanticweb.org/o2i) to promote greater awareness and faster uptake of Semantic Web technology within European industry. The Knowledge Web together with this board aims to transfer technology from research to industry, to promote ontological technologies, propose technological recommendations, and to meet industrial application needs. Within this context, they have identiied some key sectors for the early uptake of Semantic Web technologies: human resources, health and life sciences (Nixon & Mochol, 2005). We have picked up on the issue of human resource management as a potential earlier adopter and developed a Semantic Web-based scenario for e-recruitment for which we implemented a prototype.
Supporting Recruitment Processes with Semantic Web Technologies

Apart from intermediaries, the actors in a recruitment scenario are employers and job applicants. Figure 2 displays the recruitment processes from the viewpoint of an organization as well as an applicant. The actions performed by a company personnel manager or a job applicant are very similar. For instance, both processes contain the generation of a job/applicant proile and the matching of job descriptions with candidate proiles. Therefore, the next sections take mainly the viewpoint of the company into consideration. Each following subsection describes a step of the recruitment process and begins with a description of the current situation. Afterward, we illustrate the utilization of Semantic Web technologies in each step.
Requirements Analysis
The requirements for an open position are usually conceived by the operating department in cooperation with the human resources department. Nowadays, the requirements for a job posting are commonly written in free text as illustrated on the left hand of Figure 3. The usage of free text limits the machine processability of postings in the later phases of the recruitment process. In contrast to a free text description, the utilization of a widespread shared language in form of a set of controlled vocabularies as shown on the right hand of Figure 3 would facilitate communication between all parties involved. Personnel managers would use the controlled vocabularies to semantically annotate the details of a job posting, in the same way job applicants would annotate their application. This annotation step opens up the potential for the automation of a variety of tasks along the recruitment process such as inding job offers in a speciic sector of the job market.
Figure 2. Processes in the recruitment domain
Employer
Analyse the requirements
Interview for a job Write and mail applications Search manually or publish the profile Generate a job profile
Publish the job posting
Offering a job position
Publication of the Job Posting

There are a number of channels to publish a job posting on the Internet: (1) commercial job portals, such as Monster (http://www.monster.com/), JobPilot (http://www.jobpilot.com/), or StepStone (http://www.stepstone.com/) compete to publish job postings for a publication fee, (2) the employers own Web site, and (3) state labor agencies such as the German Fed-
Searching a job
Job Applicant
Receive and preselect applications Final recruitment decision
Figure 3. Example of a job description and an extract of the skill ontology
eral Employment Ofice (http://www.arbeitsagentur.de/). Various job portals target different groups and are mostly speciic to certain geographic regions. Since there are a lot of portals, a job seeker is unable to get an overview over the entire job market. Additionally, job portals can hardly gather job offers from employers Web sites. To improve market transparency several public bodies like the German Federal Employment Ofice (BA) and the Swedish National Labour Market Administration (AMS) (http://www.ams.se/) have set up projects to integrate open positions into a central database. Furthermore, they intend to establish a complete worklow for e-recruitment. In both projects, participating employers use terms from a controlled vocabulary to categorize their postings and submit them to a central database using variations of the HR-XML (http://www.hr-xml.org/) data format, for example, HR-BA-XML and HR-XML-SE, respectively. The collected postings are published through a central portal and are additionally forwarded to commercial job portals. The problem with these projects is that the entire market depends on a single central database an approach to which many market participants object and which can be quite error-prone, as experienced by the German project (Crosswater Systems Ltd., 2003). Using Semantic Web technologies to semantically annotate job postings would increase market transparency together with avoiding the bottleneck of a central database. Organizations would publish their annotated job postings directly on their corporate Web sites using controlled vocabularies and the RDF data format. The postings could then be crawled directly by job portals and thus the centralized approach would be replaced with a distributed one. Consequently, all job portals would operate on the same information and postings would reach more applicants leading to higher market transparency. Another beneit from having postings annotated with terms from a controlled vocabulary is that the terms can be combined with background knowledge about an application domain. Job portals could offer semantic matching services that allow the comparison of job position postings with applicant proiles based on domain-speciic knowledge instead of merely relying on keywords, like traditional search engines do. For example, if Java programming skills are required for a certain job and an applicant is experienced in Delphi, the matching algorithm would consider this persons proile a better match than someone elses who has skills in SQL, because Delphi and Java are closer related than SQL and Java.
Receiving and Preselecting Job Applications

According to Keim et al. (2004), a general tendency has been emerging to move away from the classical job applications toward an electronic version. The latter has the advantage that it allows both sides to check basic requirements automatically, thereby reducing the number of applications to be reviewed manually. About 59% of large-scale enterprises request that job applicants store and maintain their personal proile on the company server for a while (Keim et al., 2005). Enterprises like IBM (https://forms.bpfj.intronet.com/arm/ibm/emea/) and Volkswagen (https://www.vw-personal.de/www/de/bewerbung/onlinebewerbung.html) commonly require online forms to be submitted by job searchers. From the applicants viewpoint, this procedure is time-consuming and quite laborious, since they have to re-enter their CV data into each new application form.
Both applicants and employers beneit from the semantic annotation of job applications in addition to the classic free text application. Applicants could reuse their RDF (see third section) proiles and send them to different employers instead of having to ill in numerous Web forms. Employers use the annotations for automating the preselection process by matching their minimal requirements with the applications. After the employment of a candidate, the annotated job application could serve as the basis data for the company human resource management system. Another issue of this phase is prooing testimonials of experience and education. Testimonials could be attached to applications in the form of RDF statements that are digitally signed by the issuing university or organization. The process of validating digitally signed testimonials can be automated, further reducing costs.
Interviews and Recruitment Decision

Face-to-face contact with candidates is indispensable for the inal recruitment decision. In the interviews, the personnel manager rates the soft as well as technical skills of a candidate. Although social competencies can be expressed using ontologies, we believe that an automatic evaluation of these competencies is of little importance. Thus, this phase of the recruitment process will still be done manually in the future.
Applying the Semantic Web to Human Resource Management

In a Semantic Web-based recruitment scenario, the data exchange between employers, applicants, and job portals is based on a set of vocabularies that provide shared terms to describe occupations, industrial sectors, and job skills. This common vocabulary is deined in our human resources ontology (HR ontology). Ontologies represent domain-speciic knowledge and might be used to determine semantic similarity between resources such as job applications and job offers. Hence, ontologies are not only used to annotate job postings and applications but for semantic matching as well a technique that combines annotations using controlled vocabularies with background knowledge about a certain application domain. Ontologies form a key component of our job portal allowing the semantic comparison of job postings with candidate proiles. However, the development of ontologies is far from trivial and the decision to construct ontologies from scratch has to be well considered. In the following, we describe a methodology for ontology reuse and a model for cost estimation of the ontology development process. Thus, ontology engineers can turn to recommendations of how to proceed during ontology development and different kinds of support can be engaged.
Reuse or from Scratch?

In the case of HR ontologies, as in other ields, the potential size and complexity of the ontology requires determining a suitable ontology building methodology. There are two general ways of building ontologies: from scratch or by ontology reuse. The latter is the process by which existing knowledge sources (ontological sources) are recruited to generate new ontologies. Since ontologies are one of the main parts of our HR scenario and due to the fact that ontology development is a very time-consuming process we had to consider which way of ontology building is the proper one in our case. This decision is not a speciic challenge of our scenario. On the contrary, it is a general dificulty of every ontology-based application. Therefore, every ontology engineer, at the beginning of the ontology development process, has to estimate whether building a new ontology from the ground up or using existing sources is more eficient. Ideally, this decision is supported by comparing the costs with the beneits of these two alternatives. The ONTOlogy COst Model (ONTOCOM), described in detail in Paslaru and Mochol (2005b), estimates the costs involved in the ontology development process of the two methods (from scratch or reuse). ONTOCOM is a parametric method with a cost calculation formula to determine the cost and price (in person/monthsPMs) of the ontology development process based on the size of the ontology and other different cost drivers, for example, tool support, ontology complexity, experience of the ontology engineer, domain experts experience). On the one hand, the costs of the ontology building from scratch (PMscratch), which includes sub-tasks like speciication, conceptualization, implementation, instantiation, and evaluation can be determined, while, on the other hand, ONTOCOM calculates the price of the ontology reuse (PMreuse) by estimating the work relevant to the reuse of existing sources for the generation of new ontologies, for example, costs of inding, evaluating, and adapting existing sources to the requirements of the target ontology. If the reuse process is more economical than ontology building from scratch, the ontology engineer is likely to create the target ontology from existing knowledge, otherwise a wholly new construction of the ontology is recommended. Reusing ontologies should reduce engineering costs since it avoids re-building existing ontologies. Unfortunately, it still involves signiicant costs and work, which currently may outweigh its beneits. Based on the ONTOCOM calculation (PMreuse < PMscratch) and in an effort to support common commercial practices and to maximize the integration of job seeker proiles and job postings from various organizations, our HR ontology reuses established domain-speciic standards and classiications.
Ontology Reuse
Since ontologies are understood as a means for a shared knowledge conceptualization, reusing existing ontological sources increases application interoperability both on the syntactic and on the semantic level. It is generally accepted in the Semantic Web community that building ontologies from scratch is a challenging, time-consuming, and error-prone task. Moreover, the development of new ontologies fails to tap the full potential of the knowledge sources available on the Web. This situation can be also explained by the dificulties related to building reusable ontological sources, which must strike a balance between rich
conceptualization and application speciicity (Gruber, 1995) and the very fact that this issue has been poorly explored in existing engineering methodologies. Though most methodologies for ontology building mention the reuse of existing knowledge sources, they do not deine precisely how ontology discovery and the subsequent evaluation of candidate ontologies should be performed nor do they provide tool support so as to allow at least the partial automation of the reuse process. Within this context Paslaru & Mochol (2005a) present a methodology for ontology building whose focus is on the reuse and reusability of ontologies, which is additionally supported by a semi-automatic tool (Liebsch, 2005). The methodology starts off by considering the vocabulary of the sources (concepts, relations, and axioms), which contains lists of potential ontological primitives: a list of concept names, a list of properties, and a list of axioms. After merging the source vocabularies, generating separate lists of ontological primitives and eliminating duplicates the syntactical similarities between concept names belonging to different source ontologies are computed. This phase results in a list of concepts ranked in terms of concept frequencies, source priority, and application requirements. After identifying relevant concepts, the user selects the relevant relationships, which can be added incrementally to the ontology until a certain level of complexity has been achieved. This methodology based on an incremental process that concentrates on the concepts represented in the input sources and takes into account additional information like semantic relationships and axioms, depending on application needs, has been used for the building of our HR ontology.
Development of the HR Ontology

Currently, there is a wide range of standards to process documents and classiication schemas developed for recruitment as well as a number of taxonomies available for the description of skills, classiication of job proiles, and business sectors. The irst step toward the development of the HR ontology is the identiication of the sub-domains of the application setting (e.g., professional and educational skills, types of professions, and business ields) and inding useful knowledge sources that cover them. While developing our HR ontology we evaluated 25 ontologies in 4 representational languages (Bizer, Mochol, & Westphal, 2004) and reused some of the most relevant classiications being deployed by national and international agencies: Standard Occupational Classiication (SOC): Classiies workers into occupational categories (23 major groups, 96 minor groups, and 449 occupations). Profession Reference Number Classiication (BKZ): German version of SOC system. Classiication of Industrial Sector (WZ2003): German classiication standard of economic activities. North American Industry Classiication System (NAICS): Classiication and deinition of industry sectors for Canada, Mexico, and the United States to facilitate economic analyses.
Figure 4. HR ontology
HR-XML: Library of approximately 75 interdependent XML schemes produced by the HR-XML Consortium, which deine data elements for major processes and particular transactions in the human resource management as well as options and constraints governing the use of these elements. HR-BA-XML: Extended version of HR-XML developed by the German labor agency. KOWIEN: Skill ontology, which describes different human abilities and prospects.
These sources cover different domains and vary in their degree of formality and granularity. They are represented in different natural languages (English, German) and in a broad range of formats, for example, text iles (BKZ, WZ2003), XML schemas (HR-XML, HR-BAXML), and ontology languages (KOWIEN). Dealing with different natural languages also complicates the process of ontology development, because human-readable concept names in various languages were required to make the ontology usable in different job portals. Another important characteristic of the candidate ontologies was the absence of semantic relationships among concepts (except for the KOWIEN ontology). Consequently, we focus on how vocabularies (concepts and relations) can be extracted and integrated into the target ontology. The selection of the applicable ontologies is followed by their customization and integration into the target ontology. The HR ontology is thus divided into sub-ontologies (see Figure 4) to be used in both job posting and job application descriptions. It is mainly based on the German version of the HR-XML standard, HR-BA-XML from which we adapted the job position seeker and job position posting segments. Due to the application setting, classiication standards like the occupation classiication (BKZ) and the classiication of industrial sectors (WZ2003) used in the industry sub-ontology have been completely integrated in the new ontology. A section of the KOWIEN ontology is used to deine concepts representing skills required for an open position as well as job seeker skills. To extract the relevant fragments of KOWIEN, we compiled a small conceptual vocabulary (of approximately 15 concepts) from various job portals and job procurement Web sites and matched these core concepts to the source ontology. The human resource ontology needs to be represented in a highly formal representation language since our semantic matching approach makes use of it. To this end, the implementation of the human resources ontology was carried out by translating several semi-structured
input formalisms and manually coding text-based classiication standards to the Semantic Web ontology language OWL.
Implementation of a Job Portal

Our recruitment scenario implies three basic roles: information providers, information aggregators (information integration, matching, and presentation), and information consumers. Additional to the free text descriptions, information providers (i.e., hiring organizations) publish open positions and background information to their organization in the RDF format using controlled vocabularies. They have two different approaches for publishing annotated job postings depending on their existing software infrastructure. If they use database standard software in the backend, they can export related data directly into RDF using mapping tools like Bizer and Seaborne (2004). If they do not use any enterprise software to manage open positions, they can annotate existing HTML versions of their postings using annotation tools like (Kalyanpur, Hendler, Parsia & Golbeck, 2003) and publish RDF version of their postings using, for example, the RDF NetAPI (Moore & Seaborne, 2003) or RAP (Oldakowski & Bizer, 2004). Information consumers (job seekers) use one portal of their choice as a central access point to all published open positions and background information regarding employers, instead of collecting information fragments from different sites. Information aggregators (semantic job portals) use a directory service to locate information providers. They also use a crawler to collect published data from the different providers and then a mapping engine to integrate crawled information published using different vocabularies, for example, D2R MAP (Bizer, 2003) for mapping relational data into RDF. The semantic portals also provide semantic matching services for matching applicant proiles with job postings, which utilize background knowledge included in the HR ontology. Furthermore, the usage of URIs as an employer identiication mechanism allows the enrichment of postings
Figure 5. Scenario architecture
with information about employers from different external information sources. The gathered and matched information can be presented to the users (e.g., job seekers) according to their preferences, for example, desired format, language, or level of aggregation. A semantic portal based on the architecture presented in Figure 5 will have to store a huge amount of collected RDF data and provide high performance access to this data. An overview of specialized databases and query languages can be found in Beckett (2003).
Semantic Matching
Whereas traditional similarity computation methods treat objects (like job descriptions) as sets of concepts drawn from a lat domain and calculate their similarity based on set intersection (shared concepts), more sophisticated approaches enhance these object models by adding a taxonomy describing the relationships among concepts leading to improved measures of similarity (Ganesan, Garcia-Molina & Widom, 2003). Semantic matching is a technique that combines annotations using controlled vocabularies with background knowledge about a certain application domain. In our prototypical implementation, the domain-speciic knowledge is represented by concept hierarchies for skills and skill levels, an occupational classiication, and a taxonomy of industrial sectors. Having this background knowledge of the recruitment domain (i.e., formal deinition of various concepts and speciication of the relationships between these concepts) represented in a machine-understandable format allows us to compare job descriptions and applicant proiles based on their semantic similarity (Poole & Campbell, 1995) instead of merely relying on the containment of keywords like most of the contemporary search engines do. In our scenario, as the matching result, a ranked list of best candidates for a given job position (and vice versa) is generated. In contrast to Sure, Maedche, and Staab (2000), our matching approach utilizes metadata of job postings and candidate proiles. Furthermore, we do not apply attribute vectors but determine the similarity of two concepts based on their distance in the given concept hierarchy. Inside both a job posting as well as an applicant proile, we group pieces of information into thematic clusters, e.g., information about skills, information regarding industry sector and occupation category, and, inally, job position details such as salary information, travel requirement, and so forth. In our approach, we compare each thematic cluster from a job posting with the corresponding cluster from an applicant proile. The total similarity between a candidate proile and a job description is then calculated as the average of the cluster similarities. The rationale for property clustering is to enable sorting of the matching result not only by the total similarity but also by each single cluster similarity. The cluster similarity, in turn, is computed based on the similarities of semantically corresponding atomic concepts from a job description and an applicant proile. However, if for a given concept in the job description there is more than one semantically corresponding concept in the applicant proile (e.g., several skills), only the one with the highest similarity lows into the cluster similarity. In other words, for each required skill, only the best match of all applicant skills is being considered. The taxonomic similarity between two atomic concepts c1 and c2 is determined by the distance dc(c1, c2) between them, which relects their respective positions in the underlying
Figure 6. Skill antology segment together with corresponding milestone values

m(5) = 0.0313 High level programming languages
m(6) = 0.0156
Object oriented
Imperative procedural
m(7) = 0.0.0078
Pure object oriented
Hybrid languages
COBOL
m(8) = 0.0039
Java
Visual Basic
C++
TURBO PASCAL
concept hierarchy (Sowa, 1984; Zhong, Zhu, Li & Yu, 2002). Consequently, the concept similarity is formally deined as:
sim c (c1 , c 2 ) = 1 d c (c1 , c 2 )
Every concept in a hierarchy is assigned a milestone value (Zhong, Zhu, Li & Yu, 2002), which is calculated with the formula:
milestone (n) = 1/ 2 k l (n)
Where k is a factor larger than 1 indicating the rate at which the milestone value decreases along the hierarchy and is assigned different values depending on the taxonomy depth; l(n) is the depth of the concept n in a given hierarchy. Since the distance between two given concepts in a hierarchy represents the path from one concept to the other one over the closest common parent (ccp), we calculate the distance as follows:
d c ( c1 , c 2 ) = d c ( c1 , ccp ) + d c ( c 2 , ccp ) d c ( c , ccp ) = milestone ( ccp ) milestone ( c )
This model implies two assumptions: the semantic differences between upper level concepts are bigger than those between concepts on lower hierarchy levels (in other words, two general concepts are less similar than two specialized ones) and that the distance between siblings is greater than the distance between parent and child.
Figure 7. A detailed matching description generated by the prototypical job portal
As an example, we determine the distance between two concepts from our skill ontology: Java and C. Figure 6 shows a snippet of our skills ontology together with the milestone values (with k=2) for the corresponding ontology levels. Since the closest common parent is High-level programming languages, the distance between Java and C is calculated as follows: dc(Java, C) = = = dc(Java, High level programming languages) + dc(C, High level programming languages) (0.0313-0.0039) + (0.0313-0.0078) 0.0509
Consequently, the similarity between these two concepts equals: simc(Java, C) = 1 0,0509 = 0,9491
Our approach also gives employers the opportunity to specify the importance of different job requirements. The concept similarity is then justiied by the indicated weight, that is, the similarity between more important concepts, such as the skills crucial for a given job position, will have greater inluence on the similarity between a job position posting and an applicant proile. Figure 7 shows an example of a detailed matching result of comparing a job posting with a candidate proile generated by our prototypical job portal, depicting similarity values at cluster and concept level. Our matching approach presented in this section was implemented in SemMF, a semantic matching framework (Oldakowski & Bizer, 2005) that is used as a foundation in the implementation of the job portal (see matching engine in Figure 5). SemMF is an easy-to-use and lexible open source framework for computing semantic similarity between objects represented as arbitrary RDF graphs, based on similarities of speciied nodes. In contrast to existing matchmaking frameworks such as Generic Request Architecture for Passive Provider Agents (GRAPPA) (Veit, 2003), which relies on information retrieval oriented similarity computation algorithms, SemMF exploits domain-speciic knowledge in the form of concepts taxonomies and is a pure Semantic Web application allowing it to work with native RDF data. Moreover, SemMF computation of taxonomic similarity between concepts is customizable according to the structure of the underlying concept hierarchy. An alternative similarity computation method is presented in Zhong et al. (2002). In addition to taxonomic concept similarity, this approach considers taxonomic similarity of relations between concepts, thus matching entire graphs. (Andersen, Bulskov, & Knappe, 2003) also determine the similarity of two concepts based on their respective positions in the graph. However, they do not only rely on concept subsumption but also analyze other kinds of connections between concepts by putting emphasis on the nodes shared by the concepts to be matched.
Summary
More and more job postings are being published on the Internet. Although a large number of recruitment portals index online postings, a job seeker cannot get an overview of all relevant openings on the labor market. Since companies prefer to publish job openings as free text on their own Web sites, job portals have dificulties automatically gathering and integrating them into their database. As a solution to these problems, we propose an e-recruitment process, which takes advantage of Semantic Web technologies. We suggest improvements for several steps of the process by enhancing job postings and job applications with semantic annotations using controlled vocabularies and describe our prototypical implementation of the scenario. The core components of the prototype are a human resource ontology derived from existing standards, a data publishing and integration infrastructure, and a retrieval engine relying on semantic matchmaking.
The need to improve data exchange between enterprises results in further standardization of data schemes and classiications. Regarding our recruitment scenario, the efforts are supported by public bodies such as state labor agencies or by leading enterprises and vendors of large ERP systems such as Systems Applications and Products in Data Processing (SAP) and Oracle/PeopleSoft. Web ontologies seem to be the proper means to consolidate their efforts, because they aim at a network-based exchange of information between enterprises. The analysis of the interest of the participants showed that the main actors in the employment marketemployers and job seekerswould both beneit from the realization of the scenario. Publishing a semantically enriched job posting on the Web site of the company would guarantee that the posting would be found by job seekers and job portals. Employers would beneit from reaching more potential applicants and from cutting costs by automating the preselection of applications. After the employment of a candidate, the annotated application could serve as basis data for the company human resource management system. Assuming continuous maintenance of the staff proiles as described in Sure et al. (2000), the described technologies could improve the automatic preselection of staff according to their competencies to be assigned to new projects. Job seekers would also beneit from the increased market transparency. Assuming standardized ontologies for representing an applicant proile (e.g., basic personal data, skills, etc.), a machine-processable job application could be created once and be reused many times. Service providers could provide enhanced services, since they could identify and classify semantically annotated job postings that are published on the Web servers of the companies. As Semantic Web technologies increase in visibility and uptake, we expect enterprise data to become more structured and more semantic and that systems based on these technologies such as the prototype presented here will provide new beneits to key sectors such as human resource management.
References
Andersen, T., Bulskov, H., & Knappe, R. (2003). From ontology over similarity to query evaluation. In Proceedings of the 2nd CoLogNET-ElsNET Symposium-Questions and Answers: Theoretical and Applied Perspectives, 39-50. Beckett, D. (2003). Scalability and storage: Survey of free software/Open Source RDF storage systems. Retrieved October 31, 2005, from http://www.w3.org/2001/sw/Europe/reports/rdf_scalable_storage _report Beckett, D., & McBride, B. (Eds.). (2004). RDF/XML Syntax Speciication (rev. ed.). Retrieved October 31, 2005, from http://www.w3.org/TR/rdf-syntax-grammar Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The semantic Web. Scientiic American, 284(5), 34-43. Bizer, C. (2003). D2R MAPA database to RDF mapping language. In Proceedings of the 12th International World Wide Web Conference. (pp. 20-24) Budapest, Hungary: ACM Press Bizer, C., Cyganiak, R., Gauss, T., & Maresch, O. (2005). The TriQL.P browser: Filtering information using context-, content- and rating-based trust policies. In Proceedings
of the Semantic Web and Policy Workshop at the 4th International Semantic Web Conference. (pp. 12-20) Bizer, C., Mochol, M., & Westphal, D. (2004). Recruitment. XML Clearinghouse Report, ISSN 1861 1222. Bizer, C., & Seaborne, A. (2004). D2RQTreating non-RDF databases as virtual RDF graphs. In Proceedings of the 3rd International Semantic Web Conference. (pp 7-11) Hiroshima, Japan. Crosswater Systems Ltd. (2003). Der Virtuelle Arbeitsmarkt der Bundesanstalt fr Arbeit. Anspruch und Wirklichkeitdas Millionengrab. Retrieved October 31, 2005, from http://www.crosswater-systems.com/dossier_03_vam_fakten_v100.pdf Dean, M., & Schreiber, G. (Eds.). (2004). OWL Web ontology language-reference. Retrieved October 31, 2005, from http://www.w3.org/TR/owl-ref/ Ferris, G. R., et al. (1999). Human resources management: some new directions, Journal of Management, 25(3), 385-415. Ganesan P., Garcia-Molina, H., & Widom, G. (2003). Exploiting hierarchical domain structure to compute similarity. ACM Transactions on Information Systems, 21(1), 63-94. Gene Ontology Consortium. (n.d.) The gene ontology. Retrieved October 31, 2005, from http://www.geneontology.org/ Gruber, R. T. (1995). Toward principles for the design of ontologies used for knowledge sharing. International Journal of Human-Computer Studies, 43(5-6), 907-928. IEEE Standard Upper Ontology Working Group. (n.d.) Suggested upper merged ontology (SUMO). Retrieved October 31, 2005, from http://www.ontologyportal.org/ Kalyanpur, A., Hendler, J., Parsia, B., & Golbeck, J. (2003). SMORE - Semantic markup, ontology, and RDF Editor. Retrieved October 31, 2005, from http://www.mindswap. org/papers/SMORE.pdf Keim, T., Knig, W., Weitsel, T., Wendt, O., & von Westarp, F.V. (2004). Recruiting Trends 2004 (Working Paper No. 2004-5). einance Institut. Johann-Wolfgang-Goethe-Universitt Frankfurt am Main. Keim, T., Knig, W., Weitsel, T., Wendt, O., & von Westarp, F.V. (2005). Recruiting Trends 2005 (Working Paper No. 2005-22). einance Institut. Johann-Wolfgang-Goethe-Universitt Frankfurt am Main. Knowledge Web. (n.d.) EU Network of Excellence. Retrieved October 31, 2005, from http://knowledgeweb.semanticweb.org/ Liebsch, F. (2005). PROMIPlatform for the reuse of ontologies through merging and integration. Retrieved October 31, 2005, from http://www.adestiny.de/promi.html Monster. (2003). Recruiting Trends 2004; Institut fr Wirtschaftsinformatik der Johann Wolfgang Goethe-Universitt Frankfurt am Main; 2. Fachsymposium fr Personalverantwortliche, 2003: MonsterDeutschland and TMP Worldwide. Moore, G., & Seaborne, A. (2003). RDF net API, W3C member submission. Retrieved October 31, 2005, from http://www.w3.org/Submission/2003/SUBM-rdf-netapi-20031002/
Mlder, W. (2000). Personalinformationssysteme: Entwicklungsstand, funktionalitt und trends. Wirtschaftsinformatik, 42, 98-106. Nixon, L., & Mochol. M. (2005). Prototypical business use cases. Deliverable 1.1.2 in the Knowledge Web EU Network of Excellence, 11-15, 53-56. Oldakowski, R., & Bizer, C. (2004). RAP: RDF API for PHP. Retrieved October 31, 2005, from http://www.wiwiss.fu-berlin.de/suhl/radek/pub/RAP-oldakowski.pdf Oldakowski, R., & Bizer, C. (2005). SemMFA framework for calculating semantic similarity of objects represented as RDF graphs. Poster session presented at the 4th International Semantic Web Conference. Paslaru, E. B., & Mochol, M. (2005a). Towards a reuse-oriented methodology for ontology engineering. In Proceedings of the 7th International Conference on Terminology and Knowledge Engineering. (pp. 175-187) Paslaru, E. B., & Mochol, M. (2005b). Towards a cost estimation model for ontology engineering. In Proceedings of the Berliner XML Tage 2005. (pp. 153-159) Poole, J., & Campbell, J. A. (1995). A novel algorithm for matching conceptual and related graphs. Conceptual Structures: Applications, Implementation and Theory, 954, 293307. Quan, D., & Karger D. (2004). How to make a semantic Web browser. In Proceedings of the 13th International World Wide Web Conference. (pp. 255-265) Schraefel, M. C., et al. (2004). CS AKTive Space: Representing Computer Science in the Semantic Web. In Proceedings of the 13th International World Wide Web Conference. (pp. 384-392) Sowa, J. F. (1984). Conceptual structures: Information processing in mind and machine. Addison-Wesley. Sure, Y., Maedche, A. & Staab, S. (2000). Leveraging corporate skill knowledgeFrom ProPer to OntoProPer. In Proceedings of the 3rd International Conference on Practical Aspects of Knowledge Management. Veit, D. (2003). Matchmaking in electronic markets: An agent-based approach towards matchmaking in electronic negotiations. Lecture Notes in Computer Science, 2882. World Wide Web Consortium. (2001). Semantic Web. Retrieved October 31, 2005, from http://www.w3.org/2001/sw/ Wright, P. M., & McMahan, G. C. (1992). Theoretical perspectives for strategic human resource management. Journal of Management, 18, 292-320. Zhong, J., Zhu, H., Li, J., & Yu, Y., (2002). Conceptual graph matching for semantic search. In Proceedings of the 2002 International Conference on Computational Science. (pp. 92-106)
Ontology-Based Competency Management for Corporate E-Learning 319
Chapter XV
Ontology-Based Competency Management for Corporate E-Learning

Fotis Draganidis, National Technical University of Athens, Greece Gregoris Mentzas, National Technical University of Athens, Greece
Abstract
The synergies between ontology management, competency management, and e-learning have been explored during recent years both on theoretical and practical levels. This chapter describes the architecture, design, and deployment of a system that integrates ontologies with competency management and e-learning, as well as with other human resources functions. Moreover, a detailed description of the supportive methodology and the main lessons learned in technical and organisational areas are presented. This system currently is being deployed for research purposes in a national subsidiary of Microsoft, the information technology (IT) services multinational irm. The objective of this chapter is to provide the reader with an overview of the key concepts in competency-based management, a non-technical methodology for supporting the effective deployment of an ontology-based competency management system and an analysis of the lessons learned during the irst deployment phase.
320 Draganidis & Mentzas
Introduction
Competency-based management has become a very crucial element in the effective operation of an enterprise or an organization, due to the increased need of the latter to be agile enough to adapt to quick market changes and re-orientation of its business plans. In this situation, competency management systems (CMS) become the core human resource tool, which enables the enterprise to manage and develop the skills of their employees, recruit the most appropriate candidates, and make effective succession planning and employee development plans. Apart from enterprise competency management systems, research is being conducted on the development of ontology-based CMS, which can provide possibilities such as the easy integration and mapping of different competency ontologies. Moreover, research efforts have been realized in the development of ontological e-learning systems. However, very few if any systems exist that integrate e-learning functionality with an ontological CMS. The mapping of employee or departmental/organizational skill gap analysis with the appropriate learning objects is crucial in order to develop the correct learning paths and consequently the appropriate competencies of employees or organizations. This chapter focuses on the description of an ontology-based competency management system, which also integrates e-learning functionality in order to address this issue. The interested reader can ind an essential introductive overview of the key concepts in competency-based management, as well as a methodology that supports the eficient deployment of such a system in an enterprise. The practical experience of the authors derives from the deployment of a Microsoft .NET version of the described system in Microsoft Hellas, the Greek subsidiary of the leading IT enterprise Microsoft Corporation. In the next section, we provide a brief history of competency management, a deinition of the term and a description of its core elements. In the third section, we describe the research efforts conducted in ontological CMS and ontological e-learning systems. In the fourth section, we describe the system design and architecture, while in the next two sections we provide the supportive methodology and the lessons learned during the irst phase of the deployment. Finally, conclusions and possible topics for further research are presented.
Competency-Based Management Key Concepts

The competency approach to human resources management is not new. The early Romans practiced a form of competency proiling in attempts to detail the attributes of a good Roman soldier. The introduction of competency-based approaches was realized around 1970, and their development since then has been rapid. The distinguished Harvard psychologist David McClelland is credited with introducing the idea of competency into the human resource literature, in his efforts to assist the United States Information Agency in improving its selection procedures. The latter argued that traditional intelligence tests, as well as proxies such as scholastic grades, failed to predict job performance. McClellands counter-argument to the growing dissatisfaction with intelligence testing and the traditional job analytic approaches to personnel selection was the proposal to test for competency. As
a case study, he proposed the selection of foreign service information oficers (McClelland, 1973). In his research, McClelland found that competencies such as interpersonal sensitivity, cross-cultural positive regards, and management skills differentiated superior from average information oficers (Dubois, 1993). Throughout the years competency-based approaches have proven a critical tool in many organizational functions such as workforce and succession planning, performance appraisal, etc. The main reasons for selecting these approaches are the following: They can provide identiication of the skills, knowledge, behaviours, and capabilities needed to meet current and future personnel selection needs, in alignment with the differentiations in strategies and organizational priorities. They can focus the individual and group development plans to eliminate the gap between the competencies requested by a project, job role, or enterprise strategy and those available.
According to the HR-XML Consortium competencies schema (http://ns.hr-xml.org/2_0/HRXML-2_0/CPO/Competencies.pdf), a competency can be deined as: A speciic, identiiable, deinable, and measurable knowledge, skill, ability and/or other deployment-related characteristic (e.g. attitude, behaviour, physical ability) which a human resource may possess and which is necessary for, or material to, the performance of an activity within a speciic business context. Based on the analysis of the existing deinitions and the further study and research that we have conducted on competency management, we provide the following deinition of the term competency: A competency is a combination of tacit and explicit knowledge, behaviour, and skills, that give somebody the potential for effectiveness in task performance. The authors regard the above term as the most appropriate deinition for the scope of this chapter and the analysis of the described competency management system. A further analysis of the concept of competency brought us to the conclusion that, typically, a competency is deined in terms of: 1. 2. 3. 4. Category: A group that homogeneous/similar competencies belong to Competency: A descriptive name for the speciic competency Deinition: Statement(s) that explains the basic concept of this competency Demonstrated behaviour: Behaviour indicators an individual should demonstrate if the speciied competency is possessed
Table 1. An example of a competencys deinition

Category People management competencies Competency Building team spirit Deinition Provide team members with the excitement and desire to cooperate with each other, contributing to common goals Demonstrated Behaviour 1. Encourages help and respect to other team members Creates a common mission and a feeling of belonging to a team that aims at such Provide mentoring and experience transfer Provide feedback on strengths and weaknesses of the team members
2.
Developing people
Help team members to reach their potential in personal development
1.
2.
Table 1 depicts an example of a competencys deinition in terms of category, competency, deinition, and demonstrated behaviour. The general category of the competency is the people management competencies, which amongst other can include the competencies of building a teams spirit and developing people.
Related Work
The use of ontologies in a competency-driven e-learning system is a research area being explored over recent years. Some ontology-based, competency-based tools or prototypes have been introduced, such as CommOn (Trichet & Leclere, 2003), a framework for building competency-based systems. CommOn is based on two models (implemented with speciic tools) which guide irstly the building of competency reference systems related to particular domains such as healthcare or information and Telecommunication, secondly the identiication and the formal representation of competency proiles, and thirdly the matching of competency proiles. The CommOn framework allows one to build shareable ontologies and knowledge bases represented with semantic Web languages and to develop competency-based Web services dedicated to human resource management. Also, other systems include SMS, or skills matching system (Colucci, Di Noia, Di Sciascio, Mongiello, Donini, & Mottola, 2003), which is a prototype that is not linked with e-learning systems; and GMS (Vasconcelos, Kimble, & Rocha, 2003), which is an ontology-based competency management system for managing group competencies and which is not integrated with e-learning functionality or system. Moreover, an architectural proposal of a prototype system for ontology-based competency management is presented by Reich, Ontology-based competency management in Swiss life (Reich, Brockhausen, Lau, & Reimer, 2002), as a further step to extensible markup language (XML)-based competency management systems, such as MaSel (Garro & Palopoli, 2003). However, these prototypes do not integrate elearning functions or links with learning objects and resources.
Research work is also conducted in the usage of ontologies in learning objects in order to facilitate the discovery and reuse of learning objects stored in local and global repositories (e.g., Sicilia, Garcia, Diaz, & Aedo, 2002; Sicilia & Garcia, 2003). A learning object is deined as any entity, digital or non-digital, that may be used for learning, education or training (IEEE Learning Technology Standards Committee, 2001). Initiatives such as the IEEE learning object metadata (LOM), Dublin Core and IMS Global Consortium, are developing standards, speciications and reference models for learning objects in order to facilitate the online retrieval and reusability of the latter. In the future, developments of the discussed system lies in the automatic discovery of learning objects from global repositories and their mapping with competency gap reports in order to facilitate the user to access both local and global repositories of learning objects. Moreover, the use of ontologies in e-learning applications has been theoretically researched, for example, with the architectural proposal of a prototype system for e-learning using ontologies (Schmidt & Winterhalter, 2004; Stojanovic, Steffen, & Studer, 2001). Some research efforts in integrating competency ontologies with e-learning have taken place in theoretical background (Woelk, 2002) and have been partially implemented (Hirata, Ikeda, & Mizoguchi, 2001). Additionally, many learning management systems integrate competency management features without ontological support. For a detailed analysis of the main competency management features included in some popular learning management systems (LMS), one can refer to Draganidis and Mentzas (in press). Evidently, there is a gap in the integration of ontology-based, competency-driven e-learning systems. Our systems target is to develop and deploy an ontology-based competency management system, which will provide the possibility for further enhancements, such as succession planning and training analysis, always using the developed ontology as a reference. Moreover, through the use of Web services the system can be integrated with other human resource management or e-learning systems.
System Architecture and Design

Our research efforts focus on developing a prototype ontology-based system, which will integrate competency management with e-learning and other human resource functions, such as succession and career planning, training needs analysis and organizational planning; Figure 1 outlines the conceptual design of our system. The system is based on ontological structures: the competency ontology and the learning objects ontology. The competency ontology describes the competency model of the organization, as this is deined by the human resources director. Each competency is composed of four proiciency levels, which correspond to one or more indicating behaviors. Table 2 analyzes the four behaviors that can be observed regarding the competency team leadership. Of course, the deinition of these behaviors is different according to the needs of the organization to which the competency model refers to, and the following descriptions are used only as random examples.
Figure 1. Conceptual design of the system
Competency Ontology Skill Gap Anlaysis The desired competencies in another position Succession Planning
Employee Competency Proile
The competencies that an employee possess and their levels
The desired competencies in employees current position
Employee Development
ts an w ob s, ee e j ok oy th o T) pl n , b B m ., o ion , C e e .g at ng th n (e ul ini m ow ar si tra H le , g, to ning nin i r tra -lea e
Personalised Learning Model
Intranet Documents E-learning Courses Seminars Personalised Learning Path Learner
Learning Objects Ontology
Table 2. Analysis of the competency team leadership

Competency name: Team leadership Competency deinition: Persons who demonstrate team leadership regard themselves as people leaders. They can share a vision with the team, create excitement, and be a facilitator in the process of jointly reaching this vision. Proiciency level 1 Indicating behaviors: Provides all information required to facilitate the accomplishment of team members work tasks. Proiciency level 2 Creates team spirit. Sets processes to facilitate the team success. Proiciency level 3 Provides development opportunities for the team members. Proiciency level 4 Shares his/her vision with the team and creates excitement and commitment to the team for its accomplishment.
Figure 2. Learning objects system view
Figure 3. Relationships between competencies and learning objects
Each competency level links to one or more learning objects, and the sum of learning objects is described in the learning object ontology. In Figures 2 and 3, we present the creation of learning objects in the system and the way they are linked with competencies. Each competency can have one or more learning objects, which are categorized in different categories, such as books, seminars, online documents, and so forth. Each learning object corresponds to one or more competencies in a speciic proiciency level. Figure 4 depicts the class hierarchy in the competency ontology designed with the ontology editor Protg. The system supports two main user roles: The human resources (HR) user and the learner (or employee) user. HR has permission to construct the competency model of the organization and enter the competency assessment for each employee and the required competencies for each job position. At a second step, HR can perform a skill gap analysis for an individual, group, or organizational level in order to identify missing competencies and be able to develop the training and succession plan. The employee can review his/her missing competencies, deine the learning procedure of his/her choice (e.g., mentoring, on-the-job training, books, CBT, classroom training, etc.), and review his/her personalized learning plan, customized to his/her learning preferences.
Figure 4. Competency ontology description in Protg
Figure 5. Screenshot of the system presenting an employee skills gap report
Figure 6. Conceptual view of the pilot system
Human Resource Systems
Other Learning Systems
The process of analyzing the competency gap between the competencies possessed by an employee and those required for effectiveness in a job role is the following: The HR department analyses which competencies and which proiciency level the employee possesses. Given that the competencies and the competency levels of the job role that this employee has are already entered, the system is able to compare one by one the corresponding competency and competency levels and to create a relevant report. It also provides the employee with the learning objects required in order to further develop the competencies that are needed in his/her current role. Figure 5 depicts the results of an employee skill gap analysis, with random competencies and competency levels. A Microsoft .NET implementation of this system is currently in its pilot deployment phase as a case study in Microsoft Hellas, the Greek subsidiary of Microsoft Corporation. In Figure 6, we provide a technical conceptual view of the system. The system that is deployed in Microsoft Hellas stores all information about employees and jobs competencies and learning objects in a Microsoft SQL 2005 Server. With the help of an XML parser, this information is linked with two ontologies, the competency ontology and the learning objects ontology. The user interface enables the userhuman resource director or employeeto access information through the Web as well. Finally, the system has been designed in such a way that it can integrate through Web services with other learning systems or human resource systems.
Support Methodology
During the design and deployment of the abovementioned competency management system with the human resource management team of Microsoft Hellas, a supportive methodology has been created. This includes the following steps in the process:
A. Creation of a competency management team The irst step to a successful implementation of a competency management system is the creation of a team that will support it throughout its lifecycle. In our case, this team consists of the human resources director and other HR staff, the technical team that is in charge of developing the software solution, competency and knowledge management researchers, and employees from different departments who have deep knowledge on the jobs included in the model, and the whole project is endorsed by high executives of the enterprise. Consequently, around 10 members belong to the team, providing input and feedback, while ive of them form the core team of the project, which has the responsibility for overseeing the whole initiative. B. Creation of the competency model The second step of the presented methodology is the creation of the competency model. This includes the following steps: Formed competency management team should work together in order to develop a tentative competencies list, which will be the basis for the construction of the model: In order to develop this team, the team could review competency models used by other organizations and align them with their companys short- and long-term strategies and business plan. In our case, we have reviewed the competency models provided in the past from Microsoft and other companies and combined the review results with the three-year business plan of the enterprise, which inally resulted to a competency model consisting of 28 competencies. A proiciency scale should be prepared for each competency: A scale of four competency levels has been created in our case, in order to deine expert, superior, average, and marginal possession of the competencies included in the tentative competencies list. Deinition of competencies behavioural indicators: Each competency should be analyzed in behaviours in order to facilitate the employees competencies assessment at a next step. In order to deine the indicating behaviours of each one of the four competency levels, it is essential to collect feedback from employees who master these competencies, through interviews, focus groups, and surveys of employees and managers. Development of an initial competency model: The competency management team, based on quantitative analysis of the feedback and results that have been collected in the previous steps, should develop an initial competency model. Fine-tune the initial competency model: In this step, additional feedback can be collected from employees and high executives who have not participated in the development of the initial competency model. The team can perform a quantitative and qualitative analysis of the feedback in order to perform any necessary adjustments to the model. Validation of the model: Validation efforts begin with converting the competencies into a questionnaire that can be used for rating individual effectiveness. The individu-
als identiied earlier as superior, effective, and marginal performers are rated on this questionnaire by multiple individuals, if possible, such as managers, peers, and direct reports. The ratings on the competency questionnaire are correlated to the performance ratings to determine if each competency relates to job performance. Finalize the model: The last step involves eliminating any competencies that do not correlate with the performance measures to provide a validated model that is linked to effective performance.
C. Employees Competency Assessment The assessment of the competencies that each employee possesses, as well as of the level that each competency is possessed, is crucial to the success of the whole project. Only through correct evaluation will the system be able to perform an accurate skill gap analysis and all the other human resources functions that are based on it, such as creating a personalized learning path for employees development, deining training needs, performing succession planning, and more. What we suggest as the best approach to a successful competency assessment of employees is a combination of 360-degree feedback and performance review. In more detail, a questionnaire about an employees behaviors should be replied to by the employee herself, her peers, her manager, her team members, and possibly from people outside the company. Then, the results collected should be weighted based on different factorsfor example, the number of years that each of the responders has collaborated with the assessed employee. This initial assessment should be validated and ine-tuned, using the results of the employees performance review. For example, if an employee has received a good score on a performance review metric because several members of her team have been promoted, it is highly likely that she possesses the competency of developing people at a high level, but this has to be crosschecked and validated with the results of the 360-degree feedback. D. Correlation of the Competency Ontology with the Learning Object Ontology The competency management system that has been described above, after performing the skill gap analysis of an employee, provides her with a personalized learning path in order to increase the level of possessed competencies or acquire additional ones. The successful implementation of this function depends on the correct correlation of the competency ontology with the learning object ontology. Ideally, each competency level should be related to one or more learning objects that are appropriate only to the speciic competency level, thus facilitating the learning process and not providing the employees with learning objects that are too hard or to easy for her competency levels. To eliminate this possibility, feedback should be collected by employees who master certain competencies, on the competency level that each learning object should be assigned. This subjective correlation of competency levels with learning objects has proven the only eficient way to validate this process, even for learning objects with predeined competency levelfor example, a technical seminar predeined as level 200/500.
Lessons Learned
The lessons learned from the design, development, and deployment of the competency management system include both technical and organizational issues that proved important for the successful deployment of the system in a highly demanding multinational environment, such as in Microsoft Hellas. The main lessons learned in the technical ield include the following: Design the system taking in mind the needs of human resources staff and employees: The system will be primarily used by two groups of users, the human resources staff who will have administrative privileges and the employees who want and should have a personalized view. We have performed in-depth user requirements analysis and tried to keep the user interface for both groups as simple as possible in order to facilitate the wide usage of the system. One of the most important requests that we have received from human resources staff is the ability to have reports on as many features as possible, such as top employees per competency, ive most possessed competencies, competencies possessed by department or project team, and so forth. Consequently, the reporting mechanism of the system should be taken under serious consideration. Involve the IT department in the system design: Another lesson learned is that the IT department should be engaged from the very irst steps of the design and implementation of the competency management system, especially when this is going to collaborate with other IT systems. Use of visualization components: During our meetings with the human resources department of Microsoft Hellas, it became evident that the system should have excellent reporting tools. Moreover, to facilitate the adoption of the system by the two groups of users, the human resources staff and the employees, the system needs add-on modules visualizing competencies and competency interests. Design for integration: The competency management system is highly likely to have to collaborate in the future with other systems in the enterprise, such as a performance review system, an e-learning application, and so forth. To ensure that this will be possible in the future, it is highly recommended to use an open architecture of Web services and XML in system design.
Additionally, some of the important lessons learned as far as organizational issues are concerned include: Ensure high executives endorsement: The support of the directors team is very important in order to ensure the wide usage of the competency management team in an enterprise. Employees, due to time restraints, are usually reluctant to start using new applications, unless there is a clear beneit to their job role or personal development and are motivated by their managers.
Ensure fair employee competency assessment: The correct assessment of the competencies and competency levels of the employees is crucial to the adoption, usage and success of the system. Unless the employees feel that their competencies have been fairly assessed, they will not accept to proceed with the other functions of the system, such as completing the personalized learning path or succession planning. As described above, a combination of 360-degree feedback and performance review may prove important to the fulillment of this goal.
Conclusion and Further Research

In this chapter, we have provided a practical approach to the integration of competency management, e-learning, and ontologies, presenting a system currently under deployment in a multinational IT irm. The system design and architecture have been analyzed, as well as the main lessons learned during the design and deployment phase. These include both technical issues, such as designing for integration, making use of visualization components, and involving the IT department, as well as organizational issues, such as ensuring top executives endorsement and ensuring a fair assessment of employees skills. Moreover, we have provided our own approach to the deinition of the term competency and a further analysis of its main concepts. Further research lies in the area of expanding the system with semantic attributes, such as adding semantic annotation to the Web services exported, in order to enable it with broader integration capabilities with other ontology-based human resources systems. Moreover, by adding semantic Web services capabilities to the system, we will enable semantic Web matchmakers to accept the descriptions of our available services and match them against requirements from different requestors, in an automatic way. This process can currently be deployed manually, through registering the produced Web services to universal description, discovery, and integration directory (UDDI). Additionally, all the data that will have been gathered upon the completion of its pilot phase deployment in Microsoft Hellas will be carefully examined in order to reveal trends in the relationships of employees and competencies.
References
Colucci, Di Noia, Di Sciascio, Mongiello, Donini, & Mottola. (2003, July 2-4). Finding skills through ranked semantic match of descriptions. In Proceedings of I-KNOW 03: 3rd International Conference on Knowledge Management, Skills Management, Graz, Austria. Draganidis, F., & Mentzas, G. (in press). Competency management: A review of systems and approaches. International Journal of Information Management and Computer Security.
Garro, A., & Palopoli, L. (2003). An XML multi-agent system for e-learning and skill management. In R. Kowalczyk et al. (Eds.), Agent technology workshops 2002, LNAI 2592 (pp. 283294). Berlin Heidelberg: Springer-Verlag. Hirata K., Ikeda, M., & Mizoguchi, R. (2001, November 12-15). Total resolution for human resource development based on competency ontology. In Proceedings of the 9th International Conference on Computers in Education (ICCE2001), Seoul, South Korea. IEEE Learning Technology Standards Committee (LTSC). (2001). Draft standard for learning object metadata (version 6.1). Retrieved from http://ltsc.ieee.org. IMS Global Consortium. [www.imsglobal.org] McClelland, M. (2003). Metadata standards for educational resources. IEEE Computer, 36(11), 107-109. Reich, Brockhausen, Lau, & Reimer. (2002). Ontology-based skills management: Goals, opportunities and challenges. Journal of Universal Computer Science, 8(5), 506-515. Schmidt A., & Winterhalter, C. (2003). User context aware delivery of e-learning material: Approach and architecture [Special issue]. Journal of Universal Computer Science, 10(1), 28-36. Sicilia, M. S., & Garcia, E. (2003). On the integration of IEEE-LOM metadata instances and ontologies. Learning Technology, 5(1). Sicilia M. A., Garcia, E., Diaz, P., & Aedo, I. (2002, December 3-6). LearningLinks: Reusable assets with support for vagueness and ontology-based typing. In Proceedings of Workshop on Concepts and Ontologies in Web-based Educational Systems. ICCE. Stojanovic, L., Staab, S., & Studer, R. (2001, October 23-27). E-learning based on the semantic Web. In Proceedings of the World Conference on the WWW and the Internet (WebNet 2001), Orlando, FL. Trichet & Leclere. (2003). A framework for building competency-based systems. In N. Zhong et al. (Eds.): ISMIS 2003, LNAI 2871 (pp. 633-639). Berlin Heidelberg: Springer-Verlag. Vasconcelos de Braga, Kimble, C., & Rocha, A. (2003, July 2-4) Ontologies and the dynamics of organisational environments: An example of a group memory system for the management of group competencies. In Proceedings of I-KNOW 03: 3rd International Conference on Knowledge Management, Skills Management, Graz, Austria. Woelk, D. (2002, June). E-learning: Semantic Web services and competency ontologies. In Proceedings of ED-Media 2002 Conference.
Applying Semantic Web in Competence Management 333
Chapter XVI
Applying Semantic Web in Competence Management

Mikko Laukkanen, TeliaSonera, Finland Heikki Helin, TeliaSonera, Finland
Abstract
Eficient competence management is essential in knowledge-based companies. This chapter describes how the Semantic Web technologies can be used in managing employee competencies. Applying the Semantic Web technologies in competence management enables building systems that support highly dynamic environments, are extensible as well as interoperable between different application domains, and beneit from the use of machine-accessible semantics. Competence management systems should be available not only for managers but for all the employees of the company. As companies get larger, it becomes increasingly dificult to manage the knowledge and competencies that their employees have. Utilizing the Semantic Web opens many possibilities for building lexible systems for competence management.
334 Laukkanen & Helin
Introduction
For companies with intellectual property, it is crucial to have an environment where the knowledge can be captured and shared eficiently within the company. Competence management is becoming increasingly important in todays competitive markets. Firstly, companies are constantly re-structuring their organization to better meet the challenges of the markets, which may result in employees with critical competencies being moved away from the companys core competence areas. Secondly, when downsizing the current personnel, it is crucial not to lose core competencies from the company. Similarly, when hiring new employees, it is also important to select the best candidates in terms of the core competencies of the company. Thirdly, new products and technologies are constantly entering the markets. This requires new skills and competencies from the employees in the company. Fourthly, successful project work requires that the project group be created from the best available candidates based on the competencies needed in the project. A properly handled competence management builds a solid base for deining the business strategies for companies; the core competencies should be focused on the core business areas. Companies must decide how to arrange resources and employees to form core competencies, which then can be used to satisfy customer needs by implementing business strategies. The business strategy deines the position of a company in the industry and the relation to its competitors. A well-known model for helping to deine business strategies is Porters Five Forces Model, which outlines the primary forces that determine competitiveness within an industry: rivalry, new entrants, suppliers power, substitute products, and buyers power (Porter, 1998). In order to develop effective business strategies, managers must decide how to react to these external forces. A competence management system enables a company to place the most competent employees in the core competence areas, and, thus, have the best possible resources to meet the external forces. Traditionally, competence management systems have been aimed at managers in the company (OLeary, 1998). That is, the competencies are collected in one way or another from the employees by a human resource department, which uses a competence management system for reining and providing the information to the executives (Lindgren, Stenmark, & Ljungberg, 2003). However, an emerging trend is that the competence management systems are also designed for the entire company. In doing so, the employees of the company are able to publish and share their competencies not only for the managers but also between other employees. One way of sharing knowledgeand maybe the most common oneis to irst establish a network of contacts; that is, an employee knows what kind of competencies his/her coworkers have. After that, the knowledge can be shared by asking the person with a given competence directly. As companies get larger, it becomes increasingly dificult to manage the knowledge and competencies that their employees have. The knowledge sharing within small companies usually happens in a face-to-face fashion between the employees; everyone knows each other and the competencies of their co-workers. However, in large companies, which usually are geographically distributed between different countries and cultures, the contact network of an employee usually covers only a small fraction of the whole company. In such companies, eficient knowledge sharing is extremely challenging.
This chapter describes how the Semantic Web technologies (Berners-Lee, Hendler, & Lassila, 2001) can be used to help solve the aforementioned issues. We also introduce an innovative Semantic Web-based solution for managing employee competencies and other relevant resources, such as documents, customers, and projects. Using machine-accessible semantics is the main difference between the Semantic Web-based solutions and other seemingly similar competence management solutions. In traditional competence management systemswhether implemented by using rather simple information technology (IT) systems or (in) formal questionnairesemployees can only state their exact competencies. Because of the lack of explicit semantics, it is typically hard to infer competencies that the employees may have without explicitly knowing about them by themselves. In our Semantic Web-based approach not only can those employees who have competence directly on a competence topic be found but also employees that have competencies on some closely related competence topics. This means that we can look up persons who are the best possible candidates for a competence topic or a set of competence topics. TheSemantic Web and ontologies also enable lexible portability between different domains. For instance, in the area of competence management, a marketing department may require a totally different ontology for its competencies than a research and development department. Still, the application logic is identical for both departments. Utilizing the Semantic Web and ontologies opens many possibilities for building lexible systems for competence management. The knowledge base in our solution is modeled as ontologies; thus, it demonstrates how the Semantic Web technologies can be used in a practical application on the area of competence management. The rest of this chapter is organized as follows. First, we provide background information about competence management, the Semantic Web, and ontologies. Then we proceed to discuss how competence management systems beneit from using Semantic Web technologies. After this, we introduce a Semantic Web-based system for managing and sharing competencies; we describe the ontologies used in the system and explain the algorithms for search operations, and show how the system is used from an end users perspective. At the end, we discuss some open issues and future trends, and, inally, conclude this chapter.
Background
Knowledge and Competence Management within Organizations
Competence management is gaining importance in knowledge-based organizations for several reasons. In the following, we point out the most relevant issues in terms of this chapter. Firstly, the reductions in stafing create a need to replace informal knowledge with formal methods. This means that it is important to formalize the knowledge of the employees in order for the competence management systems to capture the knowledge and make it reusable by the other employees in the company. Secondly, competitive pressures reduce the size of the work force that holds valuable business knowledge. The downsizing of the personnel
in companies is a common trend today, which may result in key knowledge holders leaving the company and taking the knowledge with them. Thirdly, the amount of time available to experience and acquire knowledge has diminished. In order for a knowledge-based company to stay on the bleeding edge of the technology, the most recent and relevant knowledge has to be available and applicable as soon as possible. Fourthly, the increase in mobility of the work force leads to a loss or disconnection of knowledge. As the mobility of the employees increases, the knowledge sharing in traditional ways (e.g., face-to-face discussions) becomes extremely challenging. Thus, more eficient and modern knowledge and competence management systems are needed. Fifthly, changes in strategic direction of a company may result in the loss of knowledge in a speciic area. This applies especially for companies that follow the market trends and constantly adapt their strategy to meet the market needs. These issues can be extended by the following issues raised by Barclay and Murray (1997): Most of our work is knowledge based. This has been an obvious fact for IT companies, but the need for knowledge and competence management is becoming increasingly important also in non-IT companies, such as the paper industry, where the nature of work has changed from manual and physical work into controlling automated high-tech paper machines. Organizations compete on the basis of knowledge. One example of this is when the companies buy the key persons from their competitors. Products and services are increasingly complex, endowing them with a signiicant information component.
This places demands on both the product/service providers and consumers. The need for life-long learning is an inescapable reality.
When speaking about competence management, many terms are deined in various ways. In OLeary (1998) it is discussed how a companys knowledge can be captured and made available to all its members. In this context, OLeary says that enterprise knowledge management entails formally managing knowledge resources in order to facilitate access and reuse of knowledge, typically by using advanced information technology (p. 54). From this deinition, we could pick up a few issues relevant to this chapter. Firstly, knowledge (or competence) management should be formalized and made explicit so that it can be effectively analyzed and utilized. Secondly, the knowledge resources do not necessarily mean only employees (immaterial) competencies or skills. They should also cover resources that the employees arebased on their competencies and skillslinked to in one way or another. These kinds of resources are, for instance, documents they have written and projects they have participated in, but also other employees they know. Thirdly, the use of advanced information technology, as is proposed later in this chapter, is a necessity in order to build intelligent and lexible competence management systems.
One important aspect is obviously the deinition of the competence itself, which also has many deinitions and terms associated and interpreted in the literature. Harzallah and Vernadat (2002) consider a theoretical model of three components of competency, which consist of knowledge, know-how, and know-whom. The knowledge is everything that requires training or learning. The know-how, on the other hand, deals with competencies acquired by performing tasks, such as working experiences. The skills can be seen as a synonym for know-how. The know-whom is about individual characteristics, which are comprised of abilities such as creativity, communication capabilities, and interests. As compared to the theoretical deinitions of competency, a more pragmatic way of deining competency in the form of a data schema has also been introduced (Sicilia, 2005). The HRXML Consortium (Allen, 2006) proposes a common structure for competency, which aims at bringing interoperability between systems exchanging competency-related information. The competencies are deined using XML fragments, which include information such as competence name and description, importance, and possible reference to taxonomy. Although a machine is able to parse these, the XML fragments lack the model for semantics, which is the main difference to our solution. Languages like XML deine the structure of a document but lack a semantic modelintuitively an XML document may be clear, but computers lack the intuition. When operating on a syntactic level, a computer is not able to know if the terms Java and Java Language mean the same thing or not. Thus, while in HR-XML a competency name Java is of type string, we could state that it would refer to a concept of Java. That is, when we add semantics to the system, the terms Java and Java Language would relate to the same concept of Java, and a computer would be able to know that they both mean the same thing. Another advantage of adding semantics is the possibility of reasoning. In XML-based systems a computer has no way to reason whether the word Java refers to a programming language or an island in Indonesia. With semantics, we can deine that a concept Java would relate to a concept of Programming Language; thus, the computer in this case could reason that the Java actually is a kind of programming language. In this chapter, we deine the competency to be an ability to know something or perform some action about something. In our solution, this practically implies an abstract unidirectional link from a resource to a competence topic. This deinition can then be specialized to cover relations that are more concrete, such as a person can have knowledge on some competence topic. A competence topic represents either a physical object, such as a mobile phone, or immaterial thing, such as foreign language skills or abilities to lead large projects. The competence topics are structured as a semantically meaningful hierarchy. The deinition of a single competence topic follows the idea presented in Sicilia (2005), although it is not as broad as the deinition in HR-XML. However, unlike in HR-XML, our taxonomies are built into the competence topic hierarchy, and the other properties, such as importance, are related to the employees or other resources competence topic. The competence management systems in the past were like expert systems that included a process for discovering and capturing knowledge about various data sources, such as documents, Web pages, spreadsheets, presentations, and databases (Staab, Studer, Schnurr, & Sure, 2001). After being captured, the knowledge is then made available, accessible, and usable by the human resources department and managers. However, the current trend in competence management is that the competence management systems should be open for all the members of organizations. As the authors Sunassee and Sewry (2002, p. 241) argue that: the contribution of the employees in a knowledge management effort is invaluable to
any organization. Therefore, in addition to the human resource departments and managers, the employees of companies should have access to the competence management systems in order to manage their own competence descriptions and make queries about the competencies of their fellow employees. When competence management systems are made available for not only human resource experts but for all the members of a company, the system has to be as easy to use and maintain as possible. It must provide as much intelligence in terms of helping the user when maintaining his/her competencies and provide semantically sound results when making queries. This is where IT systems would play an important role, but they would require that the information in the competence management system be in a machine-accessible and understandable format. The Semantic Web technologies provide tools for implementing such a competence management system.
The Semantic Web

In our solutions, we apply Semantic Web technologies for competence management. Currently, the World Wide Web (WWW) is based primarily on documents (Web pages) written in HTML, which structures the documents using a set of tags. For instance, a text block between tags <b> and </b> means that the text is in boldface. The visualization and interpretation of these tags are meant for human users perceiving the HTML document. The HTML documents may also contain links to other HTML documents, but only a human knows what these links mean. Thus, the WWW is meant for humans, and machines are not able to understand about the information on the WWW. The same applies to languages like XML, which only deine the structure of a document and not the semantic model of the document. The Semantic Web deines the meaning for the information so that not only the humans but also the machines are able to process the information. The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries (World Wide Web Consortium, 2001). Therefore, the Semantic Web can be seen as a mesh of information linked up in such a way as to be easily processed by machines on a global scale.
Figure 1. Layers of the Semantic Web
Figure 2. Example of using RDF and RDF schema
The Semantic Web principles are implemented in the layers of Web technologies and standards. The layers are presented in Figure 1. At the lowest layer, the Unicode asserts that the information is encoded in international character sets, and the URI provides means for identifying resources in a unique manner in the Semantic Web. The syntax of URIs is governed by the IETF (Berners-Lee, Fielding, & Masinter, 1998). The XML layer with namespace and schema deinitions integrates the Semantic Web deinitions with the other XML-based standards. The XML is maybe the most popular and widely used serialization for the information in the Semantic Web (Becket, 2004), although it is not the only syntax; see, for instance, Notation3 (Berners-Lee, 2001). Whereas the two lowest layers are about identiiers and syntax, the layers above those are the ones that enable adding the meaning (semantics) to the information. The resource description framework (RDF) (McBride, 2004) is a data model that forms the third layer in Figure 1. The RDF consists of triples, which are groups of three URIs. In a triple, a resource (the subject) is linked to another resource (the object) through an arc labeled with a third resource (the predicate). Therefore, triples allow deining statements like Model XYZ is of type Laptop (see Figure 2). The RDF Schema (Brickley & Guha, 2004) is a document that controls a set of terms in an RDF document. Thus, the RDF schema is a data-typing model for RDF. The RDF schema allows deining taxonomies like Laptop and Desktop as subclasses of a Computer (see Figure 2). The ontology layer supports the evolution of RDF schema vocabularies, as it can deine relations between the different concepts. Because the ontologies are the most important building blocks of the Semantic Web, and also the most important layer from this chapters viewpoint, we will have a more detailed look at the ontologies in the next section. The top three layers, logic, proof, and trust, are currently being researched. The logic layer enables the writing of rules while the proof layer executes the rules and evaluates, together with the trust layer mechanism for applications, whether to trust the given proof or not.
Ontologies
Ontologies are intended for knowledge sharing and knowledge reuse. To enter a discussion, communicating parties need to share a common terminology and meaning of the terms used (Gruber, 1993). Otherwise, proitable communication is infeasible because of a lack of shared understanding. With software systems, this is especially truetwo systems cannot interact
with each other without common understanding of terms used in the communication. Until now, this common understanding is achieved by hard coding this information to applications, making the systems static and impractical. This is where ontologies come into the picture. Ontologies describe the concepts and their relationshipswith different levels of formalityin a domain of discourse (Gruber, 1993). An ontology is more than a taxonomy (classiication of terms) since it includes richer relationships between deined terms. For some applications, taxonomy can be enough, but without rich relationships between terms, it is impossible to express domain-speciic knowledge without deining new terms. Ontologies have been an active research area for a long time. One of the most problematic issues in developing ontologies is the actual conceptualization of the domaincompetencies and other related resources, in our case. In recent years, ontology languages based on Web technologies have been introduced. DAML+OIL (Hendler & McGuinness, 2000), which is based on RDF schema (Brickley & Guha, 2004), is one such language. It provides a basic infrastructure that allows machines to make simple reasoning. Recently, DAML+OIL language was adopted by W3C, which is taking the work further by developing it to a Web ontology language (OWL) (McGuinness & Harmelen, 2004). Like DAML+OIL, OWL is based on RDF schema, but both of these languages provide additional vocabularyfor example, relations between classes, cardinality, equality, richer typing of properties, characteristics of properties, and enumerated classesalong with a formal semantics to facilitate greater machine readability. OWL has quite strong industry support, and therefore it is expected to become a dominant ontology language for Semantic Web.
Semantic Web-Based Competence Management

In Berners-Lee et al. (2001, p. 35), the authors state that the Semantic Web is an extension of the current web in which information is given well-deined meaning, better enabling computers and people to work in cooperation. This deinition is well aligned with the current trend in competence management, where competence management systems are no longer aimed at managers and executives but at all the employees in organizations (OLeary, 1998). When applying the Semantic Web technologies in competence management, we can build systems that beneit from the use of machine-accessible semantics, support highly dynamic environments, and are extensible as well as interoperable between different application domains. These issues are discussed in the following sections.
Reasoning over Competence Ontologies

The ontologies are described using a language with formally deined semantics. For instance, OWL allows deining classes and their instances as well as properties for those. In addition, there are many other inbuilt relations that can be deined between classes, such as taxonomies (subClassOf and subPropertyOf) and restrictions (unionOf, intersectionOf, etc.). In addition to these inbuilt properties of OWL, one can of course deine the domain speciic properties for the application. Using these semantics, reasoning about ontologies and their instances
Figure 3. Example ontology for programming languages
can be made. For instance, let us consider the ontology presented in Figure 3. First, we can see that Java and C++ are object-oriented languages, and, more generally, Programming Languages. Similarly, Pascal is a procedural programming language but also a programming language, just like the Java and C++. The prog:supports property is a domain speciic property from the programming domain, which is used between two pairs of resources; the object-oriented languages supports inheritance, and programming languages support a conditional IF-THEN clause. Now, because Java and C++ are subclasses of object-oriented languages, they both also support inheritance. This can be reasoned because the properties in OWL are inherited by the subclasses. In addition, because Java and C++ are also Programming Languages, they both support conditional IF-THEN clause. Pascal on the other hand only inherits the properties from the Programming Language, thus, we can reason that Pascal also supports conditional IF-THEN clause, but not inheritance. When applying the reasoning capabilities to the competence management systems, we can extend the searches to cover situations where direct matches are not found. In this case, the people with competence on nearby topic(s) can be searched for. This is crucial today, when, for instance, new products hit the markets rapidly. Let us consider the following example, which is depicted in Figure 4. A person working for a software vendor has competence on a mobile phone model . Then, the mobile phone manufacturer releases a new mobile phone , which is based on the model but on which nobody has yet competence at the software vendor. However, at some point when competencies at the software vendor for the new mobile phone are needed, a competence management system with Semantic Web-based reasoning is able to ind the persons having competence on the phone model . Although this is not an exact match, it is a very close one, because usually the given phones are similar in functionality. Furthermore, the person having competence on phone model can be considered a candidate when searching for persons having competence on Series 80 phone models, although the competence level is lower than for the Series 60 phone models. Therefore, adding or changing the competence topic structure does not necessarily require corresponding updates to the competencies of the persons.
Figure 4. Sample ontology for mobile phones
The reasoning can be extended even further by taking into account also other relations than the ones between an employee and his/her competencies. The term social network was coined by Barnes (1954), and it means a social structure between actors, in our case, employees within companies. The social networks also have been researched by the Semantic Web community, from which a FOAF (Friend of a Friend) speciication (Brickley & Miller, 2005) has emerged. By incorporating FOAF descriptions of the employees into the competence management system and then examining the social networks of the employees and the other resources that the employees are related to enables more intelligent searches. For instance, consider a case where an employee is searching for competent Java programmers to help with his/her Java-related problem. Let us then assume that the competence management system inds a person who is not available at the time. Therefore, the employee requests the competence management system for the social network (colleagues) of that person. If these new persons have a competence on Java and they are available, the employee can contact them. Similarly, if the examining of the social network does not provide any results, the other resource of the person can be requested. For instance, the projects and documents, which the person is related to, can be looked up, and further search for the persons related to those resources is possible. Should any of these persons be competent on Java and available at the time, the employee can contact them.
Handling Dynamic Environments with Ontologies

Because the competence structure and the whole data model in the competence management system will most likely change dynamically over time, the competence management system needs to have a clean separation between the data and the application logic. While the traditional systems typically allow attaching and modifying resources to the existing data model, they usually do not allow changing the data model itself. Ontologies make it possible not only to add new concepts, but also new relationships between the concepts, without needing modiications to the application logic. The Semantic Web and ontologies also enable lexible portability between different domains. For instance, in the area of comCopyright 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.
petence management, a marketing department may require a totally different ontology for its competencies than a research and development department. Still, the application logic is identical for both departments.
Competence Manager: A Semantic Web-Based Tool for Managing Competencies

In this section, we introduce a system called competence manager, which is a tool for managing and sharing competencies. The knowledge base in the competence manager is implemented as ontologies; thus, this section describes how the theory presented in the previous sections can be implemented in practice. The main tasks for the use of the competence manager are as follows. Firstly, the competence manager can be used to track down persons with a given competence within a company. This functionality could be used when a person needs assistance or help for some topic, a project manager is gathering persons with given competencies for a project, or a document writer needs some background information on a speciic topic. Secondly, the competence manager can be used as a tool during performance appraisal between superiors and their employees; the current competencies and the competence areas needing improvements can be identiied, a development path(s) can be planned, and employee development can be later tracked. Thirdly, the competence manager can serve as a tool for company management in strategic coordination activities. The overall competence levels on various areas can be identiied, and, in the case of deining the companys business strategy, the competence manager can be used for competence gap analysis; the areas needing more or less competence can be identiied.
Ontologies in the Competence Manager

Competence manager employs OWL (McGuinness & Harmelen, 2004) for the ontologies, which are divided into general and domain-speciic parts. The general ontology deines the mandatory classes and properties for the competence topic hierarchy. The domain ontologies deine the domain-speciic resources, and must conform to the general ontology. This asserts that the reasoning engine using simple protocol for RDF query language (SPARQL) (Prudhommeaux & Seaborne, 2005) queries works correctly with the domain ontologies. The general ontology deines the topic hierarchy, resources, and their properties. Figure 51 depicts the classes and properties deining the topic hierarchy, which in the domain-speciic ontology is realized as an instance of the topic class. The parent-child relations are deined using the hasParentTopic property. The instances of the grade class specify the competence level of some resource (e.g., a persons competence) to a certain topic. The grade value in the competence manager can get integer values from one to ive. For instance, an employee having excellent knowledge on some topic may set the grade value to ive, whereas some other employee with only some background information about the same topic could set the
Figure 5. General ontology for competence topics
value to one. To the end-user, the grade values are presented using more descriptive terms. For instance, a grade value of ive could be presented as excellent competence, whereas a grade value of one could map to poor competence. In order to provide semantically sound results, the hierarchy has to be designed well. In other words, semantically similar topics in the hierarchy have to be located close to each other in the topic hierarchy. This may sound obvious, but in some cases, it may be quite tricky to deine the topic hierarchy. Consider, for instance, topics C and C++. The former is procedural language, similarly to, for instance, Pascal, whereas the latter is an objectoriented language, like, for instance, Java. Therefore, C and Pascal as well as C++ and Java would be closer to each other than C and C++. However, those that are familiar with programming languages know that C and C++ are very close to each other in a certain
Figure 6. General ontology for resources
sense. To alleviate this problem, the competence manager allows deinition of similarity links, which can be used to explicitly state that two topics are similar to each other to some extent. Thus, similarity links bring the topics closer to each other regardless of how far they are located in the topic hierarchy. The modeling of the similarity is depicted in Figure 5. The similarity class has two properties, sourceTopic and destTopic, with which the unidirectional similarity link is deined. The similarity link also has a non-negative weight deining the closeness of the two topics. The other (resource-related) classes and properties deined in the general ontology are depicted in Figure 6. The group class allows tying a group of resources together. This is useful for instance when modeling static organizational units within a company, but also when creating cross-organizational groups of persons with given competencies dynamically. The resource class is an abstract class, which is extended by concrete person, project, and document classes. HasResource is an abstract property that can be used to link domain-speciic objects (things) to the resources of the competence management ontology (see examples in Figure 7). For instance, the hasMember property is a sub-property of hasResource, and links a person to a project. Furthermore, each resource can be availablespeciied using isAvailable propertyduring a speciic time period(s), which can be speciied by the calendar class. The calendar class in the general ontology is an abstract class allowing the domain-speciic ontology and application area to use whichever calendar is preferred. The hasCompetence and hasInterest abstract properties are for deining competencies and interests, respectively, by linking a resource to a topic (through a grade instance). The hasCompetence property speciies that a resource has some competence on topic. The hasInterest, on the other hand, speciies a desire of a resource to learn more about some topic. Similarly, a negative hasInterest implies that a resource dislikes a topic and does not want to or cannot learn about nor deal with the topic. As with the hasResource, the hasCompetence and hasInterest properties can and should have more specialized sub-properties. An example of such sub-property is the hasTopic, which states that a project can be about some topic. The domain ontology allows further deining such classes, properties, and their instances, which are speciic to the application area and which are not covered in the general ontology. The resources can be freely deined as domain-speciic classes, as long as they are sub-classes of the resource class. Furthermore, the relationships (properties) between the domain-speciic resources can also be freely deined, but the properties must be deined as a sub-property of the hasResource property. Similarly, competence relationships have to be sub-properties of the hasCompetence property. The domain ontology depicted in Figure 7
Figure 7. A sample of a domain ontology
Figure 8. Sample instances for a topic hierarchy
gives examples of these kinds of specializations. Firstly, it deines sub-properties knowsAbout and knowsHowto for the hasCompetence. Secondly, it extends the person class for deining more ine-grained roles in a company by sub-classes employee, manager, and consultant. Thirdly, the document class is extended to allow a distinction between an InternalDocument and a PublicDocument. Once the general and domain ontologies have been deined, the application area-speciic data can be created by making instances of the ontologies. Figure 8 depicts a sample fragment of a topic instance hierarchy under the software technologies topic, while Figure 9 depicts a sample instances for the domain ontology. In both igures, a rectangle represents an instance, and the identiier inside < and > characters denotes the class, which the instance realizes. The properties follow the same scheme as in Figures 5, 7, and 8. In Figure 9, the instance Mikko of an employee class has competencies on topics OWL and Java, with grade levels 4 and 5, respectively. Mikko has also a calendar instance attached to it specifying the availability of Mikko. Further, Mikko is an author (authorOf) in a public document named Deliverable 2.1, which is about RDF(S). The Deliverable 2.1 again belongs to a project called Semantic Web prestudy, which is about Semantic Web.
Figure 9. Sample instances of resources linked to topic instances
Finding Closest Matches

In a case where direct matches are not found, the people with competence on nearby topic(s) can be searched for. The distance calculation between competence topics is based on formally deined semantics and an algorithm called Object Match (OM) (Stojanovic, Maedche, Staab, Studer, & Sure, 2001), which calculates the distance to the neighboring topics and returns the resources related to the closest ones. The object match uses an algorithm called upward cotopy (UC) (Stojanovic et al., 2001), which measures the distance from an object Oi to the root node of a hierarchy H by collecting all the objects Oj to a set:
UC (Oi , H ) := {O j H (Oi , O j ) O j = Oi }
(1)
Figure 10. Object hierarchy H1 with depth of 3
Figure 11. Object hierarchy H2 with depth of 4
The distance between any two objects O1 and O2 is then calculated by the object match algorithm, which is deined as:
OM (O1 , O2 , H ) := UC (O1 , H ) UC (O2 , H ) UC (O1 , H ) UC (O2 , H )
(2)
The output value of the object match algorithm is a loat value between 0 and 1; the greater the value, the closer the objects are to each other.
UC ( E , H 1 ) = E , C , A UC ( F , H 1 ) = F , C ,A
Figures 10 and 11 depict two simple object hierarchies, which we will use to illustrate the upward cotopy and the object match algorithms. Let us irst consider Figure 10. The distance between objects E and F for hierarchy H1 are calculated by irst deining their upward cotopy:
OM ( E , F , H1 ) :=
= =
| UC ( E , H1 ) UC ( F , H1 ) | | UC ( E , H1 ) UC ( F , H1 ) |
| E, C, A F , C, A | | E, C, A F , C, A | | C, A | 2 = = 0,5 | E , C , A, F | 4
The distance between E and F is then as follows:

UC ( E , H 2 ) = E , C , A, X UC ( F , H 2 ) = F , C ,A, X
OM ( E , F , H 2 ) := | UC ( E , H 2 ) UC ( F , H 2 ) | | UC ( E , H 2 ) UC ( F , H 2 ) |
= =
| E , C , A, X F , C , A, X | | E , C , A, X F , C , A, X | | C , A, X | 3 = = 0,6 | E , C , A, F , X | 5
Figure 12. Closest matches when searching for Topic Perl
One of the most powerful features of the object match algorithm lies in its ability to take into account the depth of the object hierarchy. That is, the objects deeper in the hierarchy are also semantically closer to each other. This can be seen by calculating the object match in the case of hierarchy H2 as shown in Figure 11. As we can see, by adding a new root nodethat is, making the hierarchy deeperaffects the object match value of E and F; in Figure 11 E and F are semantically closer to each other than in Figure 10. The object match is especially useful in cases where the best candidates (i.e., the persons having competence directly on the searched topic(s)) are not available, but some candidates having competence on nearby topic(s) are found. For instance, when persons with given competencies or other resources on the searched topic Perl are not available, the persons having competence onor the resources related totopics TCL, Python, or even C, which does not have the same direct parent topic with the others, might be close enough to the searched topic, and they could be still accepted. This is depicted in Figure 12, where the closest topics (with percentages) to Perl are shown. Resource match function calculates the relevance of a resource to a searched topic:
1, s O n RM ( s, O, SG, H ) = 1 OM ( s, Oi , H ) * SGi , s O i =1
(3)
where s is the searched topic, O is a set of topics to which the resource has a relation of type hasCompetence or hasInterest, and SG contains scaled grade values for each Oi 0 O. Each scaled grade value (SGi) is calculated based on the corresponding grade value and the distance between s and Oi. When applying the resource match to the example illustrated in Figure 12, the closest match for the search would be Mikko Laukkanen, who has direct competence on topic Python. This is because the topic Python matches with a percentage of 85.73, which is higher than, for instance, with Heikki Helin, whose competence on C language would produce a match with a percentage of 66.01. The resource match can also be extended to cover multiple topics. In doing so, the searcher deines a set of competence topics with importance weights, for which the best candidates should be found. The combined resource match can be calculated as follows:
1 * RM ( Si , O, SG, H ) * Wi S i =1
S
CRM ( S , O, SG,W , H ) =
(4)
where S is a set of searched topics, and W contains the importance weights for each Si 0 S. The combined resource matching is useful when employees with multiple competencies are searched for. Referring back to Figure 12, let us assume we would like to search for a person who has competence on both Pascal and Perl programming languages. In addition, the competence on Pascal is considered far more important than Perl; thus, on the scale from one to ive, Pascal is given a weight of ive, whereas Perl is given a weight of 1. Following
Figure 13. Competence manager architecture
the equation 4, the closest match in this case would be Heikki Helin with a percentage of 82.45, while Mikko Laukkanen would have a weaker match with a percentage of 69.29.
The Competence Manager Architecture

The architecture and the general parts of the competence manager are depicted in Figure 13. The users (clients) access the system through a front-end server, which renders the user interface according to the client device. The front end is implemented as a Microsoft .NET Web application, because it allows an easy and fast way to develop graphic user interfaces that are able to adapt to a variety of client devices. The back-end of the competence manager is exposed as a Web service, which the front-end server accesses in delivering the requests from the user interface. One reason for choosing the .NET for implementing the front-end is its built-in support for Web services. However, the front-end could be implemented with other technologies as well, such as Java-based application servers. Furthermore, due to the Web service interface, the competence manager could be accessed by Web service-enabled clients or devices directly. The Java Web Service Development Pack (JWSDP) is used to implement the Web service interface and the implementation, and the Tomcat application server hosts the deployed Web service. The Web service implementation forwards all the requests to the competence manager, which implements the real functionality. This loose coupling between the Web service implementation and the competence manager makes it easy to switch the JWSDP to another Web service application server, if desired. The functionality in the competence manager is divided in three modules. The search module implements all functions that are related to search requests from the user interface. The admin module, on the other hand, handles all the management functionality, such as
updates to the ontologies. The common module implements functions, which can be shared by the other modules. These functions include, for instance, initializations and wrappers for accessing ontologies and SPARQL queries (Prudhommeaux & Seaborne, 2005). All the modules have access to the ontologies and SPARQL query database, which are implemented on top of Jena (McBride, 2002).
Using the Competence Manager

The operations the competence manager supports can be divided into two parts. The management interface can be used to manage the domain ontologies, and it supports functionality to import new ontologies and modify the existing ones. The browsing interface can be used to make queries to the competence manager and display (graphical) reports about the information stored in the knowledge base. In the following subsections, we will cover these operations shortly.
Managing the Domain Ontologies

The management interface allows importing new domain ontologies to the competence manager and managing the existing ones. The competence manager supports multiple domain ontologies, which can be imported at runtime. In addition to maintaining multiple simultaneous domain ontologies, the portabilityfor instance, from one organization to anotherdoes not require changes to the implementation. The only thing that has to be done is to deine the domain ontology and import it to the competence manager.
Figure 14. Search results with exact and closest matches
Figure 15. Details of a resource and information about the other resources
The management of the existing ontologies is about adding, modifying, and deleting the topics and resources of the domain ontology. For instance, after attending a course on Java programming language, the employee may increase her competence on Java to the higher level and optionally add to the competence description about the things learned on the course. Respectively, other kind of domain-speciic resourcessuch as persons, documents, and projectscan be managed for instance by linking people and documents to projects.
Making Queries
The competence manager provides three ways of making queries to the knowledge base. The regular expression-enabled searches (see Figure 14, search area) can be used to ind some speciic resources. These kinds of searches do not make use of any ontological relationships but apply the search on all the instances in the knowledge base. The second kind of query takes a competence topic as a parameter and returns all the related resources found in the knowledge base. In this case, only the exact matchesthat is, resources referring to that speciic topic in the queryare displayed. For the exact matches, the knowledge level is displayed as bullets ranging from one to ive (see Figure 14, exact matches area). The current availability of each resource is also indicated using a graphical symbol. Competence manager also allows multiple search terms to be used, in which case the search will combine the matches. If there are no exact matches, the third kind of query allows querying the resources that are closest to the queried topic, as described above (see Figure 14, closest matches area).
Once the resource(s) has been found, the topics and other related resources can be retrieved (see Figure 15). The calendar showing availability (free/busy) information can be also queried.
Discussion and Future Trends

This chapter and the competence manager concentrate on the representationontologiesof the competencies and other related resources, and describes how this kind of system allows intelligent and lexible management and searching operations. However, a successful competence management system also needs to implement eficient ways to acquire knowledgecompetencies in our caseas easily and automatically as possible. As Abecker, Bernardi, Hinkelmann, Khn, and Sintek (1998, p. 46) claim concerning knowledge-based systems and organizational memories: Knowledge acquisition and maintenancethe main reasons why knowledge-based systems so often failed in industrial practicealso pose a serious challenge for organization memories. They also suggest that the existing data sources should be exploited and automatic tools for knowledge acquisition should be used. In our system, we expect that the end-usersemployees in an organizationmaintain their own competencies, but we acknowledge the risks of this approach in terms of usability and user acceptance. In order to keep the end-users maintaining their competencies, they should have a feeling that they beneit from doing so. In the future, we will improve our system to exploit the existing IT infrastructuree-mails, documents, groupware systems, and different kinds of system logs and tracesfor acquiring and extracting competencies (semi) automatically. Trust plays an essential role in systems like competence manager, which are open to all the employees of an organization and where the employees may state their own competencies. The essential challenge is to assure that the information published is accurate. We believe that in this kind of collaborative systems, the abusers will eventually be exposed. Reputation among superiors and other employees is important, and it would be unwise to ruin it by publishing false competencies. Although the competence manager supports multiple domain ontologies, it does not support automatic sharing and merging of those ontologies. This is not in line with the principles of the Semantic Web. In the future we will work toward more open ontology management, where different domain ontologies may borrow concepts from each other. These include both topics and other resources. For instance, the same person having competence on topics in her companys topic hierarchy could also have competence on topics deined by her gardening club. The scenarios then would be extended to cover cases where the competence manager could be used between multiple organizations. In doing so, our system could be used as a tool for inding organizations based on their real competencies.
Conclusion
The competence management is becoming increasingly important in competitive environments, especially in knowledge-based organizations. The competence management systems are made available for all the members of a company. These kinds of systems have to be as easy to use and maintain as possible. They must provide as much intelligence in terms of helping the user when maintaining his/her competencies and provide semantically sound results when making queries. This would require that the information in the competence management system is in a machine-accessible and machine-understandable format. The Semantic Web technologies provide tools for implementing such a competence management system. This chapter described how the Semantic Web technologies can be used in competence management. When applying the Semantic Web technologies in competence management, we can build systems that beneit from the use of machine-accessible semantics, support highly dynamic environments, and are extensible as well as interoperable between different application domains. In Semantic Web-based approach not only those employees that have competence directly on a topic area can be found, but also employees that have competencies on some closely related areas. This means that we can look up persons who are the best possible candidates for a competence topic. We also introduced an innovative Semantic Web-based tool, called competence manager for managing employee competencies and other relevant resources (documents, customers, projects, and so on). The knowledge base in competence manager is modeled as ontologies. Therefore, it demonstrates how the Semantic Web technologies can be used in a practical application on the area of competence management.
References
Abecker, A., Bernardi, A., Hinkelmann, K., Khn, O., & Sintek, M. (1998). Toward a technology for organizational memories. IEEE Intelligent Systems 13(3), 40-48. Allen, C. (Ed.). (2006). Competencies (measurable characteristics). HR-XML consortium recommendation. Retrieved March 14, 2006, from http://ns.hr-xml.org/2_4/HR-XML2_4/CPO/Competencies.html Barclay, R. O., & Murray, P. C. (1997). What is knowledge management. Retrieved January 25, 2005, from http://www.media-access.com/whatis.html Barnes, J. A. (1954). Class and committees in a Norwegian island parish. Human Relations, 7, 39-58. Becket, D. (Ed.). (2004). RDF/XML syntax speciication (revised), W3C recommendation. Retrieved January 25, 2006, from http://www.w3.org/TR/rdf-syntax-grammar/ Berners-Lee, T. (2001). Notation 3: An RDF language for the Semantic Web. Retrieved January 25, 2006, from http://www.w3.org/DesignIssues/Notation3.html
Berners-Lee, T., Fielding, R., & Masinter, L. (1998). IETF RFC2396: Uniform resource identiiers (URI): Generic syntax. Retrieved January 25, 2005, from http://www.ietf. org/rfc/rfc2396.txt Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The Semantic Web. Scientiic American, 284(5), 34-43. Brickley, D., & Guha, R. V. (Eds.). (2004). RDF vocabulary description language (1.0): RDF schema, W3C recommendation. Retrieved January 25, 2005, from http://www. w3.org/TR/rdf-schema/ Brickley, D., & Miller, L. (2005). FOAF vocabulary speciication, namespace document 27. Retrieved January 25, 2006, from http://xmlns.com/foaf/0.1/ Gruber, T. R. (1993). A translation approach to portable ontology speciications. (Tech. Rep. KSL 92-71). Stanford, CA: Stanford University, Computer Science Department. Harzallah, M., & Vernadat, F. (2002). IT-based competency modeling and management: From theory to practice in enterprise engineering and operations. Computers in Industry, 48(2), 157-179. Hendler, J., & McGuinness, D. L. (2000). The DARPA agent markup language. IEEE Intelligent Systems, 15(6), 67-73. Lindgren, R., Stenmark D., & Ljunberg, J. (2003). Rethinking competence management systems for knowledge-based organizations. European Journal of Information Systems, 12(1), 18-29. McBride, B. (2002). Jena: A Semantic Web toolkit. IEEE Internet Computing 6(6), 55-59. McBride, B. (2004). RDF primer, W3C recommendation. Retrieved January 25, 2006, from http://www.w3.org/TR/rdf-primer/ McGuinness, D., & Harmelen, F. (2004). OWL Web ontology language overview, W3C recommendation. Retrieved January 25, 2006, from http://www.w3.org/TR/owl-features/ OLeary, D. E. (1998). Enterprise knowledge management. Computer, 31(3), 54-61. Porter, M. E. (1998). Competitive strategy: Techniques for analyzing industries and competitors. New York: Free Press. Prudhommeaux, E., & Seaborne, A. (2005). SPARQL query language for RDF, W3C working draft 23. Retrieved January 25, 2006, from http://www.w3.org/TR/rdf-sparql-query/ Sicilia, M. A. (2005). Ontology-based competency management: Infrastructures for the knowledge-intensive learning organization. In M. Lytras & A. Naeve (Eds.), Intelligent learning infrastructures in knowledge intensive organizations: A Semantic Web perspective (pp. 302-324). Hershey, PA: Information Science Publishing. Staab, S., Studer, R., Schnurr, H. P., & Sure, Y. (2001). Knowledge processes and ontologies. IEEE Intelligent Systems, 16(1), 26-34. Stojanovic, N., Maedche, A., Staab, S., Studer, R., & Sure, Y. (2001). SEAL: A framework for developing SEmantic portALs. In Proceedings of the International Conference on Knowledge Capture (pp. 155-162). Sunassee, N., & Sewry, D. (2002). A theoretical framework for knowledge management implementation. In Proceedings of the 2002 Annual Research Conference of the South
African Institute of Computer Scientists and Information Technologists on Enablement through Technology, 235-245. World Wide Web Consortium. (2001). Semantic Web. Retrieved January 25, 2006, from http://www.w3.org/2001/sw/
Endnote
1
The notation of all the ontology igures is based on VisioOWL; see http://web.tampabay.rr.com/lynn/VisioOWL/VisioOWL.htm.
358 About the Authors
About the Authors
Miguel-Angel Sicilia obtained a degree in computer science from the Pontiical University of Salamanca in Madrid, Spain (1996) and a PhD from Carlos III University in Madrid, Spain (2002). In 1997, he joined an object-technology consulting irm, after enjoying a research grant at the Instituto de Automtica Industrial (Spanish Research Council). From 1997 to 1999, he worked as assistant professor at the Pontiical University, after which he joined the computer science department of the Carlos III University in Madrid as a lecturer, working simultaneously as a software architect in e-commerce consulting irms and serving as a member of the development team of a personalization engine. From 2002 to October 2003, he was a full-time lecturer at Carlos III University, working actively in the area of adaptive hypermedia. Currently, he works as a full-time professor at the computer science department of the University of Alcal (Madrid), Spain, and leads the Information Engineering Research Unit. * * * Alfonso Urquiza is a physicist specializing in computer sciences at the Complutense University of Madrid, Spain (1974). Alfonso has vast international experience within the IT and communications services industries. From 1974 to 1983, he worked in the design and implementation of public switching systems within the ITT group in Spain, France and Belgium. For a period of nine years (until 1996), he worked at Entel developing command and control systems for defence applications and the European Space Agency. For 20 months,
About the Authors 359
he worked testing and developing software for a radar System at Hughes Aircraft in the U.S. From 1996 until 2004, he worked in information and communications services engineering activities within the Telefonica Group. Since 2004, Urquiza has been a full-time professor at the IT High School in the Francisco de Vitoria University, Madrid. Ambjrn Naeve has a background in mathematics and computer science, and received his PhD in computer science from KTH (1993). He is presently co-ordinator of research on interactive learning environments at the Uppsala Learning Lab at Uppsala University as well as the head of the Knowledge Management Research (KMR) Group at KTH (http://kmr. nada.kth.se). The KMR group carries out research and development on how to make use of Semantic Web technology in order to enable more eficient forms of technology-enhanced learning and administration. Prominent among the KMR tools are the frameworks SCAM and SHAME, the concept browser Conzilla (www.conzilla.org), and the electronic portfolio system Confolio (www.confolio.org). The KMR group is active within several international networks in technology-enhanced learning and Semantic Web, notably Prolearn (www.prolearn-project.org), SIGSEMIS (www.sigsemis.org), and Sakai (http://sakaiproject.org). April Ng was a graduate student at Simon Fraser University, Canada. She received her BSc and MSc in 2002 and 2005, majoring in computing science and information technology. While Ng was pursuing her degree, she was very interested in the Semantic Web, information theory, ontology, and knowledge management. She is actively engaged in applying computer technology to education, knowledge sharing, and information integration. With the use of ontology, she attempted to formalize the framework on generating competency to enhance accuracy in encoding personal knowledge of individuals. This has proven important in the e-learning domain. Azzurra Ragone received the laurea cum laude degree in management engineering in 2004. She is currently pursuing her PhD at the Technical University of Bari, Italy. Her research interests include knowledge representation, description logics, and negotiation theory with applications to e-commerce and Web Services composition. Chris Kimble is a lecturer in information systems and management in the Department of Computer Science at the University of York, UK. Before moving to York, he was lecturer in IT at the University of Newcastles Business School and a researcher in both the business school and the Department of Computer Science at the University of Northumbria. His broad area of research is knowledge management. His particular area of interest is communities of practice and the problems associated with supporting distributed working in cross-cultural or trans-national contexts. He is the leader of management and information research group at York and a co-founder of the Northern Interest Group on Information Systems and Organisations. He has supervised numerous PhD students and has published more than 60 articles in journals, conferences, reports, and books. He has also organised conferences, seminars, and workshops at York and elsewhere. Kimble is a member of the editorial board of the journal Information Research and was previously a co-editor of the journal The Information Systems Review. He is currently a member of the review panel
for the journal The Learning Organisation and has acted as a reviewer for numerous other academic journals. He also has acted as a reviewer for several publishers and as consultant for two major information systems textbooks. For further details visit http://www-users. cs.york.ac.uk/~kimble/ and http://www.cs.york.ac.uk/mis/. Dietrich Albert (http://css.uni-graz.at/staff/albert/) graduated from the University of Gttingen in 1966 with an MS in psychology. In 1972, he received his Dr.rer.nat. (DSc) and in 1975 his postdoctoral degree (habilitation in psychology) from the University of Marburg/Lahn. He was a professor of general experimental psychology at the University of Heidelberg from 1976 to 1993. Since 1993, he has been a professor and the head of the cognitive science section at the University of Graz. He is member of several scientiic societies (e.g., European Association for Research on Learning and Instruction, Asia-Paciic Chapter of Association for the Advancement of Computing in Education, and the SMP) and advisory boards (e.g., Center for Psychological Information and Documentation, ZPID; Know-Center, Academy of New Media and Knowledge Transfer). Dirk Frosch-Wilke obtained a degree in mathematics and business administration from Germanys University of Trier (1990) and a PhD in computer science from the University of Trier (1994). Frosch-Wilke has been a professor of business information systems at the University of Applied Sciences Kiel since 1997. He also worked as a department head in the IT section of Deutsche Bank AG during 1994 and 1997. His research interests include the strategic applications of information technology to organizational productivity, electronic business, learning objects reusability, and software engineering techniques. He has consulted for companies on e-commerce and software development projects. He has presented papers at both national and international conferences. He is editor of the book Marketing Communication in the Internet (Braunschweig, 2002), and he has contributed chapters to other texts. Eugenio Di Sciascio received a laurea cum laude degree in electronics engineering from the University of Bari in 1989 and a PhD in 1994 from the Technical University of Bari. In 1992, he joined the University of Lecce as an assistant professor. He is currently a full professor in information technology engineering at Technical University of Bari, where he is responsible for several research projects and leads the SisInfLab. His main research interests include multimedia information retrieval, knowledge representation, and knowledge-based systems for e-commerce and Semantic Web. On these topics, he has authored or co-authored several papers in international journals and international conferences. Fotis Draganidis is a PhD candidate at the school of electrical and computer engineering at the National Technical University of Athens (NTUA), and his research interests lie in the areas of competency and knowledge management, ontologies, Semantic Web, and e-learning. Since 2000 he has been working in Microsoft Hellas as an academic programs manager. Francesco Maria Donini earned his masters degree in electronics engineering from the University of Rome La Sapienza (1988). He received a PhD in computer science from the
same university in 1992. From 1991 to 1998, he was a researcher/assistant professor at the Department of Computer and System Science at the University of Rome La Sapienza. He is currently a full professor at University of Tuscia in Viterbo. He is the co-author of many papers in international journals as well as international conferences. The papers Tractable Concept Languages, IJCAI-91 (1991), and Concept Abduction and Contraction for Semantic-based Discovery of Matches and Negotiation Spaces in an E-Marketplace, ICEC 04 (2004), received the best paper award. He has been responsible for local university research projects since 1996, as well as for CNR research projects. He is editor of the area concept-based knowledge representation for the journal ETAI-Electronic Transactions on Artiicial Intelligence, published by the Royal Swedish Academy. Gregoris Mentzas is a professor of information management at the School of Electrical and Computer Engineering at the National Technical University of Athens (NTUA) and director of the Information Management Unit (IMU), a multidisciplinary research unit at the university. During the 2006-2009 period, he is serving on the board of directors of the Institute of Communication and Computer Systems of NTUA. His area of expertise is information technology management, and his research concerns the integration of knowledge management, Semantic Web and e-service technologies, collaboration and worklow management, and corporate knowledge management in e-government and e-business settings. Heikki Helin obtained his PhD in computer science from the University of Helsinki (2003), focusing on software agent technology in nomadic environments. He is a principal researcher at TeliaSonera Finland. Previous to working at TeliaSonera, he was a researcher in the department of computer science at the University of Helsinki. His research interests include software agent technology, wireless environments, and Semantic Web technologies. He is a member of ACM and the IEEE Computer Society. Jo Boon is an associate professor of educational sciences at the Open University of The Netherlands. She specialises in the contribution of education to the labour market positions of students, alumni policies of universities, and quality assurance policies. She publishes on competency-based education, deining competencies, and evaluation of e-learning. Jos Braga de Vasconcelos is an associate professor in algorithms and data structures and software engineering in the faculty of science and technology at the University Fernando Pessoa (Porto, Portugal). He has a PhD in computer science from the University of York (UK). His main area of research is knowledge management and engineering, and he has particular interest in corporate competence management. Most recently, he has been working on healthcare information systems, such as clinical decision support systems. In this context, he is a member and a founder of the Research Group on Medical Informatics at the University Fernando Pessoa. Juan G. Cegarra Navarro is an assistant professor of business management in the Polytechnic University of Cartagena. His research is focused on e-government, knowledge management,
and marketing orientation. He is currently a member of the ELIS group (E-government for Local Integration with Sustainability). Kathleen Schlusmans is an associate professor of educational sciences at the Open University of The Netherlands. She specialises in designing and implementing curricula, and quality assurance of education. She publishes on competency-based learning, training teachers, and designing curricula. Marek Hatala is an associate professor at the School of Interactive Arts and Technology at Simon Fraser University and director of the Laboratory for Ontological Research. Hatala received his MSc in computer science and PhD in cybernetics and artiicial intelligence from the Technical University of Kosice. His primary interests lie in areas of knowledge representation, ontologies and Semantic Web, user modeling, intelligent information retrieval, organizational learning, and e-learning. His current research looks at how semantic technologies can be applied to achieve interoperability in highly distributed and heterogeneous environments, the social and technical aspects of building a distributed trust infrastructures, and what role user and user group modeling can play in interactive and ubiquitous environments. Malgorzata Mochol is a graduate research assistant at the Freie Universitt Berlin at the Institute for Computer Science. In 2003, she earned a degree in computer science at the Technical University Berlin. She is currently working on a number of research projects, including InterVal: Knowledge Nets, which uses scenario analysis to examine the impact of Semantic Web (SW) technologies on different markets; the Reisewissen project, which aims to develop a SW-based system for selecting and ranking hotels; and the KnowledgeWeb: Network of Excellence that supports the transition of SW technology from academia to industry. Moreover, she is developing the MOMA Framework (metadata-based ontology matching framework), which will allow both manual (AHP-based) and automated (rule-based) detection of suitable matching approaches for the processing of a pre-deined ontological input, as well as working on the project ONTOCOM, whose aim is to create methods and tools for estimating the costs involved in ontology engineering processes. Marcel van der Klink is an associate professor of educational sciences at the Open University of The Netherlands. He specialises in vocational and professional education, and human resource development. He publishes on competency-based learning, workplace learning, and e-learning and assessment. Marcus Specht received his diploma in psychology in 1995 and completed a dissertation at the University of Trier in Germany in 1998 on adaptive learning technology. From 1998 until 2005, he worked as a senior researcher at the Fraunhofer Institute for Applied Information Technology (FIT) in Germany. He has rich experience in educational and contextualized technologies for learning and personalized services. His main research interests are adaptive systems, contextualized computing, and intelligent interfaces. He was involved in several national and European projects for applying contextualization and personalization in the ields of e-learning and collaborative systems. Since 2005, he has worked as an associated
professor for educational technologies at the Open University of The Netherlands. His current research topics include learning design, personalized learning support, mobile and contextualized learning, and competence management. Marie-Hlne Abel received a PhD from the University of Compigne (France) in 1994. She was an associate professor for four years at the University of Amiens before joining the University of Compigne in 2000. She has been in charge of the knowledge engineering and information media specialization since 2003. She has published numerous papers in journals and conferences in the e-learning or knowledge engineering domain. She was in the scientiic committee of conferences on e-learning and knowledge management. She takes part in European research projects. She is a member of the Network of Excellence Kaleidoscope, the European Network Taconet, and the Web Intelligent Consortium. She is a World Scientiic and Engineering Academy and Society fellow. Maria Mantziou graduated from the Department of Primary Education, School of Education at the University of Ioannina, Greece. She is a studies in education postgraduate student at the Open University of Greece. Her research interests relate to the adoption of information and communication technologies in schools and the management of education. She has participated in several projects related to new technologies in schools, such as eTwinning.net. Maria Pontikaki has graduated from the Department of Informatics of the Athens University School of Economics & Business in 2001. She also holds an MSc from Hofstra University in New York and is now a PhD candidate at the University of Alcala, Spain. She completed a masters degree in health management at the University of Piraeus, Greece. She has participated in many research and development e-business projects as a project manager and IT specialist and provides expertise to several organizations. Her research interests relate to the application of the Semantic Web, knowledge management technologies, and ubiquitous networks to the management of health Mikko Laukkanen obtained his MSc in information technology from Finlands Lappeenranta University of Technology in 1999. He is currently working for TeliaSonera Finland as a senior researcher and studying for his PhD at the University of Helsinki. In his research, he concentrates on Semantic Web technologies, agent technologies, and mobile computing. He is especially interested in how Semantic Web technologies can be applied to various applications areas, such as competence management. Miltiadis Lytras is a faculty member in the Department of Business Administration at the University of Patras. His research focuses on the Semantic Web, knowledge management, and e-learning, with more than 80 publications in these areas. He has co-edited 14 special issues in international journals and has authored or edited six books. He is the founder and leader of the Semantic Web and information SIG in the Association for Information Systems (http://www.sigsemis.org). He serves as the editor-in-chief for the International Journal on Semantic Web and Information Systems, International Journal of Knowledge and Learning, International Journal of Teaching and Case Studies, and AIS SIGSEMIS Bulletin.
He is also the book series editor for knowledge and learning society in IGP (http://www. idea-group.com). Radoslaw Oldakowski is a graduate research associate at the Department of Business and Economics at the Freie Universitt Berlin, where he obtained his university degree (German: Dipl.-Kfm.) in business information systems. He is currently working on the research project InterVal-Knowledge Nets, which uses scenario analysis to examine the impact of semantic Web technologies on markets, enterprises, and individuals in various application domains. Furthermore, he is a member of the research team in the Reisewissen project, whose job it is to develop a Semantic Web-based system for the searching and booking of hotels. Oldakowski is also author of SemMF, a semantic matching framework, as well as co-developer of RAP, a Semantic Web framework for PHP. Ralf Heese is a research assistant in the Department of Computer Science, Databases, and Information Systems at the Humboldt-Universitt zu Berlin. In 2001, he received his diploma in computer science. Between 2001 and 2003, he worked on the project XML-Clearinghouse, which focused on the transfer of knowledge about extensible markup language (XML) technologies. He is now working on the research project InterVal: Knowledge Nets, which aims to evaluate the impact of Semantic Web technologies from business and technical viewpoints. Currently, his research focuses on query optimisation in RDF management systems. Rob Koper is a professor and director of learning technologies research at the Open University of The Netherlands. In this research, he studies and develop models and technologies to support the sharing of knowledge and learning resources in learning networks. This includes work on software agents, e-portfolios, competence interoperability, learning design tools, and social exchange mechanisms. Koper has more than 20 years of experience in the ield and served on numerous editorial boards, conference committees, management, and advisory boards. He was, among other things, responsible for the development of educational modeling language (EML), currently an open standard through the IMS consortium (IMS learning design). Salvador Sanchez-Alonso obtained a degree in computer science from the Pontiical University of Salamanca, Spain (1997) and a PhD in computing from the Polytechnic University of Madrid (2005). Dr. Sanchez-Alonso worked as an assistant professor at the Pontiical University of Salamanca from 1997 to 2000 and from 2002 to 2005. He also worked as a software engineer at a software solutions company in 2000 and 2001. Since 2005, he has been a lecturer of in the Computer Science Department of the University of Alcal. His current research interests include learning objects reusability, metadata, object-oriented technologies, and software and Web engineering. Shantha Liyanage is an associate professor of technology management at the business school of the University of Auckland, New Zealand, teaching technology and innovation management to graduate and undergraduate students. He holds a Bachelor of Science (Hons.) in biological sciences from the University of Colombo, Sri Lanka, and obtained his
doctoral degree from the University of Wollongong, Australia, in innovation management. Before joining the University of Auckland, Liyanage had an illustrious academic career in Australian universities. He was the principal research fellow at the University of Wollongongs Australian Research Council-funded Centre for Research Policy and contributed to the management of collaborative research of the Australian Cooperative Research Centres Programme. He was appointed to the University of Queensland, Brisbane, in 1996, and he became the director of the Technology Management Centre, faculty of engineering, at the University of Queensland in Brisbane, Australia. His contributions to Australian research and teaching efforts have been well recognised, and he was also instrumental in developing Australias irst Web-based online masters program, which launched simultaneously in Australia, Singapore, and Hong Kong. Liyanages scholarship is well known throughout Asia, Europe, and the U.S. He was a founding member and the program manager of the Asian Science and Policy Network, sponsored by UNESCO, from 1984-1995. Dr Liyanage has extensive consulting experience with various governments and international development agencies, such as the Asian Development Bank, the World Bank, UNESCO, AusAid, and New Zealands International Aid and Development Agency. Liyanage has published extensively and has extensive engagement in his ield, including editorial appointments in leading international journals, and he was the editor in chief of the International Journal of Learning and Change. He has several visiting professorial appointments including the International Business School of Nihon University, Tokyo, Japan. Simona Colucci received a laurea cum laude degree in management engineering in 2002. She is currently pursuing her PhD at Technical University of Bari. Her research interests include knowledge representation, description logics, and their applications to knowledge management. She has authored various papers on these topics for journals and conferences. Stefanie N. Lindstaedt leads the division of knowledge management at the Know-Center in Graz, Austria. She is responsible for the management of many large, multi-irm projects and the scientiic strategy of the division. For more than 10 years, she has been leading interdisciplinary, international projects in the ields of knowledge management, e-learning, and software engineering. For the last ive years, she has focused on the issue of work-integrated learning, developing the concept of AD-HOC learning and performing research on competency development and modeling. She holds a PhD in computer science from the University of Colorado (CU) at Boulder (USA) and an MS in computer science on neural networks from CU. Tobias Ley studied psychology at Darmstadt University of Technology in Germany and Melbourne University in Australia, focusing on organizational psychology and human decision-making. He spent a year abroad as a Fulbright scholar at the Krannert Graduate School of Management at Purdue University in the U.S., where he studied human resource management and management information systems. Since 2001, he has been working at the Know-Center in Graz/Austria, where he has been leading industry-based projects in the areas of knowledge management, technology-enhanced learning, and competency management. Ley holds a PhD in cognitive psychology from Karl Franzens University in Graz.
Tommaso Di Noia received a laurea degree in electronic engineering from Technical University of Bari in 2002. In 2003, he started his PhD course. Currently, he is an assistant professor in information technology engineering at Technical University of Bari. His research interests include description logics and their application to semantic Web, e-commerce, and knowledge management. On these topics, he has published various papers in journals and international conferences. He is involved in various projects on his research interests and in the organization of scientiic events.
Index 367
Index
A
absorption capability 21 Academic Search Elite 203 acquire 43 acquiring licences 64 activity 225 AD-HOC environment 111 AD-HOC methodology 111 Addeco 176 AgeLine 203 align people 43 annotations 154 anticipated reciprocity 245 Appach 164 application ontology 150, 157 approval 50 architecture 164 ARIADNE 151, 166 artiicial intelligence (AI) 150 AS400 66 assessing 57 assessing competencies 93 available competencies 85
B
B2E 58 B31.1 application 148 basic learning 32 bid response process (BRP) 261 boundary objects 214, 215 business driver 67 business processing environment (BPE) 170 business risks 48 business source premier 203
C
CA 200, 202, 206 categorization of knowledge 128 category 321 CBT 325 chief executive oficers (CEOs) 12 Cisco Systems 176 classiication of industrial sector (WZ2003) 309 CM approach 86, 91 CM developers 262 CM initiative 99
368 Index
CM management systems 44 CM methods 85 CM process 41 CM strategy 43, 47, 61 CM system 262 cognitive capacity 25 coherency 225 common goals 245 communal relationships 245 community-related 31 competence 181 competence development 236, 239 competence development programs 240 competence management system 261 competence manager 343 competence ontology 264 competence performance matrix 95 competence performance structure 97, 99 competence relation 266 competence space 107 competence state 97 competencies management 121, 124 competencies required 85 competencies types 128 competency 85, 93, 185, 186, 189, 197, 198, 202, 204, 222, 225, 321 competency-based 185 competency-based (e-)learning 129 competency-based annotation 128 competency-based curriculum 227 competency-based framework 1 competency-based HRM 89 competency-based interoperability 129 competency-based learning 185, 227, 231 competency-based project stafing 90 competency-based social networking 129 competency-based training 186 competency acquisition 147 competency assessment 5, 105 competency assignments 105 competency catalogue 86 competency deinition 104, 189 competency deinition schema 5 competency elicitation interview 86 competency gap analysi 5 competency management 41, 51, 63, 85, 180
competency management information systems 4 competency management systems (CMS) 13, 320 competency model 328 competency observatory 247 composite courseware 288 composite organizational competency 26 composition process 280 comprehensive strategy 43 computer 339 concept abduction problem (CAP) 281 concept contraction 280 concept contraction problem (CCP) 280 concept covering problem (CCoP) 283 concept names 276 conceptualisation 259 conferring 23 constructors 277 consultant 346 consultation 23 consult menu 164 countable set 162
D
DAML 8 debate 23 deining competencies 93 deining reusable competency deinition (RDCEO) 193 deinition(s) 277, 321 demerits 13 demonstrated behaviour 321 Desktop 339 destTopic 345 develop 43 discretionary enhancements 50 distributed organizational knowledge base (DOKB) 173 DLs 276 document 345 domain ontology 150 durability 225
E
e-business 41
Index 369
e-business context 42 e-business transformation 45 e-government 208, 210 e-learning 32, 48, 84, 111, 147, 230, 238 e-learning academic programmes 13 e-learning application 153 e-learning platforms 242 e-learning systems 111 e-learning technology 6, 13, 20 e-learning tools 20 e-mail system 26 E-MEMORAe 146, 148 E-MEMORAe environment 164 e-portfolios 230 e-recruiting 48 e-recruitment 300 e-service 208, 213 EBSCO 203 EDTs 214 electronic distribution technologies (EDTs) 209 electronic world 208 employee-facing relations 43 employee life cycle domain 43 employee self service (ESS) 50 engineering unit 86 enterprise resource planning (ERP) 41 environment E-MEMORAe 160 ERP 54 ERP systems 235 evaluating the models 93 ex-ante 22 experiencing 23 expertise location 91 explicit or implicit 240 exploiting capability 21 extensible markup language (XML). 158 extension of IMS QTI 243 external relationship 211
fragmented market 45 fuzzy 222
G
generic ontology 150 German Federal Employment Ofice (BA) 306 GMS framework 260 goal, obstacles, actions, and prerequisites (GOAP) 7 GOAP 7 GOAP framework 7 governance structures 19 GREEDYsolveCCoP 285 group 345 Growth and Innovation Fund (GIF) 12 grow the business 47, 49, 50
H
human capital (HC) paradigm 42 hard competencies 148 hard OM 209 hasCompetence 345, 350 hasInterest 345, 350 hasMember 345 hasParentTopic 343 hasResource 345 hasTopic 345 HC market applications 48 heterogeneous knowledge 29 heuristics 33 hierarchical knowledge 187 hierarchical task 95 high beneits 48 higher-level learning 32 HR-BA-XML 306, 310 HR-XML 174, 199, 306, 310 HR-XML-SE 306 HRM 88 HRM purpose 86 HR practices 43 HTML site 301 human capital 42 human resources (HR) 86, 235, 325 human resources (HR) management 42
F
farmer 111 inite set 162 formalism 157
370 Index
human resources management (HRM) 41, 84 human resources ontology (HR ontology) 307
I
IBM 176 ICMP 192 ICT 225 ICT environment 123 IEEE 188, 193, 204, 205 IEEE LTSC 193 ill-contextualised 22 ill-conceived 22 improving 57 IMS 8 IMS-RDCEO 174 IMS consortium 174 IMS Learning Design Speciication 235 including extranet 20 inclusion assertions 277 individual-related 31 individual competencies 20 inluence 43 informal learning and community building 239 InformationBearingThing (or IBT) 181 information model 174 information repository 254 information structuring 101 information system 180, 209 information technology (IT) 1, 15, 41, 84, 234, 273, 319, 335 inquiry 23 integration of different levels 238 InternalDocument 346 internal relationship 211 Internet protocol (IP) 192 Internet technologies (semantic Web technologies) 302 Internet technology (IT) 209 interpretation function 94 intersectionOf 340 intra-organizational learning 29 intranet 20
irreducible solution 281 IT-intensive competency systems 8 IT environment 98 IT portfolios 47 IT systems 338
J
Java 337, 341, 346 Java language 337 JavaScript/HTML+SVG 164 job posting 304 job proiles 86
K
KLC model 178 KM 88 KMCI 169 KMCI lifecycle model 169 KM framework 169 KM management systems 44 KM purpose 86 knowledge-based competencies 29 knowledge-sharing 21 knowledge assets 42 knowledge base 191 KnowledgeClaimEvaluation 181 knowledge economy 37 knowledge life cycle (KLC) 169 knowledge management (KM) 44, 84, 169, 254 Knowledge Management Consortium International (KMCI) 169, 170, 173 knowledge management lifecycle 180 knowledge management tools 15 knowledge mapping 90 knowledge production (KP) 170 knowledge resources 236, 239 knowledge sharing 37 knowledge space theory 94 knowledge states 94 knowsHowto 346 Korossys competence performance 83 KOWIEN 310
Index 371
L
laptop 339 LD 8 learner support service 241 learning activities 236, 240 learning activity repository 5 learning management systems 195 learning objectives 247 learning object ontology 329 learning object repositories (LOR) 151 learning objects 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 19 5, 196, 199, 203, 204, 205, 206, 240 learning organization 17 learning outcome 192 learning path service 241 learning path speciication 244 learning performance 32 learning practices 28 learning processes 27 learning resources 154 learning services connector speciication 244 learning strategy 24 Ley 111 life cycle 173 lifelong learning 238 Lindstaedt 111 literature on competencies 126 LMS market 63 LOM 188, 190, 194, 205 LOM (learning object metadata) 151 LOM schema 194 LOR 151 low business risks 48
manager self service (MSS) 50 many-to-one matching 279 many to many 275 many to one 275 mapping of contents 247 MCBL 222, 227 Memorae project 147 merits 13 messages notiication 50 meta-level learning 32 meta-metadata 6 meta-ontology 150 metadata 185, 186, 188, 190, 193, 194, 195, 203, 204, 205 metaphors 35 Microsoft 176 minimal interpretation 95 Model XYZ 339 MSS technology 50 multi-disciplinary 227 MySQL/PHP 164
N
navigation service 241 networks for lifelong competence development 241 Niku 47 non-discretionary 47 North American industry classiication system (NAICS) 309
O
object-oriented languages 341 object Match (OM) 347 observing 57 OIL 8 one-to-one matching 279 one to one 275 ontological 194 ontological agreement 128 ontological representations 196 ontology 150, 185, 190, 191, 192, 194, 195, 196, 197, 199, 204, 20 6, 343 ontology-based competence management model 253
M
mainstream competency-based learning (MCBL) 222 mainstream e-learning 240 maintain 43 MAMAS 291 management by training 225 management competencies 36 management system 188
372 Index
ONTOlogy COst Model (ONTOCOM) 308 ontology languages (KOWIEN) 310 ontology reuse (PMreuse) 308 OntoProper 274 OpenCyc 181 OpenCyc knowledge base 170 optical interfaces 266 optimizing 19 organization 101 organizational competencies 20 organizational competency database 5 organizational learning 13, 36 organizational learning schedule 5 organizational memory (OM) 152, 207 organizational needs registry 5 organizational planning database 5 organizational procedures 14 OS environment 66 overall frameworks 89 OWL 8, 346 OWL-DL 279 OWL-Full 279 OWL-Lite 279
product-related 31 profession reference number classiication (BKZ) 309 programming Language 337, 341 project 345 project management 101 project team stafing 90 ProPer 274 ProSight 47 PsycINFO 203 PublicDocument 346 Python 350
Q
Quantity-ScriptPerformance 181
R
rankPotential algorithm 281 RDCEO 193, 194, 204 RDF 8 RDF document 339 RDF format 311 RDF Schema 339 relection 23 relational relationship 211 reliable comprehensive process automation 67 repertory grid technique (RGT) 100 representation function 94 reputation 245 research and development (R&D) 99 resource 345 retain 43 return on investment (ROI) 242 rewards 47 right employees 43 risk 46 role expressions 277 role names 276 routing 50 Royal Charters 19 run the business 47
P
Pascal 341 PCT 216 PC technologies (PCT) 209 PDA 67 pedagogical content 152 pedagogical models 237 people-CMM process areas 45 People Capability Maturity Model 255 PeopleSoft GmbH 176 performance 95 performance assessment service 241 personnel administration 42 PMreuse 308 portfolio management 46 positioning service 241 potential development 48 Primavera 47 primitive competence 266 pro-active sharing 247 process integration 67
S
satisicing 19 ScriptPerformanceAttributeType 181
Index 373
SDCBL 222, 229 Searchable Answer Generating Environment (SAGE) 274 search menu 164 self-control of participants in learning 247 self-directed competency-based learning (SDCBL) 221, 231 self-Organization Theory 246 self-sustaining system 246 Semantic-based 271 Semantic base 280 Semantic Web 8, 301, 338 Semantic Web-based 355 Semantic Web-based tool 355 Semantic Web technologies 6, 299, 333, 355 SemMF 315 service-oriented architecture (SOA) 234 SG 350 shared values 43 sharing capability 21 skill management 279 skill management systems (SMS) 271 skills investment funds (SIF) 12 skills ontology 290 SME 67 SME-type organizations 67 SMS 276 social exchange theory 244 social software 239 soft-information 208 soft-OM 209 soft competencies 148 software agents 247 software technologies 346 sourceTopic 345 SPARQL 352 speciity 225 standard occupational classiication (SOC) 309 state-of-the-art knowledge 19 strategic asset (talent) management 42 strategic management 84 subClassOf 340 subPropertyOf 340 support methodology 327 surmise relation 94
SW 121 Swedish National Labour Market Administration (AMS) 306 synthesis 101 synthesis capability 21 system architecture 323 systematic development 23
T
TCL 350 TeamComposer 285 technical 6 TEL 121 TEN-Competence project 234 test components 247 theory of practice 23 three-tier architecture 164 TM (time control) 52 topic 343 topic maps (TM 157 trainability 225 training memory 157 transactional domain 43 transform the business 47 transmission control protocol (TCP) 192
U
units of learning (UoLs) 240 uniform resource identiiers (URIs) 302 unionOf 340 units of learning 236 usable learning objects 288 user interface 160 utilising capability 21
V
valuable knowledge 171 value 46 venture 47
W
WAP Server 67 Web-based learning 6, 15 Web-based tools 50 Web browser 67
374 Index
Web site 301 Wikipedia 230 Windows 32 66 work-integrated CM 90 workforce planning 48 worldwide HRMS 64 World Wide Web Consortium (W3C) 301
X
XML 8 XML binding 174 XML schema 176, 199
Looking for a way to make information science and technology research easy? Idea Group Inc. Electronic Resources are designed to keep your institution up-to-date on the latest information science technology trends and research.
Information Technology Research at the Click of a Mouse!

InfoSci-Online Instant access to thousands of information technology book chapters, journal articles, teaching cases, and conference proceedings Multiple search functions Full-text entries and complete citation information Upgrade to InfoSci-Online Premium and add thousands of authoritative entries from Idea Group References handbooks of research and encyclopedias! IGI Full-Text Online Journal Collection Instant access to thousands of scholarly journal articles Full-text entries and complete citation information IGI Teaching Case Collection Instant access to hundreds of comprehensive teaching cases Password-protected access to case instructor files IGI E-Access Online, full-text access to IGI individual journals, encyclopedias, or handbooks of research Additional E-Resources E-Books Individual Electronic Journal Articles Individual Electronic Teaching Cases
IGI Electronic Resources have flexible pricing to help meet the needs of any institution.
www.igi-online.com
Sign Up for a Free Trial of IGI Databases!
IGI Teaching Case Collection

The new IGI Teaching Case Collection is a full-text database containing hundreds of teaching cases related to the fields of information science, technology, and management.
g ucin d o r Int
Key Features Project background

information
Searches by keywords and

categories Abstracts and citation information Full-text copies available for each case All cases are available in PDF format with instructor files Cases are written by IT educators, researchers, and professionals worldwide
View each case in full-text, PDF form. Hundreds of cases provide a real-world edge in information technology classes or research!
The Benefits of the IGI Teaching Case Collection Frequent updates as new cases are available Instant access to all full-text articles saves research time No longer necessary to purchase individual cases Password-protected case instructor files included in the database
For More Information Visit
www.igi-online.com www.igi-online.com
Recommend to your librarian today!
A Product Of

Competences in Organizational E-Learning

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Competences in Organizational E-Learning

Uploaded by

Copyright:

Available Formats

i

Com pe t e nc ie s in Orga niza t iona l E-Le a r ning: Conc e pt s a nd Tools

I nfor m at ion Sc ie nc e Publishing

Com pe t e nc ie s in Orga niza t iona l E-Le a r ning: Conc e pt s a nd Tools

Overall View of the Book

Section I: General Concepts and Management Issues

Competencies and Organizational Learning 1

Competencies and Organizational Learning:

2 Sicilia and Naeve

Introduction and Background

Competencies and Organizational Learning 3

4 Sicilia and Naeve

Main Components of an IT-Based Approach to Competency Management

Figure 1. Main generic components for competency-based approaches

Competency gap analysis

competency gap Learning objects/ learning activity repository

Learning activity targeter

Organizational competency database

Organizational planning database

Organizational learning schedule

Competency defintion schema Human resources information database

Competencies and Organizational Learning 5

6 Sicilia and Naeve

Bridging Competencies and Modern E-Learning

Activity Orientation in E-Learning

Competencies and Organizational Learning 7

Semantic Web Technologies and Competencies

8 Sicilia and Naeve

Competencies and Organizational Learning 9

Competency Management Systems and Technologies

Competency Management Systems and Technologies 11

Competency Management Systems and Technologies 13

Managerial Attitudes toward Learning and Change

The Process of Learning

Instilling Learning as an Agent of Change

Competency Management Systems and Technologies 15

Learning as a Continuous Process

Competency Management Systems and Technologies 17

Figure 1. Modiied Kolb learning cycle for organizational learning

The Role of Higher Education Institutes

Governing Variables And Concepts

Action and Strategy

Consequences And Outcomes

Double loop Learning

Competency Management Systems and Technologies 19

What is the Role and Function of Higher Education Institutions in Learning?

Managerial Inluence in Learning

Competency Management Systems and Technologies 21

Individual and Organizational Learning

Knowledge as a Distinct Capability

Competency Management Systems and Technologies 23

Forming Effective Learning Strategies

Competency Management Systems and Technologies 25

Individual vs. Group Learning

Life-Long and Situated Learning

Competency Management Systems and Technologies 27

Industry Response to Learning

Competency Management Systems and Technologies 29

Figure 3. Components of learning process

Knowledge Sharing Behaviour

Experiential Knowledge And Action Other forms of Knowledge

Competency Management Systems and Technologies 31

Technologies for Types of Learning

Competency Management Systems and Technologies 33