MACRO LEVEL FACTORS AFFECTING THE CONSTRUCTION

PROCUREMENT SELECTION: A MULTI CRITERIA MODEL

Shiyamini Ratnasabapathy1, Raufdeen Rameezdeen2, Inoka Gamage3
shiyalk@yahoo.co.uk1, rameez@mail.ac.lk2, inokagamage@yahoo.com3
Department of Building Economics, University of Moratuwa, Sri Lanka

ABSTRACT
Construction project procurement systems practiced in the industry have been
subjected to changes resulting in many newly developed procurement systems that
could be used to meet contemporary requirements of the clients. In dealing which
procurement system to apply, there is a need to take into consideration various factors
before any practical decisions can be made. Because, the wrong selection of
construction procurement approach usually leads to project failure or general clients
dissatisfaction. Therefore, a systematic approach for the selection of the most
appropriate system is essential to aid the clients to achieve their ultimate project goals,
thus to ensure best value for their money. The results of the panel data analysis carried
out among major construction organizations in Sri Lanka clearly highlight the
dominance of Measure and Pay method throughout the period concerned. Government
as a major client and the regulator neglected the development of alternative
procurement methods. This paper presents a multi criteria construction procurement
selection model developed for Sri Lankan construction industry. Four rounds of
Delphi survey were conducted to investigate the most important factors at macro level
and their level of influence on various construction procurement systems. Based on
the Delphi survey results, a multi criteria selection model was developed. In addition,
few interviews were conducted with selected industry experts in the view of
interpreting the suitability of the model for use by clients and their consultants. The
results indicate that procurement selection procedure could be improved by
developing a structured procedure based on a set of relevant selection criteria. Clients
should establish a set of appropriate selection criteria based on their ultimate
requirements & distinctive characteristics, project characteristics and external
environment.
Keywords: Construction industry, Procurement selection, Selection criteria, Sri
Lanka

1. INTRODUCTION
Construction procurement system is the organizational framework adopted by the
client for the management of design and construction of a project. With the
introduction of new and innovative ways of construction, many new and alternative
procurement systems have been developed over the past decades. These systems could
be used to meet contemporary requirements of the clients. Different procurement
systems are used for different projects and the correct choice may help to avoid
problems and be the key to the attainment of project specific goals. Procurement
selection therefore received much attention from researchers in recent past. They
strived to develop a systematic approach for procurement selection (NEDO 1985;
581

Skitmore and Mardson, 1988; Masterman, 1992; Chan et al., 2001; Cheung et al.,
2001; Luu et al., 2003).
From the clients point of view, there are probably only three basic criteria which
include cost, time and quality. Hence, traditional procurement selection processes
result in clients prioritising the basic criteria of Time, Cost, and Quality (Seeley,
1997). The lack of structured procedures, based on good information, for the selection
of construction procurement system sometimes inhabit the opportunity for client to
choose a procurement option in a fully informed manner (Luu et al., 2003). Despite
the difficulties associated with selecting a procurement system, it has been suggested
that the main influencing factors in procurement selection are determined by the level
of clients experience (Masterman and Gamesn, 1994). Further, the selection criteria
for a project procurement system may vary with the type of clients as experience
dictates that different types of client bodies have different objectives to be secured
through procurement.
Each project in the industry has its own characteristics and the project to be success
the procurement method must address the technical features of the project alongside
the clients and contractors needs. The selection of suitable procurement method is
critical to the success of any project. Kumaraswamy and Dissanayake (1998) have
concluded that the most appropriate procurement system must necessarily depend on
the project scenario or project profile that can be derived from contextual conditions
such as external factors related to projects. Therefore, it is very much important to
investigate and analyze the influencing factors affecting the selection of a
procurement method in a dynamic construction environment. As far as Sri Lankan
construction industry is concerned, the use of alternative procurement systems are less
compared to other developing countries and therefore, there is a need to explore new
ways of delivering construction projects. This paper presents a multi criteria selection
model developed in the context of Sri Lankan construction industry.

2. PROCUREMENT SELECTION
Several previous studies have identified number of factors influencing the selection of
procurement system in construction. They have focussed on selection factors in terms
of Client characteristics (Moshini,1993; Masderman and Gameson,1994;
Molenaar,1999), Project requirements (Gordon ,1994; Ambrose and Tucker ,1999;
Rowlinson,1999) and External environment (Walker ,1989; Hughes,1989; Sheath et
al.,1994; Alhazmi and McCaffer ,2000;Kumaraswami and Dissanayake,2001).
Outcomes of the previous studies related to procurement selection revels that some
important parameters have been neglected at macro level. A list of predominant
procurement selection parameters have been identified by the Luu et al., (2003) under
the categories of client characteristics and objectives, Project characteristics and
external environment. Clients requirements will ultimately be influenced by the
context in which they operate and this implies that the selection of procurement
system may also be governed by the predominant environment. The external
environment is a structure of several other systems such as economics, politics,
finance, legal, technology, etc. The selection process is an open system which receives
information from its environment, transforms and returns as an output to the
environment (Mcdomott and Rowlinson, 1999). From the synthesis of the outcome of
582

past studies, overall factors affecting the procurement selection can be grouped under
three main criteria. The Figure 1 portrays the conceptual framework developed based
on the projects environments which determine the selection criteria.
Time
Related

Quality
Related

Cost
Related
Clients
Requirements
Clients
Characteristics

Internal
Environment

General

CONSTRUCTION
PROCUREMENT
SELECTION

External
Environment

Project
Characteristics

Figure 1 Conceptual framework for the determination of selection criteria.

There have been several models developed with the intention of facilitating the
selection of suitable procurement system for a particular project. Among the various
models, Multi Attribute Utility Technique (MAUT) received the greatest attention.
Chang and Ive (2002) discussed some of the inherent problems of using MAUT for
procurement selection. One of strongest criticisms was the selection of procurement
variables. The other is the utility value developed through opinions of industry
experts. Particularly they are critical about the subjective nature of assigning values to
procurement selection parameters to obtain mean utility values. Chan et al. (2001)
used Delphi Method and Cheung et al. (2001) used Analytical Hierarchy Process to
reduce the subjectivity of arriving at the utility values. Luu et al. (2003) on the other
hand used 34 parameters against 8-10 used by other researchers to develop an
acceptable model. However, these models use opinion to arrive at the most suitable
procurement method for a project. Another pertinent question is that, in reality does
the client or clients consultant use a structured model for procurement selection?
According to Masterman (1992) the practice of procurement selection is rather
unstructured and ad hock. This observation is of course very true for a developing
country like Sri Lanka (Shiyamini et al, 2005). In this context, this study examines the
overall factors affecting the selection of procurement system in Sri Lanka. It differs
from most of the related research on procurement selection in two respects. First, the
study focuses on the past trend of procurement use in Sri Lanka. The factors that led
to the use of a particular set of procurement methods and its change over time gave an
ex-post picture of the parameters affecting procurement selection. One of the features
of this paper is the use of this empirical data to explore procurement selection criteria.
Second, most studies have employed a large number of variables in procurement
selection. However, in reality whether such a large number of variables considered in
procurement selection is doubtful.

583

3. METHODOLOGY
The use of Multi Attribute Decision Analysis has been considered the foremost
technique for examining client needs and weighting of preferences from experts for
each procurement (Chan et al., 2001), but the major difficulty with these selection
systems lies in the lack of consensus among the experts on the utility factors. To
overcome above deficiencies, the Delphi technique was adopted. Delphi method is a
highly formalized method of communication that is designed to extract the maximum
amount of unbiased information from a panel of experts. Chan et al. (2000) has
proved that the Delphi method is a powerful and appropriate technique for deriving
objective opinions in a rather subjective area such as the Multi Attribute Methodology
for selection of procurement system. Therefore, Delphi technique has been used to
derive the expert opinion in this study. The success of the Delphi method principally
depends on the careful selection of the panel of experts. In order to meet all stipulated
requirements and to increase efficiency of the outcomes, the sample size was selected
to have upper limit of the reliable sample size (15- 35) of the Delphi technique. Thirty
five (35) members of panel represent wide distribution of professionals from several
disciplines including Project Management, Quantity Surveying, and Engineering from
consultant & contractor organizations both in public and private sectors. The
Purposive selective sampling has been used as the appropriate sampling method for
this research, because the information obtained from experts required in depth
knowledge and sound experience on various procurement options. A group of experts,
who have relevant working experience in the field of construction management and
procurement selection, were selected to get the opinion on selection criteria and their
level of influence on various procurement systems in construction. The following
Table 1 gives the profile of the experts involved in the survey.
Table 1: Profile of the experts
Category

No

Project Director
Project Manager/Manager Contracts
Project Engineer/Civil Engineer
Chief/Senior Quantity surveyor
Quantity Surveyor
Total

8
6
2
5
14
35

The Table 2 presents the formats of Delphi survey carried out in four rounds and
briefly introduces the data evaluation techniques used to analyse the data collected
from the survey.

584

Table 2 : Summary of Delphi survey and Data analysis

Round 1

Round 2

Round 3

Round 4

Instrument

Questionnaire 1
Literature review
Two weeks

Questionnaire 3
Results from round
two subject to
factor analysis
Four weeks

Questionnaire 4

Data base for

Questionnaire

Questionnaire 2
Results from
round one and
literature
Eight weeks

35

35

35

35

35

31

26

23

Utility values for

each
selection
factor
obtained
from round two
against
various
procurement
systems
Calculation
of
average
utility
values,
concordance
coefficient (w) and
related level of
significance

Re-assessed
utility values for
each
selection
factor
against
various
procurement
systems
Calculation
of
average
utility
values,
concordance
coefficient (w),
percentage
improvements of
utility values and
level
of
significance.

Duration
Number of
experts
selected
Number of
experts
responded
Findings

Data Analysis

Identification of
factors affecting
the selection of
procurement
system in terms of
main three criteria

Level of
importance of
each factor
identified in
round one
Calculation
of
Percentage,
weighted mean,
standard
deviation,
Severity index,
Coefficient
of
variation
and
factor analysis for
identified
significant factors

Results from
round three
Five weeks

4. MAIN FINDINGS AND DISCUSSION

This section discusses the results of four rounds of Delphi survey which targeted to
establish a set of exclusive selection criteria based on Sri Lankan context. Factors
were studied and analysed in terms of client requirements & characteristics, project
characteristics and external environment. Clients requirements were focussed on cost,
time, quality and other general aspects which affect the procurement system selection.
The following Table 3 summarizes the overall factors derived from the review of
literature and Delphi round one.
In the second round of Delphi, level of importance of each factor was considered. All
together, 25 clients requirements out of 34, 11 project characteristics out of 13 and 14
external environmental factors out of 19 were identified as significant factors. As
some of these significant factors are interrelated, they were consolidated using factor
analysis. The results of factor analysis brought nine factor categories of clients
requirements, six factor categories of project characteristics and five categories
external environmental factors. Third round targeted to derive the utility values for
each factor extracted from factor analysis against a wide range of various procurement
systems used in the industry.

585

Table 3: Macro level factors affecting the selection of procurement system

SELECTION CRITERIA
CLIENTS REQUIREMENTS
Cost Related Factors
Capital cost
Maintenance cost
Prequalification and tendering cost
Financial risk
Price competition
Completion within the budget
Time Related Factors
Planning and designing time
Tendering and evaluation time
Construction time
The early start of the project
Speed of construction
Time overruns
Quick response to clients new
requirements (extra work and
variations)
Maximizing of activities interfacing
Stage completion
Quality Related Factors
Design reliability
Aesthetic appearance of the building
Workmanship
Functionality (suitability for the
intended users)
Design innovation
General Needs
Allocation of responsibility
Professional team performance
Parties involvements
Accountability
Transparency
Safety requirements
Corporation and motivation
Existing building operation
Familiarity (Clients awareness of
construction procurement system)
Tender evaluation criteria
Clear express of end users
requirements
Flexibility
Consultants attitude towards clients
Types of client

PROJECT CHARACTERISTICS

Project type
Project size
Project cost
Degree of flexibility
Degree of complexity
Time constrains
Payment method
Integration of design and construction
Project funding method
Project site location
Site risk factors
Construction method
Degree of innovative technology
involvement

EXTERNAL ENVIRONMENT

586

Markets competitiveness
Government as a policy maker
Government as a major client
Regulatory feasibility
Technological feasibility
Source of finance for the project
Experienced contractor availability
Education of builders
Economic condition of the country
Availability of material
Information Technology
Influence of Intuitional bodies
Natural disaster
Industrial actions
Socio cultural differences
Goodwill of the contractor
Environmental issues
Civil war condition
Objections from neighbours / public

The questionnaire used in fourth round comprised with the average utility values
derived from 25 experts for each factors against various procurement systems. The
respondents were asked to reconsider the results of round three in order to improve the
consistency of the results. There were 23 out of 25 questionnaires received in the
fourth round and they were considered for the analysis at the end of fourth round. The
consistency of the experts utility values was subject to Kendall coefficient of
concordance analysis using SPSS package. The results were summarized and
compared separately for clients requirements, project characteristics and external
environmental factors. The following section discuses the outcomes of the Kendall
coefficient of concordance analysis separately for each selection criteria.
Clients Requirements
Utility values of all nine factors were sufficiently consistent at 0.05 level of
significance or smaller. Further, concordance coefficient (w) ranges between 0.08 0.46 for certain factors such as Risk management, Time availability & Predictability,
Accountability, Quality of works, and Responsibility & Parties involvement, indicate
less strong agreement among experts during Delphi round three. The concordance
coefficient ranges between 0.59 - 0.74 for the factors such as Price certainty, Price
competition, Flexibility for changes and Familiarity indicate the strong agreement
among experts at the fourth round of Delphi. The concordance analysis shows (Refer
Table 4) that the consistency of the experts ranking for procurement systems against
each factor has improved over the succeeding round. The coefficient of concordance
for all the factors (except for Time availability and Predictability) has improved from
28% to 174.84%. In this round, nine factors were considered sufficiently consistent
compared to third round. The results show that the experts had overcome the difficulty
in assessing the factors such as Responsibility & Parties involvement, since the
significance improved lesser than 0.05.( reduced from 0.7075 to 0.0399 in the
successive rounds). Therefore, it can be concluded that all nine factors significantly
influence the procurement selection practices in Sri Lankan context.
Table 4: Comparison of Concordance Coefficient of the utility values Clients
Requirements

Selection Factors
Clients Requirements
1
2
3
4
5
6
7
8
9

Risk management
Time availability and
Predictability
Price Certainty
Price Competition
Accountability
Flexibility for Changes
Quality of Work
Responsibility and
Parties Involvement
Familiarity

Concordance of Coefficient (W)

Significance
Level
Round Round
03
04
0.0000 0.0000

Round
03
0.1440

Round
04
0.1855

%
Improvement
28.82%

0.2548

0.2548

0.00%

0.0000

0.0000

0.3068
0.2812
0.2188
0.3271
0.1464

0.6229
0.5931
0.4661
0.6508
0.4024

103.02%
110.93%
113.04%
98.96%
174.84%

0.0000
0.0000
0.0000
0.0000
0.0000

0.0000
0.0000
0.0000
0.0000
0.0000

0.0293

0.0807

175.17%

0.7075

0.0399

0.4789

0.7397

54.45%

0.0000

0.0000

587

Project Characteristics
Utility values of all six factors were sufficiently consistent at 0.05 level of significance
or smaller. Further, concordance coefficient (w) ranges between 0.29 - 0.38 for all six
factors indicate less strong agreement among experts during Delphi round three.The
concordance analysis shows (Refer Table 5) that the consistency of the experts
ranking for procurement systems against each factor has improved over the successive
round. The coefficient of concordance for all six factors improved from 154.64% to
581.644%. In this round, six factors were considered sufficiently consistent compared
to previous round. The results shows that the experts had overcome the difficulty in
assessing the factor such as Project type, since the significance improved lesser than
0.05 ( reduced from 0.2613 to 0.0000 in the successive rounds).
Table 5: Comparison of Concordance Coefficient of the utility values - Project
characteristics
Selection Factors
Project Characteristics
1
2
3
4
5

Project Cost and

Funding method
Project Complexity
Project Type
Time Constrains
Degree of Flexibility
Payment Modality

Concordance of Coefficient (W)

Round Round
%
03
04
Improvement

Significance Level
Round
Round
03
04

0.1527

0.3827

150.64%

0.0000

0.0000

0.0952
0.0491
0.0814
0.1380
0.1025

0.3092
0.3346
0.3382
0.3558
0.2978

224.90%
581.05%
315.30%
157.94%
190.70%

0.0061
0.2613
0.0215
0.0001
0.0030

0.0000
0.0000
0.0000
0.0000
0.0000

External Environment
With respect to the external environment, utility values of all five factors were
sufficiently consistent at 0.05 level of significance or smaller. The socio cultural
suitability shows the higher level of significance as the concordance coefficient has
improved at round four. Economic condition and the fiscal policy have also subject to
changes from 0.041 to the 0.012 which indicates the increased level of significance.
The significance level of Technology, Regulatory environment and Market condition
for the project were increased to the maximum from the 0.011 and 0.001 respectively.
On the other hand, the correlation between the experts regarding the external
environmental factors has also been increased. This indicates that the respondents got
closer to the opinion of the factors regarding the procurement selection. Significant
change occurred for the socio cultural suitability from 0.049 to 0.135. The following
change has taken place for technology from 0.089 to 0.172 followed by regulatory
environment, market condition and the economic condition & the fiscal policy. These
external environmental factors have either direct or indirect influence on the selection
and use of the procurement systems. Therefore, in assessing the suitability of a
procurement system, the underplaying relationships of external environment need to
be considered to avoid the consequences in decision making.

588

Table 6: Comparison of Concordance Coefficient of the utility values External

Environment
Selection Factors
External Environment
1 Market competition
Economic condition and
2
the fiscal policy
3 Technology
4 Socio cultural suitability
5 Regulatory environment

Concordance of Coefficient (W)

Round Round
%
03
04
Improvement
0.105
0.160
52.38%

Significance Level
Round
Round
03
04
0.001
0.000

0.074

0.096

29.73%

0.041

0.012

0.089
0.049
0.089

0.172
0.135
0.147

93.26%
175.51%
65.17%

0.110
0.258
0.011

0.000
0.021
0.000

5. CONCLUSIONS
An exclusive set of multiple decisive factors in terms of main three criteria: Clients
requirements & objectives, Project characteristics and External Environment, have
been identified as being generally adequate for the procurement selection and there is
a reasonable consensus on utility values for each procurement system. This study has
used Delphi technique to ensure that the consensus is reached for the utility values
provided by the panel of experts/practitioners from the industry. The synthesis of the
outcome of the survey demonstrates intensely that there is a significant improvement
in the consistency of the utility values over the successive Delphi rounds.
The combination of the results of all four rounds revealed 20 significant factors. As
some factors are interrelated, factor analysis was performed to consolidate the related
factors from each main criterion. All together 9 factor categories of clients
requirement, 6 factor categories of project characteristics and 5 categories of external
environmental factors were derived. Among the nine factors of clients requirements;
the price certainty, price competition, flexibility for changes, familiarity,
accountability and quality of work have higher concordance of coefficient and
sufficient level of utility values leading to significant influence on all procurement
systems. Other three factors also influence the procurement selection; since those
factors have satisfactory utility values even though have lesser concordance of
coefficient. All these factors should provide a starting point for the selection process.
Six factors of project characteristics have satisfactory utility level even though they
have lesser concordance of coefficient. From this study it was found that all identified
factors in terms of clients requirements and project characteristics have greater level
of influence on procurement selection. The importance and interrelationships of
identified significant factors and the utility values provide a good background for the
development of procurement selection criteria for any kind of project scenario. In
addition to the clients requirements and project characteristics, factors from external
environment also have the significance influence on selection process. Selection of
appropriate procurement system is not a simple task. But the procurement selection
procedure could be improved by developing a structured procedure based on a set of
relevant selection criteria. Clients should establish a set of appropriate selection
criteria based on their ultimate requirements & distinctive characteristics, project
characteristics and external environment. The selection criteria should be logically
derived from projects internal and external environment.
589

6. REFERENCES
Alhazmi, T. and Mccaffer, R., 2000, Project procurement systems selection model.
Journal of Construction Engineering and Management, 126(3), 176 - 183.
Ambrose, M.D. and Tucker, S.N., 1999, Matching a procurement system to client and
project needs: a procurement system evaluator, Proceedings: Customer
Satisfaction: A Focus for Research and Practice in Construction, P.A. Bowen and
R.D. Hindle (eds), Cape Town,South Africa, University of Cape Town, 280-288.
Chan, A.P.C., Yung, E.H.K., Lam, P.T.I., Tam, C.M., and Cheung, S.O., 2001,
Application of Delphi method in selection of procurement systems for construction
projects, Construction Management and Economics, 19, pp699718.
Chang, C.Y. and Ive, G., 2002, Rethinking the Multi Attribute Utility Approach based
procurement route selection technique, Construction Management and Economics,
20, pp275-84.
Cheung, S., Lam, T., Leung, M. and Wan, Y., 2001, An analytical hierarchy process
based procurement selection method, Construction Management and
Economics,19, pp427437.
Gordon, C.M., 1994, Choosing appropriate construction contracting method, Journal
of Construction Engineering and Management, ASCE, 120 (1), 196-210.
Hughes, W., 1989, Identifying the environments of construction projects,
Construction Management and Economics, 7, 29-40.
Kumaraswamy, M.M., and Dissanayake, S.M., 1998, Linking procurement system to
project priorities, Building Research and Information, 26 (4), 223-238.
Kumaraswamy, M. and Dissanayaka, S., 2001, Developing a decision support system
for building project procurement, Building and Environment, 36 (3), 337-349.
Luu, S.D.T., Ng, T., and Chen, S.E., 2003, Parameters governing the selection of
procurement system, Engineering, Construction and Architectural Management, 10
(3), pp209-218.
Masterman, J. W. E., 1992, An Introduction to Building Procurement Systems, E &
FN Spon, London.
Masterman, J.W.E. and Gameson, R. 1994, Client characteristics and needs in relation
to, their selection of procurement systems, Proceedings: East Meets West
Procurement
Systems Symposium, CIB Publication 175, 4-7 December, Hong Kong, S. Rowlinson
(ed), 79-87.
Mcdermott, P. and Rowlinson, S., 1999. Procurement systems. A guide to best
practice in construction. London: E & FN Spon.
Molenaar, K.R., 1999, Selecting appropriate projects for design-build procurement in
Ogunalana, S.O. (Ed.), Proceedings of Profitable partnering in construction
procurement joint symposium of CIB W92 (procurement system) and CIB TG 23
(Culture in construction) , London : E & FN Spon, pp 349-60.
Moshini, R.A., 1993, Knowledge-based design of project procurement process,
Journal of Computing in Civil Engineering, ASCE, 7(1), 107-122.
NEDO, (1985), Thinking about building, Report by Building, Design Partnership for
National Economic Development Office, Building EDC, HMSO.
Rowlinson, S., 1999, Definition of Procurement Systems, In Rowlinson, S. and
McDermott, S. (Ed.,), Procurement Systems: A Guide to Best Practice in
Construction, CIB Documentation, E & FN Spon., pp27-53.
Seeley, I. H., 1997. Quantity surveying practice. 2nd ed. London: Macmillan press
Ltd.
590

Sheath, D.M., Jaggar, D. and Hibberd, P., 1994, Construction Procurement Criteria: a
multi-national study of major influencing factors, in Rowlinson, S. (Ed.), CIB W92
Proceedings: East Meets West, Procurement Systems Symposium, 4-7
December, Hong Kong, pp.361-70.
Shiyamini, R., Rameezdeen, R., and Amaratunga, D., (2005), Macro Analysis of
Construction Procurement Trend in Sri Lanka, 5th International Postgraduate
Research Conference of the Research Institute for the Built and Human
Environment, The University of Salford, UK.
Skitmore, R.M. and Marsden, D.E., 1988, Which procurement system? Towards a
universal procurement selection technique, Construction Management and
Economics, 6, E & F.N. Spon Ltd, London, pp71-89.
Walker, A., 1989, Project Management in Construction, 2nd Ed, Oxford: BSP
Professionals Books.

591