Professional Documents
Culture Documents
forMITBuildingProjects
by
PereAndreuUbachdeFuentes
CivilEngineer
PolytechnicUniversityofCatalonia,2001
SUBMITTEDTOTHEDEPARTMENTOF
CIVILANDENVIRONMENTALENGINEERINGINPARTIAL
FULFILLMENTOFTHEREQUIREMENTSFORTHEDEGREEOF
MASTEROFSCIENCEIN
CIVILANDENVIRONMENTALENGINEERINGATTHE
MASSACHUSETTSINSTITUTEOFTECHNOLOGY
JUNE2004
2004PereAndreuUbachdeFuentes.Allrightsreserved.
TheauthorherebygrantsMITpermissiontoreproduce
andtodistributepubliclypaperandelectroniccopies
ofthisthesisdocumentinwholeorinpart.
SignatureofAuthor:
DepartmentofCivilandEnvironmentalEngineering
May7,2004
Certifiedby:
FredMoavenzadeh
ProfessorofCivilandEnvironmentalEngineering
ThesisSupervisor
Acceptedby:
HeidiNepf
AssociateProfessorofCivilandEnvironmentalEngineering
Chairman,DepartmentalCommitteeonGraduateStudents
ValidationoftheProjectDefinitionRatingIndex(PDRI)for
MITBuildingProjects
by
PereAndreuUbachdeFuentes
SubmittedtotheDepartmentof
CivilandEnvironmentalEngineeringonMay7,2004
inPartialFulfillmentoftheRequirementsfortheDegreeof
MasterofScienceinCivilandEnvironmentalEngineering
ABSTRACT
The thesis presents the work performed to validate the managerial tool called
Project Definition Rating Index (PDRI). An improved methodology is presented.
This new methodology produces much better linear correlation results between
PDRI scores and Cost performance of the projects (R2 = 0.957) than previous
validationefforts.
The projects used for validation purposes have been developed by MIT for its
CapitalProjecteffortthatstartedofficiallyintheyear2000.Whiletheresultsare
recommendedtobeapplied tofutureMITprojects, theauthorexplains howthe
PDRIworksthusfacilitatingfuturevalidationtasks.
Intheprocess,theauthorproposesamodifiedversionoftheweightsofthePDRI
scoringsheetthatbetteradjuststothedefinitionprinciplesofthetool.
ThesisSupervisor:FredMoavenzadeh
Title:ProfessorofCivilandEnvironmentalEngineering
ACKNOWLEDGMENTS
Iwouldliketothankmyadvisorforthehighdegreeofconfidencehehas
putinmywork.ThishasgivenmethefreedomtocreateaworkinthedirectionI
originallylookedforwhenIcametoMIT.
I want to express my infinite gratitude to all the people who have
encouraged me and participated actively in the development of my research.
Particularlyincludingtheprojectmanagersanddirectorsofthebuildingsstudied
for the research: Martha Ondras, Hugh Colasacco, Milan Pavlinic, John Hawes,
MichaelOwu,JonathanHimmel,andNancyJoyce.AlsoIhaveaspecialregardto
Mr.JosephGifunwhohassupportedmyresearcheffortsunconditionallyandhas
givenmetheresourcestocompleteitsuccessfully.
IamalsogratefultomypeerandfriendIsraelRuizwhohasbeenalways
availableforinformationalinterviews,andhasdispelledmanydoubts.
MyroommatePanayiotisLemonidisalsodeservesaspecialmentionforhis
spontaneoushelpproofreadingthispalimpsest.
Catalina,thankyouformakingmefeellikethewholeresearchefforthas
been effortless. Not being that enough, I have enjoyed with you the marvelous
experienceofexploringtheprimrosepathofloveandatthesametimediscovering
newgroundsofknowledge.WithyouIamamuchbetterperson.
IfinallywanttowritedownintopaperhowmuchIlovemybrothersand
parents,whodespitethedistancehavebeenabletoshowmetheirsupportdayafter
dayandkeptupmymoraleevenintheworstmoments.
Thisworkhasbeenpossiblethankstothefundingof"laCaixa"Foundation
thathassponsoredmystudiesattheMassachusettsInstituteofTechnology.
Contents
1INTRODUCTION......................................................................................................15
1.1Justification........................................................................................................16
1.2ProjectScopeDefinition.....................................................................................17
1.3ThesisObjectives................................................................................................18
1.4OrganizationOfTheThesis...............................................................................19
2BACKGROUND.........................................................................................................21
2.1TheConstructionIndustryInstitute.................................................................21
2.1.1ProjectDefinitionRatingIndex..................................................................23
2.1.1.1TheFirstTool.......................................................................................23
2.1.1.2PDRIForBuildingProjects:TheBirthOfASaga.............................25
2.1.2TheDevelopmentOfThePDRI..................................................................25
2.1.2.1DefinitionProcess................................................................................25
2.1.2.2WeightingProcess................................................................................26
2.1.2.3ValidationProcess................................................................................27
2.2TheMassachusettsInstituteOfTechnology.....................................................29
2.2.1TheMITCapitalProject.............................................................................29
2.2.2TheMITDepartmentOfFacilities.............................................................30
2.2.3TheMediaLaboratoryExtensionProject..................................................31
3METHODOLOGY......................................................................................................33
3.1SampleOfProjects.............................................................................................33
3.1.1SimmonsHall..............................................................................................35
3.1.270PacificStreetDormitory........................................................................37
3.1.3BuildingE19Renovation...........................................................................38
3.1.4DreyfusChemistryBuilding.......................................................................40
3.1.5224AlbanyStreetGraduateDormitory.....................................................40
3.1.6ZesigerSportsAndFitnessCenter.............................................................41
3.1.7StataCenter................................................................................................42
3.2TheQuestionnaire..............................................................................................43
3.2.1VariablesUsed............................................................................................45
3.3Interviews...........................................................................................................48
3.4StatisticalAnalysis.............................................................................................49
3.5ProjectRatingInformation................................................................................51
3.6Summary.............................................................................................................53
4RESULTSANDANALYSIS.....................................................................................55
4.1ResultsOverview................................................................................................55
4.2BoxPlots.............................................................................................................56
4.3ScatterPlotsAndCorrelationFactors..............................................................58
4.4RegressionAnalysis............................................................................................61
4.5CriticalAnalysisOfResultsAndComparison..................................................63
5CONCLUSIONS........................................................................................................67
5.1ReviewOfObjectives..........................................................................................67
5.2Recommendations...............................................................................................69
6APPENDIXA.............................................................................................................71
7APPENDIXB.............................................................................................................75
8REFERENCES...........................................................................................................93
ListOfFigures
Figure3.1ViewofthefacadeofSimmonsHall..........................................................36
Figure3.2InteriorviewoftheSimmonsHallmainlobby.........................................37
Figure3.3Viewofthemainentranceof70Pacificstreet..........................................38
Figure3.4TheDreyfusChemistryBuildingislocatedattheheartofthecampus..39
Figure3.5224Albanystreetisalsoknownas'TheWarehouse"...............................41
Figure3.6InteriorviewoftheZesigerCenterfacilities.............................................42
Figure3.7TwoviewsoftheStataCenterfromVassarStreet...................................43
Figure3.8Snapshotofablankspreadsheetdevelopedtoweighttheelementsofthe
PDRI(Tab1).............................................................................................................52
Figure4.1Boxplotsofthe6differentsuccessvariablesused...................................56
Figure4.2Boxplotsofthe6differentsuccessvariablesusedafterremovingthe
ZesigerSportsandFitnessCenterproject..............................................................58
Figure4.3Scatterplotsshowingthecorrelationbetweenthedifferentperformance
variablesandthePDRIindexes..............................................................................59
Figure4.4LinearregressionoftherelationshipbetweenCostandPDRI(R2=0.957).
...................................................................................................................................62
Figure4.5QuadraticregressionoftherelationshipbetweenCostandPDRI..........63
ListOfTables
Table3.1Variablesusedformeasuringprojectperformance....................................47
Table 4.1 Values of the performance variables used and PDRI scores for every
project............................................................................................................................56
Table6.1WeightedListforsectionIofthePDRI......................................................72
Table6.2WeightedListforSectionIIofthePDRI....................................................73
Table6.3WeightedListforSectionIIIofthePDRI...................................................74
ListOfEquations
Equation4.1LinearregressionoftheCostPDRIcorrelation....................................61
Equation4.2QuadraticregressionoftheCostPDRIcorrelation..............................61
1 INTRODUCTION
InthefieldofConstructionManagement,researchhastakenmanyforms
and directions. Sometimes authors conclude their works with best practice
guidelinesandsometimestheydevelopmanagerialtools.Inallcasesthesuccessof
theirworkdependsontheadoptionoftheseguidelinesandmanagerialtoolsbythe
constructionindustry,whichnotalwayshappens.
Managerialtoolsfortheconstructionindustryhavebeendevelopedmany
timesforproprietarypurposesbytheconstructioncompanies.Thereasonbehindit
is that most of them use sensible information based on the experience gained
throughtheexerciseoftheirbusiness.Thefactthatprivatecompaniesarereluctant
toreleasedatabasedontheirperformancemakesitdifficultforuniversitiestolead
research to develop tools based on expertise. However there are times in which
universityresearchersareabletodeveloptheirsoughtmanagerialtools.Buteven
researchleadersfacetheproblemofcollectingenoughdatafortheirstudies,added
withthedifficultiesfacedwhileretrievingdatafromactualworksitesandprojects.
Constructioncompaniesconfirmtheaccuracyoffindingsandestimatesratherthan
givingtheirowndataforacademicuse(Slaughter1999).
INTRODUCTION
Whenthepurposeoftheresearchistogatherempiricaldataandstudyit,
every effort should be made to treat that data with the maximum care. Other
researchershavealreadystatedhowdifficultitistocollecttheempiricaldata.Good
statistical studiescanyetbemadewithsmall amountsofdata,butspecial care
needstobetaken.
1.1 JUSTIFICATION
In many other fields of science it is common to see the industry leaders
supporting applied research and development done in universities, or even
suggesting the directive lines for such investigation efforts. This yields to large
leapsinthestateofknowledgeofthosefields.Inthequestforappliedresearch,the
enterprises,andtheprivatesectoringeneral,havealottosay.Theyarebehindthe
everydaysuccessorfailureoftheeconomicrelatedactivityoftheirgivendiscipline.
Thus,theyknowinfirsthandwhattheirneedsaretomakebusinessesworkbetter.
They can tell investigators in which direction should the efforts be focused to
improveoperativeresults.InConstructionManagementthisphenomenonseldom
happens,anditistheresearchers'responsibilitytobuildmomentumaroundthese
fewexamplesofindustryinput.
Thisreflexionshouldnotalienate,inanycase,thethoughtsofthosewho
rather believe that one of the main functions of a researcher is to set the
development lines of his/her field and to decide in which direction to push the
borders of knowledge. There are different research goals and different ways to
accomplishthem.
16
JUSTIFICATION
1.2 PROJECTSCOPEDEFINITION
Differentauthorshaveidentifiedthevitalimportanceofperformingwellthe
taskstobedoneduringthedevelopmentoftheproject.Thisshouldbetakeninto
account ever since the start of the preplanning process because it has a great
impact on the outcome of the project. The cause is that since the cost of these
preliminarytasksisreducedcomparedtotheoverallcostoftheproject,andthe
impactontheoutcomeisbig, itpaysofftospendsomemoreontheseplanning
tasksinordertoensureasatisfactorydevelopmentoftheprojectduringsubsequent
phases (Neale and Neale 1989, Cho 2000, Best and de Valence 1999, Pinkerton
2003).Infact,oneofthemainreasonsforwhichprojectshavecostoverrunsisthat
changes occur during construction. There are scope changes requested by the
differentpartsinvolvedintheproject(i.e.endusers,owner,architect,oreventhe
contractor attending to market driven forces). These scope changes reflect the
uncertainty that reigns during the early stages of a project, and the different
objectiveseachpartyhasfortheproject.Behindthesechangesareeitheralackof
definitionoftheitemssubjecttochange,orapoorideaofhowtheprojecthadtobe
handled.Includedinthislastitemwecanfind:theexpecteduseofthebuilding,the
architecturaldesign,managementoftherisksfaced,thedeliverymethodchosen,or
thedocumentationanddeliverablesrequired.
17
INTRODUCTION
Havingwelldefinedguidelinesonhowtodefinebuildingprojectsshouldbe
a common objective within organizations. It is crucial that there is a deep
discussion around all of the aspects about the project undertaken, but this
discussioncould,manytimes,goonandon.
How much is enough? What items should be stressed? The Project
DefinitionRatingIndexisatoolthatprovidesawaytofollowtheplanningprocess
ofabuildingprojectandanswersthesequestionsinaneasyandeffectiveway.
1.3 THESISOBJECTIVES
Thefirstpurposeofthedissertationistodevelopspecificguidelinesthat
complementthemanagerialtoolcalledProjectDefinitionRatingIndex(PDRI)for
BuildingProjects.Morespecifically,theseguidelinesaredirectedfortheuseofthe
MassachusettsInstituteofTechnology(MIT).Thedevelopmentoftheseguidelines
canbeunderstoodasavalidationprocessthatwillmakethePDRIusefulforMIT
as an organization that will undertake more projects in the future. Taking
advantageofoneofthelargestcapitaleffortsevertakenbytheMIT,inwhichits
campusisevolvingandchangingatapacehardtocopewith,thisthesiswantsto
collect,atleastatabasiclevel,theexperiencegainedthroughthiseffort.Itisa
firmpurposeofthisessaytobringnewknowledgeonhowtoperformtheplanning
ofbuildingprojects.
Thesecondgoalofthethesisistoenlargethedatabaseofprojectsusedto
validatethePDRI.Asoftheyear2000,theResearchTeam155oftheConstruction
IndustryInstitutehadusedatotalof33projectstovalidatethePDRIforbuilding
projects.SincethisisasmallsampletorepresenttheUnitedStatesconstruction
18
THESISOBJECTIVES
industryasawhole,Cho(2000)recommendstoincreasethesizeofthesampleto
improvethevalidationprocess.Healsosuggeststhatparticularorganizationscould
benchmark the PDRI and enhance its accuracy. In this sense the objective is
twofold:ononesidethedocumentprovidesnewinputforvalidationpurposestrying
touseasamplerepresentativeofMIT'sbuildingprojects,andontheothersidethe
thesisfollowsCho'srecommendationandusesinternalknowledgeforthebenefitof
theorganization.Thisthesiscontributeswith6newprojectsundertakenbyMIT
anddrawnfromtherecentcapitaleffortitisundertaking,andthatrepresentsmore
than$1billionthatwillbeinvestedin10differentbuildingsforthecampus.This
investmentisplannedtobedoneduringaperiodof10years.
The third purpose of the thesis is to better understand the working
principlesofthePDRI.Thiswillpermitperformingacriticalanalysisofitsfunction
forprojectmanagers'use.Fromabetterunderstandingandtheanalysisofthedata
obtainedtheauthorexpressestheobjectiveofimprovingthePDRImodel.
1.4 ORGANIZATIONOFTHETHESIS
Inthefollowingchaptersofthisdocumenttheauthorexplainsallthesteps
thatguidedthepresentedworktoitscompletion.Inchapter2theauthorintroduces
the background that motivates the research effort and puts the reader into a
positiontounderstandthecontextoftheresearch.Aspecialpartofthischapter
takes care of explaining the previous investigation on which the present one is
based.Thefirstpartofthechapterfocusesdownstartingattheorganizationthat
generated the PDRI for building projects and the roots of the PDRI. It finishes
explaining, at a low level of detail, the development of the PDRI and the
19
INTRODUCTION
correspondingvalidationpreviouslydone.Thesecondpartofthechapteralsoscales
down from the organization for which the PDRI for building projects is being
validated,MIT,totheparticularinvestmenteffortthatmotivatesthisresearch.
Inchapter 3 the authorexplains the methodology used for collecting the
necessarydata.Eachprojectusedforthestatisticalanalysisisdescribed,andthe
chapteralsoincludesadescriptionofoneprojectthat wasnotincludedintothe
analysisandtheargumentsthatsupportedthisdecision.Thereisanoverviewof
thetoolsusedforcollectingthedataandtheprocedureisalsodescribed.Thelast
partofthechapterexplainshowtheempiricaldataismanipulated,accordingtothe
basic principles and assumptions of the PDRI and following standard statistical
procedures.
Chapter4 reviews the results of the recollection of data and proceeds to
analyzethemasdescribedinchapter3.Everystepisexplainedandreasoned.At
theendofthechaptertheauthordiscussesindepththefindingsandprovidesa
theoretical explanation of the empirical findings. All being embedded into the
generaloriginalframeworkofthePDRIforbuildingprojects.
Chapter 5 reviews the accomplishment of the objectives expressed in
chapter 1 and presents the conclusions of the research. In this chapter,
recommendationsforfutureresearchmayalsobefound.
20
2 BACKGROUND
ThischapterdescribestheoriginsoftheProjectDefinitionRatingIndexfor
buildingprojects,whileitalsoprovidesanintroductiontothegenerationofthetool
basedonthethesisofChungSukCho(2000)andstartsalightdiscussionofthe
decisionsmade.
Thesecondpartofthechapterintroducesthenatureofthedatathatwillbe
usedforthestudyofthisthesis:theCapitalProjectoftheMassachusettsInstitute
of Technology. Finally, the chapter introduces the motivations of the author to
conductthisthesisandtheproblemsitintendstosolve.
2.1 THECONSTRUCTIONINDUSTRYINSTITUTE
TheConstructionIndustryInstitute(CII)isanorganizationthatinvolves
membersthatareeithercontractorsorownersintheconstructionbusiness,andhas
partnerships with several American universities. The main goal of the CII is to
develop best practices for the construction industry to improve the business
effectiveness. By having involved representatives of the different sides in the
industry,theyensureparticipation.
BACKGROUND
The Project Definition Rating Index for industrial projects is one of the
differentresearchtopicsthatwaspromotedbytheCIIinordertoimprovethescope
definition level of industrial projects (Dumont, Gibson and Fish, 1997). After
realizing the good reception that the construction industry gave to this newly
developedtool,andacknowledgingtheneedofasimilartoolforbuildingprojects,a
newresearchteamwasformedwiththepurposeofadaptingthetooltobuilding
projects(Cho,FurmanandGibson,1999).Thiskindofprojectswasnotincludedin
theapplicationrangeofthepreviouswork.Thereforeitwasfeltthatanewresearch
team should be formed in order to develop a similar tool focused on building
projects.
Thisnewteamcollectedtheexperiencegainedinthepreviouseffortand
restartedtheprocessofdevelopingdefining, weighting andvalidating another
PDRI.ItissurprisingthateveninsuchaparticularorganizationastheCIIis,in
which more than 40 contractors and 40 owners are involved, only 33 different
projectscouldbeusedforvalidationpurposesofthebuildingversionofthePDRI
(moreoverwhenthisstudywaspromotedbasedonfeedbackfromaverysimilar
research and which was felt very useful). Taking into account that the
organizationsbackingtheCIIincludesomeofthelargestownersandcontractorsin
theUnitedStates,thepotentialdatabasewouldbeeasilyofhundredsofdifferent
projectsforeverysingleyear.Thefactthatonly33projectscouldbeusedshowsthe
high difficulty researchers find in conducting their research in the field of
constructionmanagementastheauthoralreadystatedinthefirstchapter.
22
THECONSTRUCTIONINDUSTRYINSTITUTE
2.1.1 ProjectDefinitionRatingIndex
Thepresentessayisfundamentallybasedonpreviousresearchdonejointly
bytheCIIandtheUniversityofTexasatAustin.Thatworkstartedin1994with
the development of the Project Definition Rating Index (PDRI) for Industrial
Projects(GibsonandDumont,1996).
2.1.1.1 TheFirstTool
ThePDRIforIndustrialProjectswastheresultoftheeffortsmadebythe
research team 1132 of the CII in a joint work with the University of Texas at
Austin.Thisearlyefforttodevelopamanagerialtoolthatwasbothusefulandeasy
touse,waswellreceivedbytheconstructionindustry(Dumont,GibsonandFish,
1997).Thetoolwasdevelopedusinginputfromprofessionalsintheconstruction
industrywhodefinedalistof70relevantelementsinthescopedefinitionprocessof
anindustrialproject.Theseelementsarerelevantinthesensethattheresultofa
project depends heavily on how well defined they are. The 70 elements were
carefully described so that they were meaningful to the different construction
professionals in the industry. The main feature is the fact that each of the 70
elementshasadifferentweightinanattempttospecifytherelativeimportance
theyhaveonthefinalprojectsuccess.Theweightingprocessbenefitedfromthe
experienceofover50professionals.
Thephilosophyofthistoolistoallowprojectplannerstoeasilydetermine
thelevelofdefinitionneededforeachoftheelementsinthelist.Ifacertainoverall
levelofdefinitionisattained,thenprojectmanagersshouldbeabletoassesshow
successful the the project will be. This success will undoubtedly depend on the
23
BACKGROUND
executionofthelaterstagesoftheprojectthatinvolvealargerexpenditure,but
theirpreparationisdoneaspartofthedefinitionprocess.Thelaterstagesofthe
project referred to are: the development of construction documents, or complete
design, and the construction of the actual project. The elements in the list will
receiveadifferentratingdependingontheirdefinitionstatusatagivenpointin
time.Havingalltheitemsaddedprovidesanoverallratingfortheproject'slevelof
definitionatthepointintimeconsidered.Theobjectiveistoachieveasufficient
levelofdefinitionwithoutputtinganexcessiveeffortonthepreplanningprocessin
ordertomaximizethechancesofasmoothdevelopmentofthelaterstagesofthe
project.
Thisindustrialversionofthetoolwasthenvalidatedusingatotalof40
completedindustrialprojects.Thevalidationprocesswasintendedtocheckwhether
thetoolpredictedaccuratelyagreatersuccessfortheprojectsthatreceivedabetter
rating. However the linear correlation that the researchers probably expected
wasn'texperimentallyobserved(R2 =0.39).Itispossiblethatusingmoreprojects
forthevalidationprocess,wouldhaveyieldedbetterresults.Toproceedwiththe
validationthen,theresearchersseparatedthesampleofprojectsintotwodifferent
groupsbysettingadividingscoreandcomparedresultsbetweenthetwogroupsof
projects(Dumont,GibsonandFish,1997).Bystudyingtheperformanceofthese2
groupsofprojectstheyfoundthattheprojectswithbetterPDRIscores(low)were
moresuccessfulonaveragethanprojectswithworsePDRIscores(high).Results
fromthisdivisionshowedthatthegroupofprojectswithlowscoreswerecompleted
withaveragereductionsof5%incostand1%inschedule.Besides,projectsinthe
groupwithhighscoreshadaverageincreasesof14%incostand12%inschedule.
Atthesametimetheprojectsscoringbelowthedividingscorehadonaverage6%
24
THECONSTRUCTIONINDUSTRYINSTITUTE
lessamountofchangeordersthantheprojectsscoringabovethedividingscore.
This data is particularly interesting because change orders are one of the main
sourcesofbenefitforcontractorsgiventhatthenegotiationpoweroftheowneris
greatly reduced once construction is underway. Therefore, owners typically pay
morethanthemarketvalueforachangeorderthanforwellplannedwork.
2.1.1.2 PDRIForBuildingProjects:TheBirthOfASaga
Followingtheeffortsofpreviousyears,andencouragedbytheacceptanceof
thePDRIforIndustrialProjects(ChoandGibson,2001),theCIIdecidedtodevelop
a similar tool to address scope development of buildings. With the experience
obtainedintheprocessofdevelopingtheIndustrialProjectsversionandbasedon
that previous knowledge, the new CII155 research team was created in 1998.
AccordingtoCho (2000),eventually aPDRI forInfrastructureprojectsmight be
developedbytheCII.AbriefdescriptionofthecreationprocessofthePDRIfor
buildingprojectsfollows.
2.1.2 TheDevelopmentOfThePDRI
The PDRI forbuilding projects consists of a list of 64 different elements
mostlikelytoberelevantinthesuccessoftheproject.Thatis,duringtheplanning
phase of a project, those elements that require special attention to reduce the
uncertaintiesabouttheoutcomeoftheproject.
2.1.2.1 DefinitionProcess
Taking advantage of the knowledge gained after the development of the
PDRIforindustrialprojects,researchersattheUniversityofTexascontinuedthe
previouseffortwithapreliminarylistof75elements.After2iterationsandinput
25
BACKGROUND
fromover30industryrepresentatives,thelistfeatured71elements.Theresearch
team155oftheCIIusedthisdraftasstartingpoint.Then,Cho(2000)usedthe
expertise of the team members and a series of workshops with industry
representativestodevelopafinalversionofthelistthatincluded64elementsin11
differentcategories.
2.1.2.2 WeightingProcess
Forthelisttobeusefulinthesametermsastheindustrialversionwas,the
elementshadtobeweighted.Thatis,weightshadtobeassignedtoeachelementto
reflecttherelativepotential influenceoftheelementtotheprojectoutcome.the
processofassigningweightswasdoneusinginputfrommorethan69participants
inaseriesofworkshopsheldin7differentcitiesacrosstheUnitedStates.
Thecriteriathattheparticipantswereaskedtouseinordertoweightthe
differentelementswastheamountofcontingencytheydeemedappropriateforan
element when evaluating its current level definition considering that they were
abouttobeginthedevelopmentofconstructiondocuments.Thelevelsofdefinition
thatwereusedforevaluatingeachelementwasLevel1[CompleteDefinition]and5
[Incomplete orPoorDefinition] (Cho 2000). Theauthorrecalls that the criteria
used is exclusively economic and more specifically cost oriented. This poses the
questiononwhetherthetooliswelldesignedtomeasureotheraspectsofaproject
that are not economic. Another aspect to pay attention to is that Cho (2000)
collectedweightsonlyforthedefinitionlevels1and5,andassumedthatweights
forthelevels2,3,and4varylinearly.Whilethisassumptionseemswellfunded
based on previous research (Gibson and Dumont, 1996), it also determines the
behaviorofthePDRI.Therewillbeadeeperdiscussiononthistopicinchapter4.
26
THECONSTRUCTIONINDUSTRYINSTITUTE
After collecting the input from the workshops, Cho (2000) performed a
statistical analysis with all the data and filtered out sets of data that skewed
significantlythesample.Inthisprocessheeliminated8setsofdataoutofthe67he
hadoriginallycollected.Withthe59remindingsetsofdatainput,Chocalculated
themeanforeachelementandforthetwolevelsofdefinitionpreviouslymentioned.
Afterthat,allthenumberswerenormalizedtoprovidealevelofcoherencewiththe
previous PDRI for Industrial Projects. That is, the sum of all the weights
correspondingtoalevelofdefinition5hadtoresult1000,andperformingthesame
operationforthedefinitionlevel1,70.Thereforetherangeofthescoresofthe2
PDRItoolswouldbethesame.Afterthat,allthenumberswereroundedtothe
closestintegerandthelevels 2, 3, and4 interpolatedandroundedaswell. The
author recalls that rounding the interpolated numbers no longer necessarily
maintainslinearityoftheweights.
2.1.2.3 ValidationProcess
The members of the research team 155 of the CII made an effort to
demonstratetheusefulnessoftheirnewlydevelopedtool.Toaccomplishthisgoal
they conducted a validation process aimed tocorrelate PDRI scores with project
success.Thefirstproblemtheyencounteredwastodecidewhichmeasureofsuccess
touse.Afterareviewoftheliterature,theyconcludedthatthemostrepresentative
variables were: budget achievement, schedule achievement, design size
achievement,andfacilityutilizationachievement.Thisdecisionwasheavilybased
on the experience collected during the previous researches done at the CII. But
instead of combining them using the same formula to find a success index, the
27
BACKGROUND
membersoftheteampreferredtocomeupwithaformulatakingintoaccountthe
correlationsofthedifferentsuccessvariableswiththePDRI.
The CII research team (1999) conducted a survey of successful and
unsuccessfulprojectsamongCIIorganizationmembers.Thesurveyconsistedofa
questionnairethatwassentto68chosenprojectsdrawnfromtheCIIdatabase.38
questionnaires were returned and only 33 of the returned could be used for
validation purposes. The data collected in these questionnaires served to check
whetherthedifferentvariableshadacorrelationwiththePDRIscores.Althougha
linearcorrelationfactorwasnotcomputedforeachvariableconsidered,statistical
significanceprovedthatprojectswithlowerPDRIscoresshowedbetterperformance
indexesthanprojectswithhigherPDRIscores.
InanefforttocorrelatePDRIscoreswithperformancevariables,theCII
researchteam(1999)chosethepreviouslymentioned4variablesandgroupedthe
project results in threedifferent categories foreach variable. The authorrecalls
that grouping data reduces the information of the sample. The CII team, for
example, decided to rate the project performance in each area with the discrete
values1,3,and5.Sincetheoriginalvariablesweredefinedinacontinuousscale,
thatmeansgroupingthevariables.Furthermore,thegroupingcriteriawasselected
completelyarbitrarily,assigningthevalueof3oratperformancetothosevalues
ofthevariablesthatdifferednomorethan2%betweentheplannedvalueandthe
actual value. And assigning 1 or 5 to values falling either below or above that
interval.Oncethedatavaluesweregrouped,Cho(2000)computedthecorrelation
coefficientbetweenthePDRIscoresandeachoftheperformancevariables.Thebest
correlation found was R2 = 0.121 which indicates a weak linear correlation.
28
THECONSTRUCTIONINDUSTRYINSTITUTE
Nevertheless,Chowasabletoproposealinearprojectsuccessequationthathada
correlationfactorofR2 =0.1536,bycombiningthethreemostheavilycorrelated
performancevariableswithequalweights.Withacurvilinearmodelheimproved
thecorrelationwithR2=0.2313,butitisstillverylittlesignificant.
2.2 THEMASSACHUSETTSINSTITUTEOFTECHNOLOGY
TheMassachusettsInstituteofTechnology(MIT)isaworldwiderenowned
educationinstitutioncommittedtoprovideaneducationthatexcelsinallfieldsof
science, technology and other human sciences. MIT is a private university that
housesover900facultyand10,000graduateandundergraduatestudentsfromall
over the world. In addition, 47 alumni, faculty, researchers and staff have won
Nobelprizes.
2.2.1 TheMITCapitalProject
MIThasembarkedinthemissionofrenovatingitscampuswithamajor
capital investment onnewandrenovated building facilities. TheCapital Project
representsatotalbudgetof$1.4billion,10buildings(renovatedandnew)housing
aboutonemillionstateoftheartsquarefeet,andseveralinfrastructurefacilities
andlandscapeimprovements.Inadditionmostofthebuildingsarearchitecturally
unique, continuing the MIT tradition of building a unique campus that fosters
creationandinnovationineverypossibleway.Infactatleastthreeoftheselected
architectsarePritzkerprizewinners(ENR,2003).Thisspeaksofthequalityand
representativity of the projects everybody at MIT is talking about. This is an
ongoingeffortandatremendousopportunitytolearnandgainhandsonexperience
inmanyways.Suchapoolofuniquebuildingprojects,representinganinvestment
29
BACKGROUND
of$1billion,happeningatthesametimecannotgowastedintermsoflearning.In
thissense,MIT'sCapitalProjectrepresentsauniqueopportunitytotesttheories
andhypothesissincealltheprojectshavecommoncharacteristicsthatmakethem
easilycomparable.Theyareallsponsoredbythesameowner.Theyareallbuiltin
the same campusandTheywill all bebuilt within a shortperiod of time, thus
undercomparableeconomicandenvironmentalconditions.Allthesefactorsreduce
the possibility of dispersion of the data caused by factors that are not directly
relatedtotheprojects.
2.2.2 TheMITDepartmentOfFacilities
TheDepartmentofFacilitiesofMITmanagesthedesign,constructionand
maintenanceofallthebuildingsownedbyMIT.Itworkswithatightbudget,and
becausefundsfornewconstructioncomemainlyfromdonations,allthedecisions
madeareinsomewayinfluencedbytheneedtokeeptothebudget.However,since
MIT is an educational institution that performs on a time basis, it needs some
facilitiesataduedate,andthecostofnothavingthefacilityattheexpecteddate
cansometimesbeprohibitive.Therefore,decisionsaremanytimesacomplexissue
attheDepartmentofFacilities.
To prevent eventualities to the maximum extent, the department has
adopted thepolicy ofperforming completedesign ofitsbuildings beforestarting
construction.Thishasadoubleoutcome.Ononesideitreducesthebidsreceivedfor
theworkbecauseofthehigherlevelofdetailprovidedtothesubcontractors.Onthe
othersideitreducestheriskofcostoverrunsandscheduledelaysthatcomefrom
incomplete definition because it ensures that all possible problems have been
studiedbeforeembarkingintoconstruction.
30
THEMASSACHUSETTSINSTITUTEOFTECHNOLOGY
2.2.3 TheMediaLaboratoryExtensionProject
OneoftheseveralbuildingsprojectedinthecapitalrenovationofMITisthe
extensionofoneofthemostwellknownlaboratories.TheMediaLaboratory(Media
Lab)focusistodevelopthetechnologiesthatsoonerorlateraretobecomepartof
oureverydaylives.Duringthedecadesofthe1980'sand90'stheresearchdoneat
theMediaLabwasrecognizedastheleadingedgeofthisfront.Attheentranceto
the21st centurytheMediaLabwantstoexpanditsfacilitieswithanewbuilding
thatwillhostsevenresearchlaboratories,displayspaces,lectureandconference
rooms,andacafe(MakiandAssociates,2003).Amongtheuniquefeaturesofthis
buildingtheauthorrecallsthecustommadecurtainwallthatgivesthebuildinga
speciallookbyidentifyingthedoubledecklaboratorieswiththeexteriorenclosure
ofthebuilding.
The author started the focus of this thesis by studying the reasons that
caused delays in the start of the construction of the building. Permission for
startingconstructionhasbeendelayedatleast3timesatthetimeofwritingthe
presentthesis(Fall2003,Winter2004,andSpring2004).Additionally,theproject
had to be modified because the actual budget didn't meet early expectations of
donations.Acomprehensiveprocessofvalueengineeringthenstartedtoidentify
possiblewaystosignificantlyreducethecostofthebuilding.
Afterreviewing theassumptionsforscopedefinition,themembersofthe
planning team accepted that those assumptions could be modified without
compromisingtheperformanceofthebuildingnorthesatisfactionofthecustomer
butyetreducingtheneedofbuildingmechanicalsystemutilities.Somuchso,that
awholefloorlocatedinthebasementcouldberemovedfromthedesign.Thatalso
31
BACKGROUND
meant that the depth of the slurry wall could be significantly reduced. For
engineeringinterest,itisworthnotingthattheMITcampusislocatednexttothe
CharlesRiver,upstreamoftheregulatingdamthatcontrolsthewaterlevelofthe
river in relation to the Atlantic Ocean, and prevents the penetration of saline
waters.
These changes in scope definition of the project, that are expected to
ultimately free the path for construction of this new building, lead the author's
interesttostudythescopedefinitionprocessofprojectsandfindwaystoguarantee
theirsuccess.Havingfoundintheliteraturetheexistenceofarecentlydeveloped
PDRIforbuildingprojectswithapparentlyalotofpotentialandbackedbytheCII,
theauthorfellcompelledtoturnitintoausefultoolfortheuseoftheMIT.Iexpect
itsusagewillproveusefultotheDepartmentofFacilitiesandotherorganizations
thatmightfinditappealing.
32
3 METHODOLOGY
ThemethodologytheauthorusesforvalidatingthePDRIforMITbuilding
projects shares many aspects of the previous validation efforts made by Cho,
Furman and Gibson (1999), Cho (2000), Cho and Gibson (2001). During those
efforts, the process consisted of collecting data from various projects using an
extensivequestionnairethatwasaddressedtothepeopleinthemanagingteamof
theprojects.Similarlytowhatpreviousresearchershaddone(Cho,Furmanand
Gibson1999),theauthorlookedforavailableprojectswithintheorganizationand
thathadbeencompletedwithinafewyears.Thatwouldensurethepossibilityto
collectthenecessaryobjectiveandsubjectivedatarequiredinthequestionnaires.
Theauthorusedthesamequestionnairesdescribed inChapter2 andaddressed
themtotheprojectmanagerswhowererepresentingMITduringthedevelopment
processthroughtheDepartmentofFacilities.
3.1 SAMPLEOFPROJECTS
Asitwasmentionedinthepreviouschapter,theMITcampuspresentsa
uniqueopportunitytostudytheperformanceofbuildingprojectsduetothemajor
capital development underway. The MIT campus had not seen any major
constructionprocessforaperiodof15years(ENR,2003)andnowstartsaperiodof
METHODOLOGY
heavyrenovationandnewconstructionthatwillspanover10years,withatotalof
10newandrenovatedbuildingsthatwillcostabout$1billion.Whiletotalcapital
investmentwill reachover$1.4 billion.Thisisthenextmajorbuildingprogram
sincethe1970'sinMIT(MITNews,2002).Theprogramofficially startedinthe
year2000,andatotalof6projectshavebeencompletedatthetimeofwritingthis
paper.Theseprojectsare:theStataCenter,theSimmonsHall,theZesigerSports
andFitnessCenter,theDreyfusChemistryBuilding,224AlbanyStreetGraduate
Dormitory,and70PacificStreetDormitory.Theprojectswhicheitherareunder
construction,orhavenotyetstartedtheconstructionphaseatthetimeofwriting
thethesisare:theBrainandCognitiveSciencesproject,theBakerHouse,theEast
Campusproject,andtheMediaLabExtension.
TherenovationoftheexistingAlumniPool,andtherenovationofBuilding
E19areexamplesofotherminorconstructionprojectswhicharenotincludedinthe
CapitalProject.
Tohave asamplethatstatistically representsthecapital effortmadeby
MIT, the author wanted to gather data from as many completed projects as
possible.Inordertoaccomplishthis,the6completedprojectsmentionedabovewere
surveyed.However,afteraninterviewwiththeprojectmanagerfortheSimmons
Hall building, the author and the project manager agreed that this particular
projectwasnotsuitableforastatisticstudybecauseitsuniquenesseswouldskew
the sample in excess. The author managed to keep the sample at 6 projects by
eliminating the Simmons Hall project, and including the E19 renovation project
intothestudy.AtthispointtheauthorwouldhavegladlyincludedtheAlumniPool
renovationaswell,butitwasnotcompletetoprovidethedataneeded.
34
SAMPLEOFPROJECTS
3.1.1 SimmonsHall
Althoughnotfinallyincludedinthesample,SimmonsHallisworthbeing
describedduetotheuniquenessfactorsthatmakeitnotsuitableforthestudyhere
presented.Inmany waysthebuilding was bornas an architectural experiment.
Withthepurposeofhousing350undergraduatestudents,andthusincreasingthe
capacityofoncampushousinguptothegoalofbringingoncampusallfreshmen
students, the architectural team was assigned the mission of creating an
environmentthatwouldpushandincentivizethecreativenessofMITstudents.As
MITpresidentDr.CharlesVestsaidatthededicationceremonyofthebuilding:
[thebuildingis]unlikeanythingoncampusorelsewhere.
Besidesthestructuralshapeofthebuilding(see Figure3.1)inwhichthe
architectchallengesthespatialvisioncapabilitiesofresidentsandpedestrianswith
open holes, atriums and cantilever corners all over the building, the facade is
uniqueineverysenseandiswhatconfersanopencharactertothebuilding.Instead
of showing standard size windows, the anodized aluminum facade features
hundredsoftinylittlewindowsthatgivethebuildingaporosityappearance.Every
singleroominfacthas9openablewindows.Thisisnotaregularbuildingbyany
means,andevenfortheMITstandardsitrepresentsamilestoneinthecampus
architecture.Thiswasreflectedintheconstructionprocess.Fortheconstructionof
thefacadethewindowsweremountedinsetsonprecastunits;dozensofthemwere
35
METHODOLOGY
required,buttheywerenotinterchangeable.Allthesefeaturesalongwithmany
othersaddeduptothesingularityofthisremarkablebuilding.
Figure3.1ViewofthefacadeofSimmonsHall.
Butbesidesallthesefacts,themostcumbersomethingthatconvincedthe
authorofnotusingthisprojectfortheresearchwastheassertionbyitsproject
directorthatwhilethefirstteamofarchitectsdidagoodjobinconceptualdesign,
thesecondteamthat implementedthisdesigndida poorjobunderstandingthe
needsofsuchasingularbuildingandjeopardizedthefinaloutcomeoftheproject.
36
SAMPLEOFPROJECTS
Figure3.2InteriorviewoftheSimmonsHallmainlobby.
3.1.2 70PacificStreetDormitory
AlsoknownbyitsresidentsasSidney&Pacificbecauseitislocatedatthe
cornerofthesetwoCambridgestreets,thebuildinghasbecomethehubofgraduate
residentiallifeatMIT.Thedormhousesover700studentsinoneroom,tworoom,
anddoublequadapartments(4studentsinatwobedroomapartment).Besides
thebasicamenitiesoftheseapartmentssuchasindividualbathroomsandkitchens,
thebuildingalsoofferscommonkitchensforeveryfloor,awellassortedgameroom,
gymnasium,courtyard,differentcommonroomsfororganizingactivities,lounges,
andstudyrooms.
37
METHODOLOGY
Figure3.3Viewofthemainentranceof70Pacificstreet.
Thebuildingis9storieshigh,has412,000ft2,andMIThadtobuilditanew
withtightrestrictionsonbudgetandschedule.Inadditiontherewerepermitting
problemsthatwerefinallysolvedwiththegiftfromMITtothecityofCambridge,of
the park lying across Sidney street. This was exchanged for the building rights
associatedwiththeparcelthatallowedtoconstructmoredenselyinthedormsite.
3.1.3 BuildingE19Renovation
TherenovationprojectofbuildingE19involvedthecompleterenovationof
the5thfloorofthis7storiesbuildingintoaresearchandlaboratoryspace.Thenew
floorwith20,000ft2wastohousethePicowerCenterforLearningandMemorythat
38
SAMPLEOFPROJECTS
lateronwouldmovetothenewStataCenter,andtheMcGovernInstituteforBrain
Research(MIBR).TheexistingspacewasoccupiedbytheOfficeoftheController.
Inaddition,3,000ft2ofexistingvacantspaceinthe3rdfloorweretransformedinto
officeandconferencespacefortheheadquartersoftheMIBR.
Duetotheresearchscheduleneedsofthenewoccupants,theprojecthadto
bedesignedandconstructedin7months.
Figure3.4TheDreyfusChemistryBuildingislocatedattheheartofthe
campus.
39
METHODOLOGY
3.1.4 DreyfusChemistryBuilding
AlsoknownasBuilding18,becauseofthetraditionatMITtogivenumbers
toeverything.Thisbuildingisalaboratoryspaceforresearchinchemistryenclosed
inamodernistshelldesignedbyI.M.Peiin1967(see Figure3.4).Theresearch
facilityhadtobecompletelyrenovatedwithallitssystemsreplaced.
Theprojectwasn'tshortontechnicaldifficulties.Butthemaindifficultywas
causedbyitsuserswhodidn'twanttoleave thebuilding duringitsrenovation.
Therefore,themanagershadtocomeupwithaplantorefurbishthebuildingwhile
2/3ofthebuildingwereoccupiedwithallitssystemsrunning.
Thebuildinghasacapacityof264occupantsin14facultygroupsandatotal
area of 132,000 gross ft2. This project was granted the Merit Award for the
SuccessfulConstruction,andthearchitectreceivedtheResearchandDevelopment
DesignoftheYearAward.
3.1.5 224AlbanyStreetGraduateDormitory
Talkingaboutalias,herewehaveagoodexample.InMIT'snomenclature
thisbuildingiscalledNW30,butduetoitsoriginaluseitisalsoknownas"The
Warehouse".Thebuildingwasbuiltin1890asanindustrialwarehouseandinthe
lastyearswasusedforstoragespace.Therenovationofthisbuildingstartedinthe
year2000withthegoalofconvertingitintoa120singlesuiteapartmentsusingthe
89,000ft2 available.Theseapartmentsareoccupiedbygraduatestudentsduring
thefallandspringsemesters,andduringthesummertheapartmentsareofferedto
conferencevisitors.
40
SAMPLEOFPROJECTS
Figure3.5224Albanystreetisalsoknownas'TheWarehouse".
3.1.6 ZesigerSportsAndFitnessCenter
The new sports facility of MIT brings in a 11,000 ft2 stateoftheart
gymnasium,anOlympicclasspool,atrainingpool,6squashcourts,amultiactivity
court,administrativeofficesandamedicalarea.Thebuildingisbuiltbetweenand
connects to 2 other existing buildings that also house sports facilities for the
campus members. The project included the construction of the new 3 stories
buildingof125,000ft2,andtherenovationof37,000ft2ofexistingspace.
41
METHODOLOGY
Figure3.6InteriorviewoftheZesigerCenterfacilities.
3.1.7 StataCenter
Withatotalof716,000ft2theStataCenterbecomesoneofthemammothsof
thecampus.It hasbeendesignedbythearchitectFrankGhery,andthisisthe
program's as well as the architect's largest project (ENR, 2003). This is a truly
custombuiltfacilitythathousestheLaboratoryforComputerScience,theArtificial
IntelligenceLaboratory,theLaboratoryforInformationandDecisionSystemsand
theLinguisticsandPhilosophyfaculty.
Althoughtheprojecthasrecentlybeencompletedandnewoccupantsare
justmovingin,theauthorconsidersreallyimportanttoincludethisprojecttothe
databasebecauseofitssizeandrepresentativityofwhattheMITCapitalProjectis.
42
THEQUESTIONNAIRE
Figure3.7TwoviewsoftheStataCenterfromVassarStreet.
3.2 THEQUESTIONNAIRE
Theauthorusedthesamequestionnairethatwasusedinpreviousresearch
(Cho 2000). The reason for that decision is that the questionnaire that was
developedfortheCIIwasverygenericthusapplicabletoawiderangeofprojects.
Furthermore,itisverycomplete,soitcollectsallthenecessarydatatoperformthe
validationanalysisofthePDRI.ThequestionnairecanbeviewedinAppendixB.
Thequestionnaireisdividedinto9differentsections:
1. ProjectBackgroundInformation
43
METHODOLOGY
2. GeneralProjectInformation
3. ScheduleInformation
4. CostInformation
5. ChangeInformation
6. Financial/InvestmentInformation
7. OperatingInformation
8. CustomerSatisfaction
9. ProjectRatingInformation
Sections1and2gatherthenecessarydatatounderstandthenatureofthe
project,itspurposeandits organizational structure.Sections3, 4 and 5 ask for
objective measures of success with several closed ended questions and an open
endedquestioninsections3and5.Sections6,7and8askforsubjectivemeasures
ofsuccessusingclosedendedquestionsthatevaluatesuccessina5degreescale,
andopenendedquestions.Section7alsoasksforobjective measuresofsuccess.
Section9consistsoftheunweightedPDRIsheetwhichistobecompletedbythe
managingstaffoftheproject,basedonthestateofdefinitionattheendofdesign
development.
This questionnaire was based on the questionnaire developed for the
industrialversionofthePDRI(DumontandGibson,1996)andmodifiedaccordingly
totheneedsofthebuildingversion(Cho,2000).
44
THEQUESTIONNAIRE
3.2.1 VariablesUsed
Outofallthequestionsaskedinthequestionnaireonlyahandfulproved
relevantforperformingthestatisticalanalysis.Thereasonsfordiscardingmostof
the questions for statistical analysis is that the questions were not strictly
applicabletothedataavailableinthewaytheywereformulated.Let'stakethe
questionsregardingcostinformation,forexample.Notalltheorganizationsusethe
samecriteriatoallocatethecostpacketsintothedifferentgroups.Thistranslates
intothedifficultythatrepresentstrackingeverycostcategoryandreorganizethem
into the groups suggested by the questionnaire. For this reason, from the cost
breakdowninformationinquestion4.1ofthequestionnaire,onlytotalprojectcosts
wereconsideredfortheanalysis.
Some of the questions included in the questionnaire, are more geared
towards providing the researchers a more accurate idea on the project
characteristics and increasing their understanding of the project rather than
providingdatatobeprocessedstatistically.Themainreasonisthatsomequestions
are not accurate enough for that purpose. An example of this can be found in
question5.5whererespondentsareaskedtoprovidedataonchangesafterproject
authorization that represent more than 1% of the project budget. The question
doesn'tspecifywhetherrespondentsneedtospecifyifthereismorethanonechange
meeting thecriteria, thus it is not possible totrackthe specific nature of those
changes.
FreemanandBeale(1992)foundinabibliographicalreviewthatthemost
cited measures of project success are technical performance efficiency of project
45
METHODOLOGY
execution (the ability to meet cost and schedule targets). Nonetheless, they
proposedmeasuringsuccessbasedonfinancialperformance.
In any case this research does not intend to assert which is the most
accurateorvalidmeasureofprojectsuccess,butrathertofindoutwhichoneis
mostusefultounderstandPDRIscores.
In conclusion, the questions that were finally used for further statistical
studywere:
PlannedConstructionDurationfrom3.1
ActualConstructionDurationfrom3.1
PlannedTotalProjectCostsfrom4.1
ActualTotalProjectCostsfrom4.1
NumberofChangeOrdersfrom5.1
SubjectiveFinancialPerformanceassessmentfrom6.1
SubjectiveOriginalIntentMatchingassessmentfrom8.1
SubjectiveSuccessassessmentfrom8.2
Thedatacollectedfromthefourfirstquestionshadtobemathematically
treatedinordertoremovetheinfluencethatthesizeoftheprojectswouldexercise
overthesample.Theoperationperformedwastosubtracttheplannedvaluefrom
theactualvalueanddividetheresultbytheplannedvalue.Seeexpressionbelow:
Actual ValuePlanned Value
Planned Value
46
THEQUESTIONNAIRE
Thisoperationyieldsanondimensionalvaluethatcanbeeitherpositive,
negative,orzero.Whentheresultiszerothenthereisnovariationbetweenthe
plansandtheactualimplementationoftheproject.Whentheresultispositive,it
meansthattherehasbeenanincreaseincostorinschedule,andtheabsolutevalue
representsthepercentageoftheincreaserelativetotheplannedvalue.Whenthe
resultisnegative,itmeansthattherehasbeenareductionincostorinschedule,
andtheabsolutevaluerepresentsthepercentageofthereductionrelativetothe
plannedvalue.
Afterperformingthisoperation,theauthorobtains6differentvariablesto
measure project performance and find a relationship between PDRI scores and
projectsuccess.ThevariablesarelistedinTable3.1.
Table3.1Variablesusedformeasuringprojectperformance.
47
METHODOLOGY
3.3 INTERVIEWS
The author worked closely with the Assistant Director of Facilities for
Infrastructure and Special Projects of the Department of Facilities in MIT to
identifypotentialprojectstobeusedforthestudyandtocontacttheprojectleaders
thatwouldbeinterviewedtocollectthenecessarydataforthestudy.Assoonasthe
authorwasgiventhecontactinformationforeachoftheprojectdirectors,anemail
wassenttorequestanappointmentforaninterview.
Thegoaloftheinterviewswastopresenttheobjectivesoftheresearchto
theprojectdirectorsthatweretoprovidethenecessarydata,andtointroducethe
questionnaire.Inthisway,theauthorwasabletoquicklydispeldoubtsonseveral
questions and at the same time engaged the interviewees to help in the
development of the research. In some interviews the project director showed
satisfactionandimmediatelyafterbrowsingthequestionssaidthathewouldgladly
completethequestionnaireandsenditbacktotheauthor.Inothercasestheproject
directorwould try tosatisfy the data required on the spot as much as possible.
However,inallcasessomeofthequestionsrequiredcheckingthearchiveswhich
alwaysimpliedthattheinterviewhadtofinishwithoutthoroughcompletionofthe
questionnaire.
Incertaincasesthequestionnairewasreturnedtotheauthorwithsome
fieldsincomplete.Ifthathappenedtheprojectdirectorwasimmediatelycontacted
byemailandaskedtoreviewthemissingdata.Ifhedeclaredhimselfunableto
provide the required information he would suggest another member of the
managingteamthatwouldlikelybeabletoprovidetherequiredinformation.In
thoseparticularsituations,thedatacollectionprocesssloweddownconsiderably.
48
INTERVIEWS
Anotherfactorthatsloweddowndatacollectionwasthetimeconstraintsof
theprojectdirectors.BecauseofthehugeeffortbeingmadebytheDepartmentof
Facilities,thetimeavailabilityofitsprojectdirectorsisverylittle,soscheduling
interviewswasdifficult.
Anecdoticallytheauthorrecallsthatforcertainfinancialinformationthat
wasnotreadilyavailabletothestaffintheDepartmentofFacilities,theauthor
approachedapersonalfriendofhisintheOfficeofBudgetandFinancialPlanning
ofMITtoaskforthatinformationinamoreexpeditemanner.Inexchangeforthis
favorstheauthorbribedhisfriendwithasailingrideintheCharlesRiver.
3.4 STATISTICALANALYSIS
Thefirststepistoanalyzewhetherthedatacollectedisstatisticallyvalidto
beusedinthestudy.Forthatpurpose,andtorepresentthedistributionofeachset
ofdata,theauthorusesboxplotswhichprovidealotofinformationinagraphical
manner.Abox plotfeaturesdifferentelementseachhasavery specificstatistic
meaning.Thethreecentralhorizontallinesrepresenteach:thefirstquartile,the
medianandthethirdquartile,whicharerespectivelydisplayedfrombottomtotop.
Outliersarerepresentedaspointsoutofthebox.Fromthebottomandthetopof
theboxmayextendtwohorizontallinesrepresentingthedatapointsoutsideofthe
boxthatarenotoutliers.Outliersaredatathatliebeyondadistancefromthebox
equalto1.5timestheinterquartiledistance,orthedistancebetweenthefirstand
thirdquartiles.
The correlation factor r is a statistic parameter that describes how
consistentthelinearcorrelationbetweentwosetsofdatais.Thetwosetsmusthave
49
METHODOLOGY
thesamenumberofobservations.Thecorrelationfactorvaluescanvarybetween1
and1.Whentheabsolutevalueofthecorrelationfactorapproaches1,itmeansthat
thereisevidenceofastronglinearrelationshipbetweenthetwosetsofdata.Onthe
contrary,whenitsvalueapproaches0,itcannotbeinferredthatalinearrelation
exists.Apositivevalueimpliesthatthereisadirectrelationship,thatis,thehigher
the value in the first dataset, the higher the corresponding value in the second
dataset.Onthecontrary,ifthefactortakesanegativevaluethenthereexistsan
inverserelationship,thatis,thehigherthevalueinthefirstdataset,thelowerthe
corresponding value in the second dataset. A related statistic parameter to the
correlationfactorrisR2,whichiscomputedtakingthesquareofr.Thisisusedto
arguetheexistenceofalinearrelationship,sincethevaluesnowvaryonlybetween
0and1.Valuesof R2 willalwaysbelowerthanabsolutevaluesof r,makingthe
evidenceofalinearrelationshipmoredemanding.
Thescatterplotisastatistictoolusedtorepresentdatagraphicallyandto
identify,inaneasyway,relationshipsthatnumberscouldotherwisehinder.Itis
normally used to represent 2 different sets of data and to show relationships
betweenthem.
Theregressionanalysisisyetanotherstatisticaltoolusedwhencorrelation
between two or more sets of data exists. Its most common use is to infer the
equationofthelinethatapproximatestherelationshipbetweendata.Howeverit
canalsobeusedtoinferequationsthatarenotnecessarilylinear.
All the statistical operations and tools described above are done for this
thesisusingthesoftwareMaple8fromWaterlooMapleInc.initsversionforLinux.
50
PROJECTRATINGINFORMATION
3.5 PROJECTRATINGINFORMATION
Inordertoefficiently processtheinformation from theunweighted score
sheetsforeveryproject,theauthordesignedacomputerspreadsheet(see Figure
3.8)thatautomaticallyweightedthedefinitionlevelsforeachofthe64different
elementsinthePDRIlist.Thespreadsheetisdividedinto6differenttabs:3for
enteringdefinitiondataand3containingtheweightinginformation.The3tabsfor
entering definition data are organized following the Sections structure of the
PDRIelementslist.Theother3tabswithweightinformationarethecorresponding
tabs of the data input tabs. Therefore, Tab 1 is designed to enter the data for
Section 1 of the PDRI; Tab 2, for Section 2; and Tab 3, for Section 3. Tab 4 is
designedtocontaintheweightinformationforSection1;Tab5,forSection2;and
Tab6,forSection3.
It is worth noting that the only information included in the weight
information tabs are the weights associated to definition levels 1 and 5 of each
element. The author has conserved the original argument made by Cho (2000)
concerningtheexistenceofalinearrelationshipbetweenthelevelofdefinitionof
each element and its corresponding weight. Thus, the spreadsheet interpolates
linearly the corresponding value for every definition level without rounding the
value.ThisdiffersfromwhatChodidinhisresearch,andhasimplications.The
mostdirectimplicationisthatbyroundingtheinterpolationsfordefinitionlevels2,
3 and 4, the model would no longer follow the assumption of linearity between
definitionlevelsandweights.
Withthisspreadsheetitisveryeasytoenterthenumericvaluesofeach
definitionlevelandgetthecorrespondingweightinstantaneously.Inaddition,the
51
METHODOLOGY
totalscoreforeverycategoryandsectionisalsoautomaticallycomputed.Finally,
the person entering the data can watch in real time how the total PDRI score
changesasheintroducesthedata.
Thespreadsheetwaseditedusingtheopensourceandfreesoftwarenamed
CalcfromOpenOffice.org1.0.
52
SUMMARY
3.6 SUMMARY
In this chapter the author has presented the methodology used for
validating thePDRI forMITbuilding projects.Theprojectsusedforthesample
have also been described. Justification for the projects not used has also been
provided.Thetotalnumberofprojectsusedis6:
3researchandlaboratoryspaces(buildingE19,theDreyfusChemistry
BuildingandtheStataCenter).
2graduatedormitories(70PacificStreetand224AlbanyStreet).
1recreationalandathleticfacility(ZesigerSportsandFitnessCenter).
53
4 RESULTSANDANALYSIS
This chapter presents the results of the research and the analysis
performed.Italsocomparesthisresultstopreviousresearch.
4.1 RESULTSOVERVIEW
Sincetheamountofrelevantdatacollectedissmall,theauthorprefersto
presentithereratherthaninaseparateappendix,sothatdiscussioncanfollowa
morenaturalpath.Thedataanalyzedinclude:
thePDRIscoresforeverybuildingproject
theperformancevariablerelatedtocost
theperformancevariablerelatedtoschedule
theperformancevariablerelatedtochangeorders
theperformancevariablerelatedtofinancialperformance
theperformancevariablerelatedtoachievementoforiginalintent
theperformancevariablerelatedtoownersatisfaction.
RESULTSANDANALYSIS
70PacificStreet
E19Renovations
DreyfusChemistryBuilding
224AlbanyStreetDorm
ZesigerCenter
StataCenter
Cost
Schedule
0.17
0.05
0.16
0.28
0.4
0.04
0
0
0.1
0
1
0.31
Change
Orders
3
1
12
13
22
N/A
Financial
Performance
5
3
4
3
4
2
Intent
Success
PDRI
5
5
5
4
5
5
5
5
5
4
5
5
115.75
163.25
204.25
120
92
182
4.2 BOXPLOTS
Figure4.1Boxplotsofthe6differentsuccessvariablesused.
56
BOXPLOTS
ThereisanoutlierintheboxplotcorrespondingtotheSchedulevariable.
The point in the box plot corresponds to the Zesiger Sports and Fitness Center
Project.IfwelookwithmoredetailattheotherboxplotgraphsinFigure4.1and
comparethemtothedatainTable4.1,wecanseethatthissameprojectskewsthe
dataagaininthenumberofchangeordersandthecostvariables.Atthesametime
thisisthebestrankedprojectwithaPDRIindexof92.Itisverysuspiciousthata
projectthathasrankedsowellindefinitionlevelcausessomanyproblemsatthe
timeofexecution.Acloserlookatthecorrespondingquestionnairerevealsavery
interesting fact: the only single cause for a 100% schedule delay is a funding
change.Therefore,thedelayisnotcauseddirectlybyalackofscopedefinition.This
kind of troubles are particular and characteristic of MIT's culture of building.
BecauseMITreliesheavilyongenerousdonations,thatkindofthingscanhappen.
However,thepurposeofthePDRIisnottopredictwhetheradonationforanMIT
projectisgoingtobechangedduringthelifeoftheproject,butrathertoidentify
potentialareasofproblemsthatstemfromalackofprojectdefinition.Becauseof
allofthesereasons,theauthordecidestoremovetheZesigerSportsandFitness
Centerfromtheprojectdata.WiththeZesigerSportsandFitnessCentertakenout
ofthedataset,anewsetofboxplotsisneeded,seeFigure4.2.
Oncemore,wefindanotheroutlierinscheduleperformance.Inthiscasethe
projectinquestionistheStataCenter.Nevertheless,sinceitdoesnotskewany
other parameter and there is no particular reason that justifies removing this
projectfromthedataset,theauthorpreferstoconservethisprojectandcontinue
withtheanalysis.
57
RESULTSANDANALYSIS
Figure 4.2 Box plots of the 6 different success variables used after
removingtheZesigerSportsandFitnessCenterproject.
4.3 SCATTERPLOTSANDCORRELATIONFACTORS
Thissectionprovidesanothertypeofgraphicalrepresentationofthedata.
Nowthepurposeistoobservehoweachoftheperformancevariablesconfrontsto
thePDRIratings.Toaccomplishthistheauthoremploysbivariatescatterplots
witheachofthepairofdatausingthePDRIratingsastheindependentvariable.
Also,theauthorcalculatesthecorrelationfactorexistenttogiveanumericalvalue
toanypossiblecorrelation.
58
SCATTERPLOTSANDCORRELATIONFACTORS
Figure4.3Scatterplotsshowingthecorrelationbetweenthedifferent
performancevariablesandthePDRIindexes.
59
RESULTSANDANALYSIS
The first plot represents the scatter plot of PDRI scores versus the cost
performancevariable.Inthisplotwecanidentifyclearlyalineartendencyofthe4
bottompoints.Howeverthetopleftpointinthegraphdestroysthatline.Thispoint
correspondstothe224 Albany Streetproject.Takingacloserlooktotheproject
questionnaire,theauthorrecallsthattheprojectsufferedfromatotalof13change
ordersforanamountequivalentto10.6%ofthetotalestimatedcostoftheproject.
Additionally, after project authorization there were 3 scope/design changes that
increasedthecostoftheprojectforatotalvaluerepresenting8.4%oftheplanned
constructioncosts.Thisisanobjectiveindicatorofalackofdefinitioninpreproject
planning.Anotherconsiderationthattheauthormakesisthataprojectscoringas
lowas120inthePDRI(being70thelowestpossibleand1000thehighestpossible)
one would expect that the project performed better. In addition, out of all the
projectssampledthisisthesingleonethatdoesnotscore5inthetwosubjective
questions regarding customer satisfaction. That is, the performance variables
measuringoriginalintentmatching,andprojectsuccess.Finally,whencomputing
thelinearcorrelationfactorofthisvariablerespecttothePDRIwithandwithout
the224AlbanyStreetproject,thedifferenceinresultsisdramatic.Withtheproject
takenintoaccount r receivesavalueofjust0.174,whilethevalueof r jumpsto
0.978 if the project is removed from the sample. Taking all these factors into
account,theauthordecidestoremovethe224AlbanyStreetProject,forthepurpose
offindingacorrelationbetweenthetwonamedvariables.
Noclearrelationcanbedrawnfromtheotherscatterplots.Inallcasesthe
spacialdistributionofthepointsiseitherrandomlooking,orwithatendencyto
uniformity (in the owner satisfaction related questions), which does not imply
correlationofdata.
60
SCATTERPLOTSANDCORRELATIONFACTORS
One observation the author makes however, regarding the second plot
representingtherelationshipbetweenPDRIscoresandprojectschedules,isthatall
theprojectsscoringbelow170inthePDRIdidnotpresentanyscheduledelays.
4.4 REGRESSIONANALYSIS
Thissectionpresentstheregressionanalysisthattheauthorperformsto
correlate the PDRI ratings with the projects' cost performances. In the previous
section has been shown that a strong linear relationship exists between the 2
variables.Thereforeitisinterestingtowriteanequationforthatrelationshipand
studyitscharacteristics.
Theequationofthelineapproximatedusingtheleastsquaresmethodis:
Cost Variation %=60.30.36 PDRI
Equation4.1LinearregressionoftheCostPDRIcorrelation.
ThisequationisrepresentedinFigure4.4withthepointsofthedatathat
makeacorrelationr=0.978,andR2=0.957.
Alsoamoreaccuratequadraticmodelcanbeadjustedbytheequation:
Cost Variation%=7.90.54 PDRI 0.0029 PDRI 2
Equation4.2QuadraticregressionoftheCostPDRIcorrelation.
WherethegraphicalrepresentationisshowninFigure4.3.
61
RESULTSANDANALYSIS
Thisquadraticmodelcanbeinterpretedasrepresentingthephysicallimit
thatexiststoreducingthecostoftheproject.Itcanbeobservedthatthecurvature
oftheequationincreasesasthecostsarereducedmoreandmore.Attheotherend,
theequationcanbefairlyapproximatedbyastraightline.
Oneparticularinterestingpointtostudyinbothcasesisthe0cuttingpoint,
sincethatpointdefinesthedefinitionlevelatwhichtheprojectwillpresumablybe
completedonbudget.ForthelinearmodelthatpointisPDRI=167.5,whereasfor
thequadraticmodelthatpointliesatPDRI=170.2.
62
CRITICALANALYSISOFRESULTSANDCOMPARISON
Figure4.5Quadraticregressionoftherelationship
betweenCostandPDRI.
4.5 CRITICALANALYSISOFRESULTSANDCOMPARISON
The results presented in this chapter differ substantially from those
presentedbypreviousresearchersonthetopic(Cho,FurmanandGibson1999,Cho
2000,ChoandGibson2001).Thepreviousresearchdiscussedinthefirst2chapters
ofthisthesisproposedagenericmodelforpredictingtheperformanceofabuilding
projectbasedonthedefinitionlevelduringthepreprojectplanningphase.While
theformulationofthemodelwasveryserious,thevalidationpartlackedcoherence
intheassumptionsandstrictnessintheconclusions.
63
RESULTSANDANALYSIS
In chapter 2 the author mentions that the criteria used to generate the
weightsofthedifferentelementsinthePDRIscorelistisessentiallyeconomic.The
author also recalls that the relationship established by the model between the
differentlevelsofelementdefinitionandthecorrespondingweightsisforcedtobe
linearbydefinition.Fromthese2basicmodelcharacteristics,onewouldexpectin
thevalidationprocesstofindalinearrelationshipbetweentotalprojectscoresand
successbasedoneconomicdata.Thiscanbeexplainedbecauseofthewaythetotal
scoreiscomputed.Itisdonebythesimpleadditionofthescoresprovidedbyeach
singleelementofthelist.Giventhepropertyofadditivityconsistingofthefactthat
linearfunctionsaddedtogetherresultinanotherlinearfunction,itispossibleto
inferthatthescoresthataprojectisgivenbythePDRIarelinearlyrelatedtothe
necessarycontingencytocovertheriskofindefinitionintheproject.
However,thiswasneitherprovednorproposedbythepreviousresearchers
whoformulatedthemodelintheliteraturereviewtheauthorhasfound.Infact
fromthestatisticaldatacollectedbythoseresearchers,theycouldfindcorrelation
levelsashighasR2=0.121.Erroneouslytheyclaimedthatsuchvaluerepresented
positive proof of existence of a linear relationship between PDRI scores and
performancevariablesbasedoncost.Theauthorposesthequestionwhetherthe
groupingofthedatausedbypreviousresearchersinto3discretegroupsmighthave
ruinedthestatisticalvalueofthatdata.Itwouldbeapityifsuchtheoryturned
true.
Ontheselfcriticexercisetheauthormustsaythatinordertoaccomplisha
good linear relationship representation had to discard one observation from the
samplewhosecontributionwasdoubtfultothegoodnessofthedataset.However,
64
CRITICALANALYSISOFRESULTSANDCOMPARISON
enoughargumentswerepresentedsupportingthatdecision.Inaddition,thefact
thatthosebadpointsappearedinthesamplemustbeunderstoodasawarningon
theuseofthePDRI.Theevaluationhastobemadebasedonseriousprinciplesand
withtheaimofbeingimpartialatthemomentofratingtheproject.Thisisvery
importantbecausetheprojectmanagersareratingtheirownjobandmustmakea
deepexerciseofselfcriticismeverytimetheyfaceaPDRIscoringsheet.
The other observation to make is that the PDRI does not solve all the
problemsthatmayariseduringthedevelopmentofaproject.Theearlyrejectionof
the Simmons Hall project as being a suitable project for the study, or the later
rejectionoftheZesigerSportsandFitnessCentershouldsufficeasproof.
Outofthemodelsproposedinthischapter,theauthorprefersthelinear
modelbecauseofthequalitativevalueitbringstodemonstratetheprinciplesofthe
developmentoftheProjectDefinitionRatingIndex.However,thequadraticmodel
alsoaddsanimportantfeatureasitlimitsthepossiblecostreductiontoacredible
value.Inaddition,thisquadraticmodelcanbeeventuallyapproachedbyalinear
oneonthehigherspectrumofthePDRIrangeofvalues.
65
5 CONCLUSIONS
The work here presented has made interesting findings related to the
managerialtoolforbuildingprojectsnamedProjectDefinitionRatingIndex.The
threemainobjectiveshavebeenaccomplishedsatisfactorily.
5.1 REVIEWOFOBJECTIVES
Thankstothecriticalanalysismadeinthepreviouschapter,itisnoweasier
tounderstandtheworkingprincipleofthePDRI.ThePDRIscoresoftheprojects
relatetothecostperformanceoftheproject.Thisrelationshipappearstobelinear.
Thereasonforthisrelationshipisentirelybasedonthedefinitionofthetoolitself.
Firstlybecauseofthecriteriausedbytheresearcherstoweightallthe64different
elements that compose the PDRI list. And secondly because of the decision of
interpolatinglinearlytheweightsforthedifferentdefinitionlevelsofeachelement.
In the research the author presents evidences that this relationship exists, and
theredoesn'tappeartobeanyotherreasontoexplainit.Previousresearchersmade
the hypothesis that PDRI scores could be related to other variables measuring
projectsuccess.Theauthorinsteadclaimsthatlinearrelationshiphadneverbeen
proved before with such strong correlation factors as in the present thesis. The
CONCLUSIONS
authoralsoconcludesthatPDRIscoresarenotrelatedtoanymeasureofsuccess
otherthancostvariables.
Anotheraccomplishmentofthethesisistoenlargethepoolofprojectsused
tovalidatethePDRIforbuildingprojects.Beforethepresentwork,andbasedon
theliterature,thePDRIhadbeenvalidatedusing33differentprojectswithatotal
cost of approximately $900 million. This project contributes with 6 new projects
thatrepresentatotalcostofmorethan$500million.Thisincreasesof18%inthe
numberofprojectsand55%inthecostoftheprojectssurveyed.
Finally,theauthorconcludeswithafewguidelinesforfutureuseinMIT
buildingprojects.Basedontheresultspresented,theauthorestimatesthatthepre
planningeffortofMITprojectsshouldleadtodefinitionlevelsscoring170orlower
inthePDRIforbuildingprojectsasitwasdefinedbyCho(2000).Thisisconsidered
a good target for 2 reasons. First of all, the two regressions performed in the
analysisyieldequationsthatcrossthezerovariationlineincostatpointsnearthe
170 value. Thesecondreasonisthat all theprojectsconsideredandthatscored
below170inthePDRIwerecompletedonschedule.
Theauthoremphasizesthat thefindingsbasedintheregressionsdonot
have to be valid for every organization. The analysis was performed using
exclusivelyprojectsdevelopedundertheownershipofMIT,andthereforeresults
canonlybeexpectedvalidforMITprojects.However,themethodologypresented
should be applicable to any organization willing to use the PDRI to assess the
definitionleveloftheirprojects.Thismethodologyalsotakesadvantageofthelittle
variationintheperturbationsthatmayaffectthesampleused(i.e.location,market
68
REVIEWOFOBJECTIVES
conditions,ownerexperience,etc.),giventhatalltheprojectsweredevelopedbythe
sameowner,inashortperiodoftimeandinnearbylocations.
5.2 RECOMMENDATIONS
The author strongly recommends the use of the PDRI to assess the
managingteamofaparticularproject.Thisrecommendationbecomesevenstronger
iftheteambelongstoanorganizationdevelopingseveralprojectsthushavingthe
opportunitytovalidatethePDRIusingthemethodologypresentedinthisthesis.In
this sense, it is crucial to maintain the linear characteristics of the tool by not
roundingtheinterpolatedvaluesoftheweightsused.Itisalsoveryimportantnot
togroupdatavaluesintodiscretegroups,thusreducingtherichnessofthedata
collected. There are no proofs of these actions being pernicious, but by avoiding
themtheauthorhasachievedverygoodresults.
Anotherrecommendationtheauthormakesistobeverycriticalwhenever
usingthePDRItoscoreaproject.Sincethepeopleevaluatingtheprojectwiththe
PDRIarethemembersofthemanagingteam,itbecomesaselfevaluationprocess,
inwhichhonestyiscriticaltoyieldgoodresults.
FinallytheauthorwantstoexpresshisinterestinasimilarPDRItoolbe
developed for infrastructure projects. Industrial projects and building projects
already enjoy the existence of this tool, and by creating the Project Definition
RatingIndexforinfrastructureprojects,allthedifferentfieldsoftheconstruction
industrycouldbenefitfromthisusefultool.
69
6 APPENDIXA
This appendix displays the appropriate weights to be used for the PDRI
elements in order to maintain linearity between definition levels and element
weights.
The definition of the elements has been conserved equal to the original
developedbyCho(2000).Theonlychangesthereforearetheweightscorresponding
tolevels2,3and4ofthoseelementsforwhichtheroundingprocesswouldhave
altered the linearity of the weights' sequence. As has been shown in the
dissertation, it is convenient not to round the values of these intermediate
definitionlevelsbecausethatpermitstransferringthelinearrelationshipbetween
definitionlevelsandweightsfromtheelementleveltothetotalprojectlevel.
Since the definition of the weights for each element is based solely on
recommendedcontingenciesforeachelement,dependingonthedefinitionstatusof
thatelement,thenwecaninferthatthedefinitionlevelofthewholeprojectwillbe
solelyrelatedtothenecessarycontingencyforthewholeproject.Thequantitative
differencemaynotbeverylarge,butqualitativelythisisanimportantpointthat
theauthorwantstorecalltheattentionto.Inthefollowingpagesthelistofthe
elementswiththecalculatedweightsisdisplayed.
APPENDIXA
Definition Level
2
3
22,5
14
14
11,5
16,5
11,5
14,5
8
33,25
20,5
20
16,25
23,75
16,75
21,25
11,5
44
27
26
21
31
22
28
15
5,25
4,75
4,5
5,5
9,5
8,5
8
10
13,75
12,25
11,5
14,5
18
16
15
19
5,5
6,75
7,5
5
6,5
8,25
10
12,5
13
9
11
14,5
14,5
18,25
18,5
13
15,5
20,75
19
24
24
17
20
27
0
0
0
0
0
0
0
0
1
1
2
2
2
1
1
1
11,75
7,5
8
6,75
9,25
6,25
7,75
4,5
0
0
0
0
1
1
1
1
0
0
0
0
0
0
1
1
2
1
2
2
A BUSINESS STRATEGY
A1 Building use
A2 Business Justification
A3 Business Plan
A4 Economic Analysis
A5 Facility Requirements
A6 Future Expansion / Alteration Considerations
A7 Site Selection Considerations
A8 Project Objectives Statement
B OWNER PHILOSOPHIES
B1 Reliability Philosophy
B2 Maintenance Philosophy
B3 Operating Philosophy
B4 Design Philosophy
C PROJECT REQUIREMENTS
C1 Value-Analysis Process
C2 Project Design Criteria
C3 Evaluation of Existing Facilities
C4 Scope of Work Overview
C5 Project Schedule
C6 Project Cost Estimate
Table6.1WeightedListforsectionIofthePDRI.
72
APPENDIXA
Definition Level
2
3
7,5
7,5
10,5
7,5
8,5
7
4,5
6
10,75
10,75
14,75
10,75
12,25
10
6,25
8,5
14
14
19
14
16
13
8
11
4,75
6
3,25
4
4,5
4
3,25
2,75
3
4,5
4
4,25
1,25
8,5
11
5,5
7
8
7
5,5
4,5
5
8
7
7,5
2,5
12,25
16
7,75
10
11,5
10
7,75
6,25
7
11,5
10
10,75
3,75
16
21
10
13
15
13
10
8
9
15
13
14
5
1
1
1
2
1
1
1
1
4,25
6,25
5,25
6,5
4,5
3,25
4,25
3
7,5
11,5
9,5
11
8
5,5
7,5
5
10,75
16,75
13,75
15,5
11,5
7,75
10,75
7
14
22
18
20
15
10
14
9
1
1
1
4,5
3,25
3,5
8
5,5
6
11,5
7,75
8,5
15
10
11
0
0
0
0
0
0
0
0
1
1
2
1
1
1
1
1
4,25
4,25
6,25
4,25
4,75
4
2,75
3,5
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
D SITE INFORMATION
Dl Site Layout
D2 Site Surveys
D3 Civil Geotechnical Information
D4 Governing Regulatory Requirements
D5 Environmental Assessment
D6 Utility Sources with Supply Conditions
D7 Site Life Safety Considerations
D8 Special Water and Waste Treatment Requirements
E BUILDING PROGRAMMING
El Program Statement
E2 Building Summary Space List
E3 Overall Adjacency Diagrams
E4 Stacking Diagrams
E5 Growth and Phased Development
E6 Circulation and Open Space Requirements
E7 Functional Relationship Diagrams Room by Room
E8 Loading and Unloading Storage Facilities Requirements
E9 Transportation Requirements
E10 Building Finishes
E11 Room Data Sheets
E12 Furnishings Equipment & Built-Ins
E13 Window Treatment
G EQUIPMENT
G1 Equipment List
G2 Equipment Location Drawings
G3 Equipment Utility Requirements
Table6.2WeightedListforSectionIIofthePDRI.
73
APPENDIXA
Definition Level
2
3
7,5
6
10,75
8,5
14
11
1
2,5
2
4
3
5,5
4
7
1
1
1
1
1
2,75
4
4,25
5,25
3
4,5
7
7,5
9,5
5
6,25
10
10,75
13,75
7
8
13
14
18
9
1
1
1
1
1
3,25
3,5
4,5
4,5
3
5,5
6
8
8
5
7,75
8,5
11,5
11,5
7
10
11
15
15
9
0
0
1
1
4,25
3,5
0
0
0
1
0
0
0
0
0
0
0
0
0
0
H PROCUREMENT STRATEGY
H1 Identify Long Lead Critical Equip. & Materials
H2 Procurement Procedures and Plans
J DELIVERABLES
J1 CADD / Model Requirements
J2 Documentation Deliverables
K PROJECT CONTROL
K1 Project Quality
K2 Project Control Cost
K3 Project Schedule Control
K4 Risk Management
K5 Safety Procedures
Table6.3WeightedListforSectionIIIofthePDRI.
74
7 APPENDIXB
VALIDATIONQUESTIONNAIREPROJECT
DEFINITIONRATINGINDEX(PDRI)FORBUILDING
PROJECTS
ConstructionIndustryInstitute(CII)PDRIforBuildingProjectsResearchTeam
1PROJECTBACKGROUNDINFORMATION
1.1Date:
1.2PointofContact
1.2.1Name:
1.2.2Title:
1.2.3Address:
1.2.4Tel.No.:Fax.No.:
1.2.5Email:
APPENDIXB
2GeneralProjectInformation:
2.1ProjectName:
2.2ProjectIDNumber(ifapplicable):
2.3Inwhattownorcityistheprojectlocated?
Inwhatstateorprovince?
2.4Whattypeoffacilityisthisproject?(choseone)
[]Apartments
[]Offices
[]Banks
[]Parkingstructures
[]Churches
[]Publicassembly/Performancehalls
[]Dormitories
[]Recreational/Athleticfacilities
[]Hotels/Motels
[]Research&Laboratoryfacilities
[]Industrialcontrolbuildings
[]Schools
[]Institutionalbuildings
[]Stores/Shoppingcenters
[]Lightassembly/Manufacturing
[]Transportationterminals
[]Medicalfacilities(Airport,Train/BusStation,etc.)
[]Nursinghomes
[]Warehouses
[]Other(pleasespecify)
2.5Whataretheprimaryusesorfunctionsofthisfacility?(pleasecheckallthat
apply)
76
[]Foodservice
[]Medical
[]Research
[]Institutional
[]Multimedia
[]Residential
[]Instructional
[]Office
APPENDIXB
[]Retail
[]Lightmanufacturing
[]Recreational
[]Storage
[]Other(pleasespecify)
2.7Didtheprojectconsideredinvolve(ifrenovationcostisgreaterthan50%,
consideritasarenovation)
[]Newconstruction
[]Renovation
2.8Whatwastheleveloftheprojectcomplexity?
[]High
[]Average
[]Low
2.9Wasthereanythinguniqueaboutthisproject?(pleasecheckallthatapply)
[]Newprocessortechnologyfortheorganization/location
[]Firstofakindtechnologyfortheindustry
[]Largest(scale)
[]Other(e.g.,specialdevelopmentrequirements,equipment,location,
execution,etc.)Pleasedescribe:
[]Notapplicable
77
APPENDIXB
2.10Whatwasthedeliverymethodthatyouusedonyourproject?
[]GeneralContractor
[]ConstructionManager
[]MultiplePrimes
[]DesignBuild
[]Turnkey
[]BuildOperateTransfer
[]Other(pleasespecify)
2.11Whatwasthetypeofcontractused?
[]LumpSum
[]UnitPrice
[]CostPlus
[]GuaranteedMaximumPrice
[]Other(pleasespecify)
78
APPENDIXB
3ScheduleInformation:
3.1Pleaseprovidethefollowingscheduleinformation:
Item
Planned
Actual
(mm/dd/yy)
(mm/dd/yy)
StartDateofConstruction
DocumentsDevelopment
EndDateofConstruction
DocumentsDevelopment
StartDateofConstruction
DateofSubstantial
Completion
3.2Iftherewereanyscheduleextensionsorreductionsduringdevelopmentof
constructiondocumentsandconstruction,pleaseindicatethereason(s)inthe
appropriate box(es) below by supplying the duration(s) of the change(s) (in
months)andwhetheritwasanextension(Ext)orreduction(Red).
(pleasecheckallthatapply)
79
APPENDIXB
Delay
Scope/Design
Months Ext
[]
Red
[]
Delay
Months Ext
Engineering
Red
[]
[]
[]
[]
[]
[]
Productivity
FundingChange
[]
[]
Master.
Shortage/Delivery
LaborShortage
[]
[]
Technology
Change
RegulatoryChange
[]
[]
DesignError
[]
[]
ContractDispute
[]
[]
DifferingSite
[]
[]
[]
[]
[]
[]
[]
[]
[]
[]
Conditions
Equipment
[]
[]
Availability
Weather
Design
Coordination
[]
[]
Unreal/Inaccurate
Schedule
Construction
[]
[]
Productivity
Strike
Commissioning
Problem
[]
[]
Other(please
specify)
_________________
_________________
3.3Do you have any additional comments regarding any causes or effects of
schedulechanges(e.g.,specialcauses,freakoccurrences,etc.)?
80
APPENDIXB
4CostInformation:
4.1Pleaseprovidethefollowingcostinformation:(Ifthepersonfillingoutthis
sectiondoesnothavetheinformation,pleasestateDontknow,ifitwas0,
stateas0.)
Item
EstimatedCostsatStart
ActualCostsAfter
ofConstructionDocument
ConstructionComplete
Development
TotalDesignCosts1
ConstructionCosts
FitupEquipmentCosts2
SoftCosts3
Owner'sContingency
Other
TotalProjectCosts
1. TotalDesignCostisArchitectstotalfeeswhichincludeprogramming,
schematic design, construction document development, and design
developmentfees
2. FitupEquipmentcostsincludenoncoreequipmentandFFE
3. SoftCostsincludeinterest,duediligence,andotherconsultingservices
(notincludingland)
81
APPENDIXB
5ChangeInformation:
5.1Whatwerethetotalnumberofchangeordersissued(includingduringboth
constructiondocumentdevelopmentandconstruction)?
5.2Whatwerethetotaldollaramountsofallchangeorders?$
5.3What was the net duration change in the completion date resulting from
changeorders?months
5.4Did the changes increase or decrease the length of the original project
duration?[]Increased
[]Decreased
5.5Werethereanyindividualchangesafterprojectauthorizationthatexceeded
1%oftheprojectbudget?
[]No
[]YesIf"Yes,"whatwerethetotalcumulativeeffectsandthedirectionof
thesechangeson:
a)Cost:$[]Increaseor[]Decrease
b)Schedule:months.[]Increaseor[]Decrease
c) How many changes comprised 1% of the original contract amount or
greater?
d)Whatwerethereasonsforthechanges?(Pleasecheckallthatapply.)
82
[]Scope/DesignChange
[]FundingChange
[]LaborShortage
[]RegulatoryChange
[]ContractDispute
[]EquipmentAvailability
[]Weather
[]Constr.Productivity
[]Strike
[]Engr.Productivity
APPENDIXB
[]MaterialShortage/Delivery[]TechnologyChange
[]DesignError
[]DifferingSiteConditions
[]DesignCoordination
[]Unrealistic/InaccurateSched.
[]Other(pleasespecify)
6Financial/InvestmentInformation:
6.1The decision to design and construct a facility relies heavily on specific
project financial performance measures such as capital turnover, return on
investment,benefit/costratio,returnonequity,returnonassets,etc.Forthe
majorfinancialcriteriausedonthisprojecttodate,howwellhastheactual
financial performance matched the expected financial performance
measurementusingthescalebelow?
Usingascaleof1to5,with1beingfallenfarshortofexpectationsto5being
farexceededexpectationsatauthorization,pleasecircleonlyone.
1
fallenfarshort
3
matchedclosely
5
farexceeded
83
APPENDIXB
6.2Whattypeofspecificprojectfinancialmeasurementwasusedtoauthorize
theproject(e.g.,ReturnonAssets,ReturnonEquity,InternalRateofReturn,
Benefit/CostRatio,PaybackPeriod,etc.)?
7OperatingInformation:
7.1Whatpercentofdesignsize(e.g.,squarefootage,numberofoccupants,etc.)
was planned or anticipated (at the time the project was ready to begin
developmentofconstructiondocuments)andactuallyobtainedaftertheendof
commissioning(e.g.,ifthefacilitywasplannedtobe50,000gsf,butwasbuilt
at40,000gsf,thepercentagewouldbe80%)?
Obtained
a)DesignSizeat
Commissioning
b)Wasthefacilityscopereduced/increasedinanysignificantmanner?
Ifso,pleaseindicate:
84
APPENDIXB
Buildshell,notinteriorfinish
Size
Materialsofconstruction
Equipment
Other(pleasespecify)
7.2.Whatpercentoffacilityutilizationwasplannedoranticipatedandactually
obtained6monthsaftertheendofcommissioning?
Planned
Obtained
FacilityUtilization6months
aftercommissioning
% %
Doyouhaveanyadditionalcommentsregardingfacilityutilization?
Utilizationisdefinedas"thepercentageofthefacilityactuallybeingusedversusthat
anticipated at the time the project was conceived or approved for design and
construction."Forexample,therentaloccupancymayhavebeenplannedat75%
aftersixmonthsofoperationandactuallywasonly50%.
85
APPENDIXB
7.3Afterthefirst6monthsofuse,haveoperationalandmaintenancecostsfor
thefacilitybeen:(circleonlyone)
1
fallenfarshort
3
matchedclosely
5
farexceeded
8CustomerSatisfaction:
8.1Basedontheoriginalscopeofworkforthefacilitysetpriortothebeginning
ofconstructiondocumentdevelopmentandconstruction,ratehowthefacility
matchestheoriginalintent:(circleonlyone)
1
verydifferent
2
3
4
>
5
closelymatches
Ifthefacilityisverydifferentfromtheoriginalintent,pleasespecifywhatcaused
thechanges
8.2Reflectingontheoverallproject,ratehowsuccessfulyoufeeltheprojecthas
beenusingascaleof1to5,with1beingveryunsuccessfulto5beingvery
successful:(circleonlyone)
1
very
unsuccessful
86
2
3
4
>
5
verysuccessful
APPENDIXB
8.3Doyouhaveanyadditionalcommentsregardingcustomersatisfaction?
9ProjectRatingInformation:Next,pleasecompletetheProjectRatingInformation
formlocatedonthenextfewpages.Detailedinstructionsforcompletingthisform
areexplainedbelow.
INSTRUCTIONSFORRATINGAPROJECT
87
APPENDIXB
ThePDRIconsistsofthreemainsections,eachofwhichisbrokendowninto
aseriesofcategorieswhich,inturn,arefurtherbrokendownintoelements.Scoring
isperformedbyevaluatingandratingtheindividualelements.Elementsshouldbe
ratednumericallyfrom0to5basedonitslevelofdefinitionatthepointintime
priortobeginningdetaileddesignandconstruction.Thinkofthisasazerodefects
typeofevaluation.Elementsthatwereaswelldefinedaspossibleshouldreceivea
perfectratingof"one".Elementsthatwerecompletelyundefinedshouldreceivea
rating of "five". All other elements should receive a "two", "three", or "four"
dependingontheirlevelsofdefinition.Thoseelementsdeemednotapplicablefor
theprojectunderconsiderationshouldreceiveazero.Theratingsaredefinedas
follows:
0NotApplicable
1CompleteDefinition
2MinorDeficiencies
3SomeDeficiencies
4MajorDeficiencies
5IncompleteorPoorDefinition
Torateanelement,firstreaditsdefinitionintheDescriptionsectionofthe
64PDRIElementsdocument.Someelementscontainalistofitemstobeconsidered
whenevaluatingtheirlevelsofdefinition.Theselistsmaybeusedaschecklists.Note
88
APPENDIXB
thatsomeoftheseitemsmaynotbeapplicableforyourproject.Next,refertothe
Project Rating Information form and locate the element. Please choose only one
definitionlevel(0,1,2,3,4,or5)forthatelementbasedonyourperceptionofhow
well it was defined when the project was authorized. Once you have chosen the
appropriatedefinitionlevelfortheelementpleasecheck()thecorrespondingbox.Do
thisforeachofthe64elementsinthePDRI.Besuretorateeachelement.
89
APPENDIXB
A BUSINESS STRATEGY
A1 Building use
A2 Business Justification
A3 Business Plan
A4 Economic Analysis
A5 Facility Requirements
A6 Future Expansion / Alteration Considerations
A7 Site Selection Considerations
A8 Project Objectives Statement
B OWNER PHILOSOPHIES
B1 Reliability Philosophy
B2 Maintenance Philosophy
B3 Operating Philosophy
B4 Design Philosophy
C PROJECT REQUIREMENTS
C1 Value-Analysis Process
C2 Project Design Criteria
C3 Evaluation of Existing Facilities
C4 Scope of Work Overview
C5 Project Schedule
C6 Project Cost Estimate
90
Definition Level
2
3
APPENDIXB
Definition Level
2
3
D SITE INFORMATION
Dl Site Layout
D2 Site Surveys
D3 Civil Geotechnical Information
D4 Governing Regulatory Requirements
D5 Environmental Assessment
D6 Utility Sources with Supply Conditions
D7 Site Life Safety Considerations
D8 Special Water and Waste Treatment Requirements
E BUILDING PROGRAMMING
El Program Statement
E2 Building Summary Space List
E3 Overall Adjacency Diagrams
E4 Stacking Diagrams
E5 Growth and Phased Development
E6 Circulation and Open Space Requirements
E7 Functional Relationship Diagrams Room by Room
E8 Loading and Unloading Storage Facilities Requirements
E9 Transportation Requirements
E10 Building Finishes
E11 Room Data Sheets
E12 Furnishings Equipment & Built-Ins
E13 Window Treatment
G EQUIPMENT
G1 Equipment List
G2 Equipment Location Drawings
G3 Equipment Utility Requirements
91
APPENDIXB
H PROCUREMENT STRATEGY
H1 Identify Long Lead Critical Equip. & Materials
H2 Procurement Procedures and Plans
J DELIVERABLES
J1 CADD / Model Requirements
J2 Documentation Deliverables
K PROJECT CONTROL
K1 Project Quality
K2 Project Control Cost
K3 Project Schedule Control
K4 Risk Management
K5 Safety Procedures
92
Definition Level
2
3
8 REFERENCES
Best,Rick&deValence,Gerard.1999. BuildinginValue.JohnWileyand
SonsInc.,605ThirdAvenue,NewYork,NY101580012.
Cheong, Seay P. et al. 2003. Improving Construction Client Satisfaction
through Functional Briefing. Conference Proceedings of "Construction Research
CongressinConstructionWindofChange:IntegrationandInnovation",Honolulu,
2003.
Cho, ChungSuk & Gibson, G. Edward. 2001. Building Project Scope
Definition Using Project Definition Rating Index. Journal of Architectural
Engineering,7(4),115125.
Cho, ChungSuk, Furman, Jeffrey C. & Gibson, G. Edward. 1999.
DevelopmentoftheProjectDefinitionRatingIndex(PDRI)forBuildingProjects
AreporttotheConstructionIndustryInstitute.TheUniversityofTexasatAustin.
TheUniversityofTexasatAustin,December1999.
Dumont, Peter R., Gibson, G. Edward & Fish, John R. 1997. Scope
Management Using Project Definition Rating Index. Journal of Management in
Engineering,13(5),5460.
REFERENCES
94
REFERENCES
URLaddress:http://web.em.doe.gov/em6home/empdri.html
95