Conceptual Modelling for Product Configuration …orbit.dtu.dk/files/139515799/Final_Thesis_SARA_recent_changes.pdfCONCEPTUAL MODELLING FOR PRODUCT CONFIGURATION SYSTEMS Sara Shafiee

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from orbit.dtu.dk on: Jul 08, 2018

Conceptual Modelling for Product Configuration Systems

Shafiee, Sara; Hvam, Lars; Mortensen, Niels Henrik

Publication date:2017

Document VersionPublisher's PDF, also known as Version of record

Link back to DTU Orbit

Citation (APA):Shafiee, S., Hvam, L., & Mortensen, N. H. (2017). Conceptual Modelling for Product Configuration Systems.Kgs. Lyngby: Technical University of Denmark (DTU).

http://orbit.dtu.dk/en/publications/conceptual-modelling-for-product-configuration-systems(fdd0ba04-99cc-484d-a6f7-f38af202bffb).html

CONCEPTUALMODELLING

FORPRODUCTCONFIGURATIONSYSTEMS

SaraShafiee

OperationsManagement

DepartmentofManagementEngineering

TechnicalUniversityofDenmark

ThisdissertationissubmittedforthedegreeofDoctorofPhilosophy

April2017

ii

i

SaraShafieeDTUManagementEngineeringTechnicalUniversityofDenmarkDivisionofManagementScience,OperationsManagementGroupProductionstorvetBuilding436-130FDK-2800Kgs.LyngbyDenmark

SupervisorLarsHvam,Professor,PhDDivisionofManagementScienceDepartmentofManagementEngineeringTechnicalUniversityofDenmark

Co-supervisorNielsHenrikMortensen,Professor,PhDDivisionofEngineeringDesignandProductDevelopmentDepartmentofMechanicalEngineeringTechnicalUniversityofDenmarkAssessmentCommittee

PeterJacobsen,AssociateProfessorDepartmentofManagementEngineeringTechnicalUniversityofDenmarkÉliseVareilles,AssociateProfessorORKIDTeamleaderIndustrialEngineeringLab,Écoledesminesd’Albi-CarmauxMartinMalis,PhD,ProgramManagerITOperationServicesNNIT

ii

iii

ABSTRACTIndividual customization of goods and processes in different industries leads tocomplexitydue to agrowingmixofproductsboth regardingcharacteristicsofproductsandsupportservices.Inordertoeliminatecomplexityandchallengesinproduct/processcustomizing, smart IT systems called Product Configuration Systems (PCS), have beenproposedas thesolutionbothbyresearchersandpractitionersandvariousbenefitsarementioned from utilizing PCSs. Based on the latest literature, there are challengesreported in all phases of PCS projects including planning, development, anddocumentation. Moreover, the challenges become more serious when it involvescomplicatedproducts/processes in engineer-to-order (ETO) companies.Thepurposeofthis thesis is to contribute to the existing knowledge of managing PCS projects byproposingframeworksandtoolstoaddresssomeofthemainchallenges.

First,thisresearchfocusesonthereportedbenefitsandchallenges indifferentphasesofPCSprojectsalignedwiththegaps inthecurrent literature.Second,thestudypresentsasurvey in order to have a comprehensive overview to assess the most importantchallengesinthearea.Third,inordertoovercomedifferentchallengesinthePCSprojects,the study contributes to the literature in forms of different frameworks, tools and ITsolutions.Addressingthedefinedchallenges,thefollowingframeworksareproposed.1)AframeworkisprovidedforbusinesscasesinPCSprojectsinordertoestimatetheneededinvestmentsandfinancialreturn-on-investment.2)Furthermore,theresearchproposesaframework and different tools to scope the whole PCS project from planning to themaintenance phase. 3) Afterwards, the study suggests a framework to manage theknowledge inPCSprojectsduetoreportedchallenges.4)Inordertomake itpossibletomodel, maintain, communicate, and document complicated products/process, aframeworkalignedwithan IT tool isdeveloped inclosecollaborationwith industry.5)Finally,thestudycontributestothedirectionofintegrationofPCSandotherITsystemsbyshowingtheautomationimpactofthisalignment.

Thetoolsandframeworksdevelopedhavebeenevaluatedbasedonexistingliteratureandbyempirical tests incompanies.Furthermore,areas for further investigationhavebeenidentified.

Conceptual Modelling for Product Configuration Systems – Sara Shafiee

iv

DANSKRESUME

Individuel kundetilpasning af produkter og processer i forskellige brancher fører tilstigendekompleksitetpå grund af et stadig størreprodukt sortiment,bådehvad angårprodukterneskarakteristikaogservice-support.Foratfjernekompleksitetenogimødegåudfordringernevedprodukt/proces-kundetilpasningharbådeforskereogpraktikeresomløsning foreslået intelligente IT-systemer kaldet ProduktKonfigurering systemer (PCS).FordelevedatbenyttePCSerdiskuteretilitteraturen.Dennyestelitteraturfremhæveratder er udfordringer relateret til alle faser af PCS-projekter, herunder planlægning,udviklingogdokumentation.Endviderebliverudfordringernestørre,nårderer taleomkomplicerede produkter/processer i ordre initierede ingeniør virksomheder (ETO).FormåletmeddenneafhandlingeratbidragetildeneksisterendevidenomledelseafPCS-projektervedatforeslåmetoderogværktøjer,derkanhjælpemedat løsenogleafdisseudfordringer.

Først,fokusererdenneafhandlingpådemetoder,deranvendessamtudfordringerveddeforskelligefaserafPCS-projekterset i forholdtilmanglerne ideneksisterende litteratur.Dernæstpræsenteresenspørgeskemaundersøgelse,hvisformåleratgiveoverblikoverogidentificeredevigtigsteudfordringer,dererpåområdet.Fordettredjebidragerstudiettil litteraturen i form afmetoder, værktøjer og IT-løsningermed formål at afhjælpe deforskellige udfordringer, som man har ved PCS projekter. Følgende værktøjer erpræsenteret:1)EnmetodeforudviklingafbusinesscasesiPCS-projektertilvurderingafde nødvendige investeringer og økonomiske investeringsafkast. 2) Derudover foreslårafhandlingenenmetodeog forskelligeværktøjer tilvurderingafprojektafgrænsningafhelePCS-projektet fraplanlægningsfasen tilvedligeholdelsesfasen.3)Dernæst foreslårstudietenmetodeforviden-styringiPCSprojekterpåbaggrundafkendteudfordringer.4)For at kunne modellere, vedligeholde, kommunikere og dokumentere kompliceredeprodukter/processererderitætsamarbejdemederhvervslivetudvikletenmetodetilatdokumentere modellerne, der er implementeret i et IT-værktøj. 5) Endelig bidragerstudiet til at give retningslinjer for integrering afPCSog andre IT-systemerved atviseautomatiseringseffektenafdennetilpasning.

De metoder, værktøjer og IT-løsninger, der er udviklet til at styre PCS projekter erevalueret i forhold til eksisterende litteratur og ved empiriske test i virksomheder.Desudenerderidentificeretområdertilyderligereundersøgelse.

v

ACKNOWLEDGEMENTSDuringthesethreeyearsofmyPhDjourney,Ihavebeengivenuniqueopportunitiestakenadvantageofthemandencounteredmanychallenges,pleasantmomentsandproblemsinthetimeIwasinvolvedwiththisproject.Atanyrate,Ihavefinished,butnotalone.IcouldnothavesucceededwithouttheinvaluablesupportofseveralremarkableindividualswhoIwouldliketoexpressmysinceregratitude.

First of all, I would like to thank Prof. Lars Hvam as my academic supervisor for hiscontinuous support,his inspiration andhis immenseknowledge.Ihavebeen extremelyluckytohaveasupervisorwhocaredsomuchaboutmywork,andwhorespondedtomyquestionsandqueriessopromptly.Iwould like to thankmyco-supervisorNielsHenrikMortensenforalltheguidanceandinsightfuldiscussionsandsuggestions.IhavetothankAndersHaug,CiprianoForza andAlexanderFelfernigwho taught and supportedme intheoreticalcontributions.

Ialsoappreciate thesupportofmy industrialsupervisors,PeterBoElverkovandSannaTinaKaarinaStrand.IamalsogratefulforyoureffortsinputtingourITdevelopmentandmanagement concepts intopractice.ThomasChristensen and JesperRiis, thankyou forproviding me with the opportunity to conduct this study in your teams, and for thesupportinthecourseofthestudy.Iwouldhavebeenlostwithoutyou.

Iwould like toexpressmydeepestgratitudetothewhole community ofHaldorTopsoeand all my colleagues in corporate IT and business. It was wonderful four years ofpricelessexperienceandvaluablecollaborationwithyou. Iwould liketothankthemanypeoplefromthecasecompanieswhodidnothesitatetohelpmeandsharetheknowledgewithme.Next, I am particularly indebted my colleagues, Anna Myrodia, Katrin Kristjandsottir,Martin Bonev, Sara Markworth Johnsen, Fatemeh Rahimi, Alexandria Moseley, ManuelKorell, and Daniel Sepúlveda Estay. Their scientific and personal support made thisjourneyevenmoreenjoyable. Ialsoappreciate theopportunityofbeing aresearcher inTechnicalUniversityofDenmarkandManagementEngineeringDepartment.

Iwouldliketothankmyfamilyforalltheirloveandencouragement.Formyparentswhoraisedmewithaloveofscienceandsupportedmeinallmypursuits.Formysisters,MitraandSaeedeh,andmystepbrotherAfshinwhowerealwaysthesourcesofinspiration.

Lastly,mydeepestthanksgoestomybestfriend,soul-mate,andhusband. Imarriedthebestpersonoutthereformesixyearsago.TherearenowordstoconveyhowmuchIlovehim.Alirezahasbeenatrueandgreatsupporterandhasunconditionallylovedmeduringmygoodandbadtimes.Itrulythankyouforstickingbymyside,evenwhenIwasirritableand depressed. I feel thatwhatwe both learned a lot about life and strengthened ourcommitmentanddeterminationtoeachotherandtolivelifetothefullest.Thankyou.


vi

CONTENTS

1 introduction............................................................................................................................................................1

1.1 Productcustomization............................................................................................................................1

1.2 Challengesofproductcustomization..............................................................................................2

1.2.1 Productconfiguration:asmartITsolution........................................................................2

1.3 Problemstatement...................................................................................................................................4

1.3.1 Challengesofproductconfigurationprojects..................................................................5

1.4 Aimandscopeoftheresearch............................................................................................................7

1.5 Structureofthethesis.............................................................................................................................8

2 Researchdesign.................................................................................................................................................11

2.1 Researchobjectivesandresearchquestions...........................................................................11

2.2 Researchmethodology........................................................................................................................15

2.2.1 Basicscienceresearch...............................................................................................................15

2.2.2 Designresearchmethodology...............................................................................................16

2.3 Researchplan...........................................................................................................................................18

2.3.1 Researchstages.............................................................................................................................18

2.3.2 Casestudies.....................................................................................................................................19

2.3.3 Surveys...............................................................................................................................................21

2.3.4 Communicationoftheresearch............................................................................................22

2.3.5 Schedulingtheresearch............................................................................................................25

2.4 Chaptersummary...................................................................................................................................26

3 TheoreticalFraming........................................................................................................................................28

3.1 DifferencesbetweenITandPCSprojects..................................................................................28

3.2 StrategiesforPCSProjects.................................................................................................................30

3.3 BenefitsfromusingPCS......................................................................................................................31

3.4 BusinesscasesinPCSandavailabletools..................................................................................32

3.5 ScopinginPCSprojects.......................................................................................................................33

3.6 KnowledgemanagementinPCSprojects...................................................................................34

3.7 ModellinginPCSprojects...................................................................................................................36

3.7.1 Documentation,maintenance,andcommunicationinPCSprojects.................39

3.8 PCSprojectandintegrationswithotherITsystems............................................................40

3.9 Chaptersummary...................................................................................................................................42

vii

4 Researchsummary..........................................................................................................................................43

4.1 Publicationsoverview..........................................................................................................................44

4.2 Planningproductconfigurationsystems...................................................................................46

4.2.1 Paper A: How to Identify Possible Applications of Product ConfigurationSystemsinEngineer-to-OrderCompanies..........................................................................................46

4.2.2 Paper B: Economic Value Creation from Using Product ConfigurationSystems–ACaseStudy.................................................................................................................................50

4.2.3 Paper C: Industrial application of configurators: From motivations torealizedbenefits................................................................................................................................................52

4.2.4 PaperD:Themainchallengesformanufacturingcompaniesinimplementingandutilizingconfigurators..........................................................................................................................55

4.2.5 PaperE:HowtoframebusinessCasesFrameworkforProductConfigurationsystemProjects..................................................................................................................................................59

4.3 Scopingofproductconfigurationsystems................................................................................64

4.3.1 Paper F: Scoping a Product Configuration Project for Engineer-to-OrderCompanies............................................................................................................................................................64

4.4 Knowledgemanagement....................................................................................................................72

4.4.1 PaperG:How to scope configurationprojects andmanage theknowledgetheyrequire.........................................................................................................................................................72

4.5 Modellinganddocumentation.........................................................................................................77

4.5.1 Paper H: The documentation of product configuration systems: AframeworkandanITsolution...................................................................................................................77

4.5.2 PaperI:Productconfigurators:Astudyofthe impactofapplyingITformalproductmodellingtechniques...................................................................................................................83

4.6 PCSsinintegrationwithotherITsystems................................................................................86

4.6.1 PaperJ:AutomaticIdentificationofSimilaritiesacrossProductstoImprovetheConfigurationProcessinETOCompanies...................................................................................86

4.6.2 PaperK: Improvedperformance andqualityof configuratorsby receivingreal-timeinformationfromsuppliers....................................................................................................89

4.7 Chaptersummary...................................................................................................................................92

5 Conclusions..........................................................................................................................................................94

5.1 Contributiontotheory.........................................................................................................................94

5.2 ContributiontoPractice...................................................................................................................101

5.3 Researchlimitations..........................................................................................................................104

5.4 Opportunitiesforfuturestudies..................................................................................................104

5.5 Concludingremarks...........................................................................................................................106

6 References.........................................................................................................................................................107


viii

7 Appendedpapers...........................................................................................................................................118

ArticleA.........................................................................................................................................................................119

ArticleB.........................................................................................................................................................................120

ArticleC.........................................................................................................................................................................121

ArticleD........................................................................................................................................................................122

ArticleE.........................................................................................................................................................................123

ArticleF.........................................................................................................................................................................124

ArticleG.........................................................................................................................................................................125

ArticleH........................................................................................................................................................................126

ArticleI..........................................................................................................................................................................127

ArticleJ..........................................................................................................................................................................128

ArticleK........................................................................................................................................................................129

ix

LISTOFTABLESTable1-1Productcustomizationchallengesandtherealizedbenefitsfromusingproductconfigurationtechnology(AlexanderFelfernigetal.,2014)..................................................................3Table1-2ThecategoriesofthechallengesreportedforPCSprojects.............................................6Table2-1Overviewofthecasecompanies...................................................................................................20Table2-2Overviewofpublications..................................................................................................................22Table2-3GantttableofPhDschedule.............................................................................................................26Table3-1ApproachestoPCSscoping..............................................................................................................34Table3-2LiteraturebaseforthemainstepsintheKMinPCS..........................................................35Table3-3literatureresultsformodellingtechniques............................................................................38Table4-1Overviewofpublications..................................................................................................................45Table4-2ROIfromapplyingPCSoverafive-yearperiod.....................................................................51Table4-3Percentagesofcompaniesexpressing(oneormore)challengesgroupedintothecategories........................................................................................................................................................................56Table4-4Theimportanceofthemaincategoriesofchallenges........................................................57Table4-5PriorityTableexamplebasedonWeigers(1999)...............................................................71Table4-6KMframeworksforITprojects.....................................................................................................73Table 4-7 Comparison of companies using the CPM procedure and less formal or nomodelling techniquesaccording to theirability todocumentproductknowledge,controlproductvariants,qualityandemployeesatisfaction...............................................................................84Table5-1PhDstudy’scontributiontopracticeandtheoreticalchallenges.............................101


x

LISTOFFIGURESFigure1-1PCSandautomationofthesalesandengineeringprocess..............................................3Figure 1-2The object-oriented project life cycle, adjusted from (Booch,Rumbaugh andJacobson,1999)..............................................................................................................................................................5Figure1-3Structureofthethesis..........................................................................................................................9Figure2-1ResearchobjectivesforselectedphasesofPCSprojects................................................12Figure2-2Differentformsofengagedscholarship(VandeVen,2015)........................................16Figure 2-3 The design research methodology (DRM) adjusted from (Blessing andChakrabarti,2009).....................................................................................................................................................18Figure2-4Linkbetweenpapersandresearchquestions......................................................................25Figure3-1DifferentphaseofPCSprojectsanddifferencesfromITprojects.............................29Figure3-2FourbasicrepresentationsofproductmodellingforPCS(DuffyandAndreasen,1995).................................................................................................................................................................................40Figure4-1TheproposedframeworktoidentifyapplicationsofPCS.............................................47Figure 4-2 The interaction between sales and technical product configuration systems(PCS)..................................................................................................................................................................................48Figure4-3Achievingsynergybygeneratingdocumentsusing information fromproductconfigurationsystems(PCS)acrossdepartments.....................................................................................48Figure 4-4 Simplified overview how the sales and engineering processes could besupportedbyproductconfigurationsystems(PCS)andinteractionwithotherITsystems...............................................................................................................................................................................................49Figure4-5UnifiedProcess(Kruchten,2007)..............................................................................................66Figure4-6ThegeneraliterationprocesstemplateforPCSs................................................................67Figure4-7TO-BEprocessexample...................................................................................................................68Figure4-8Usecasediagramexample..............................................................................................................69Figure4-9Themodelsubjectlayerandintegrationwithothersystems.....................................70Figure4-10TheinitialPVMstructureexample..........................................................................................70Figure4-11TheproposedframeworkforknowledgecollectioninPCSs.....................................74Figure4-12.Anexampleofausecasediagram(Hvametal.,2008)...............................................74Figure4-13MethodologyfordevelopingtheframeworkandtheIT-baseddocumentationsystem...............................................................................................................................................................................78Figure4-14Theprocessofgeneratingdocumentation from thePCS (thephotoswillbeshowninthecasestudy).........................................................................................................................................79Figure 4-15 The relations between the product model and common structure incommercial PCSs. PVM and CRC Cards structure (a & b) (Mortensen, et al., 2008);traditionalstructureavailableincommercialPCSs(c)..........................................................................80Figure4-16 Interactionsbetween theContentManagementSystemand theUIRenderer(documentationsystem).........................................................................................................................................81Figure 4-17 Knowledge generation in the PCS and PVM/CRC structure inside the ITdocumentationsystem.............................................................................................................................................82Figure4-18Structureandinformationflowofthecomparisonsysteminthecasecompany..............................................................................................................................................................................................87Figure4-19Finaluserinterfaceofthecomparisonsystem.................................................................88

xi

Figure4-20Thetechnicalsetupatthecasecompany:thesupplierintegrationviaAPIwebservices............................................................................................................................................................................90Figure5-1IndividualphasesofPCSprojectsandthesiscontribution...........................................95


xii

LISTOFABBREVIATIONSANDACRONYMS

BOM Billofmaterial

BC Businesscase

CAD ComputerAidedDesign

CRCCard Class,ResponsibilitiesandCollaborationCards

DRM Designresearchmethodology

ERP Enterpriseresourceplanning

ETO Engineer-to-order

IT InformationTechnology

KM KnowledgeManagement

MC MassCustomization

PCSProductConfigurationSystem

PVM ProductVariantMaster

ROI ReturnOnInvestment

RUPRationalUnifiedProcess

RQ ResearchQuestion

UML UnifiedModellingLanguage

1

1INTRODUCTION

InordertodealwiththechallengesfromMassCustomization(MC),literatureisconcernedwithdifferent implementationmethods ofMC (Blecker et al., 2004).Other researcherssuggest to facilitate modelling methods and dedicated IT systems to increase theeffectivenessandefficiencyofthespecificationprocess(Hvametal.,2006;SalvadorandForza,2004).Withtheadvancementofinformationtechnology(IT)systems,thetrendfordiversificationisreinforcedbytheabilitytocustomizeindividualaspectsonproductsinamore efficient manner and thus became an important aspect for today’s competitivemarket (Pine, 1993). Product Configuration Systems (PCS) are attracting attentions tosupportthetasks involved inthecustomer-orientedcustomizedprocessesrelatedtothespecification of products (Hvam et al., 2008). There are, however, challenges reportedregarding planning, developing, and maintaining PCSs. This PhD thesis focuses onchallenges indifferentphases inPCSprojectsandproposesframeworksandITsolutionstoovercomesomeofthemainchallengesindifferentphasesofPCSprojects.

Theremainderofthischapterisasfollowing.Insection1.1and1.2theconceptofproductcustomizationandthesignificanceofthesupportofITsystemsforproductcustomizationin industry are elaborated.Moreover, the reasons for solving thementioned challengesconcernedwithproductcustomizationwithuseofPCSaresupportedwith theevidencefromcurrentliterature.Insection1.3,thegapintheliteratureishighlightedregardingthemain challenges in different phases of PCS projects that highlight the need for morestandardizedtheframeworksforPCSprojects.TheaimsandscopeofthePhDresearchisexplained in section 1.4 while section 1.5 illustrates the expected structure of thefollowingreport.

1.1ProductcustomizationProduct customization or personalization has recently replaced the mass productionmarket(Kotler,1989).Customizationofproductsistheconceptofconfiguringproductsbyselecting feasible combinations from predefined compositions connected with rules


2

(ElMaraghyetal.,2013).Thedemand towardhigherproductvarietyandcustomizationhasbeen increasingevenfurther inrecentyears.Industriestrytocompetebyprovidingmoreoptionsfortheircustomersinordertoestablishstrongereconomicbenefits(Piller,2004).On theotherhand,customersvalueestablishing theiruniqueproductsbasedontheir own preference (Schreier and Prügl, 2011). Apart from offering a wide productvariety, to react in minimum time to deliver the requirements is of great importance(SalvadorandForza,2004).Thequalityofthedeliverableproductshouldnotbeundertheinfluenceoftimeorvariantchoiceintheproductevenforcomplicatedproducts(Trentinetal.,2012a).

1.2ChallengesofproductcustomizationTheincreaseinvarietyanditisparticularlyevidentontheexampleofengineer-to-order(ETO) companies leads to an increase in complexity range and product variety. ETOcompaniescorebusinessistocreatehighlyengineeredcomplexproductvariantsthatareengineeredtothespecificrequirementsofacustomer(WiknerandRudberg,2005).ETOvariants are generally reflecting a particular highly individual customer requirement.Managing variety throughout the entire products supply chain is one of the majorchallenges in customizing the products. Grouping of similar variants into families is afundamentalenablerofdesigning,planningandproducingvariantsefficiently(ElMaraghyeta.2013).Productfamiliescontainvariantsoftheproductsandtheirparts,componentsandconfigurations(ElMaraghy2009;ElMaraghy2007)inordertobettermeetthediverseneeds of today’s highly competitive globalmarketplace, increase variety, shorten lead-times,andreducecosts(Kumaretal.,2009).OneofthehighlyrecommendedsolutionsisPCSasexpertsystemsthatsupportproductcustomizationbydescribinghowpredefinedentities (physical or non-physical) and their properties (fixed or variable) can becombined(Hvametal.,2008).ThenextsectionintroducesPCSasanITsmartsolutiontosolvethechallengesinproductcustomizationinETOcompanies.

1.2.1 Productconfiguration:asmartITsolutionPCScanbeappliedtosupportthedecision-makingprocessesinthesalesandengineeringphasesofaproduct,wherethemost importantdecisionsregardingproductfeaturesandproduct costs aremade (Hvam et al., 2008).PCS enable companies todevelopproductalternativestofacilitatethesalesandproductionprocesses(Felfernig,Hotz,etal.,2014)by incorporating information about product features, product structure, productionprocesses,costsandprices(ForzaandSalvador,2007a).This information ismodelled inPCSduringtheirimplementation(ForzaandSalvador,2007a).Automationofengineeringprocesses is increasingly prevalent in multiple lifecycle phases such as design,manufacturing and service support (Verhagen et al., 2015). The PCS receives theindividualcustomerrequirements,configures theproductbasedon thedefinedsolutionspace,performsthecalculationsandfinallygeneratesallthenecessaryoutputsincluding:proposals, bills ofmaterial, price calculation sheets or even CAD drawings (see Figure1-1).Widelyusedinvariousindustries,PCScanbringsubstantialbenefits,suchasshorterlead times forgeneratingquotations, fewererrors, increasedability tomeetcustomers’requirements regarding product functionality, the use of fewer resources, optimised

3

productdesigns,lessroutineworkandimprovedon-timedelivery(Ardissonoetal.2003;Barkeretal.1989;Hvametal.2006;Petersen2007;Forza&Salvador2006).

Figure1-1PCSandautomationofthesalesandengineeringprocess

The key points to successfully deploy configuration technologies are: 1) Avoidingerroneous and suboptimal offers, 2) Avoiding MC, 3) Complexity handling of needselicitation,4)Knowledge integration,and5)Efficientknowledgemanagement(Felfernig,Hotz, et al., 2014). In Table 1-1, themain challenges companies implementing amasscustomizationstrategiesfacearelistedalongwithhowPCSscanhelpcompaniestosolvethosechallenges(Felfernig,Hotz,etal.,2014).

Table1-1Productcustomizationchallengesandtherealizedbenefitsfromusingproductconfigurationtechnology(Felfernig,Hotz,etal.,2014).

MCchallengesinETOcompanies HowPCSscanaddressthechallengesofMC

1)AvoidingErroneousandSuboptimalOffers(Heiskalaetal.2009;Barkeretal.1989;Haugetal.2011)

Preventfire-fightinginmanufacturingfromerrors in the sales specification, shorterlead times needed for specifications,increasedsale, lowerpersonnelcost,stableeffort estimation, improved and moreuniform quality, stable schedules, lesscustomer complaints, less complexity inthe sales process, reduction in productmodelobsolescence,savecustomertime inspecification

2) Mass Confusion (Heiskala et al. 2009(Huffman&Kahn1998;JuhaTiihonenetal.1996; Von Hippel 2001; Felfernig et al.2006)

Customerparticipationindesign,Corporatememory, Finding right amount of offeredcustomization

3)ComplexityHandlingofNeedsElicitation(Heiskalaetal.2009;Ardissonoetal.2003;Tiihonenetal.1998a;Tiihonen&Felfernig2010;Falkneretal.2011)

Increased sales, Preference knowledge,Open innovation, Reduction in inventions,Componentsharingacrossproductlines

4) Knowledge Integration (Liliana Reliable demand prediction, Decreased


4

Ardissono et al. 2003; JuhaTiihonen et al.1996;Tiihonenetal.1998a)

timetomarket,Lesserrors

5)EfficientKnowledgeManagement(Mittal& Frayman 1989; Heiskala et al. 2009;Heiskalaetal.2009;Felfernigetal.2006)

Efficient development, Efficientmaintenance, Extent of organizational andculturalchanges,Longtermmanagementofdelivered product individuals,reconfiguration, Configuration knowledgechangesfrequently

Challenge 1:Avoiding Erroneous and suboptimal offers.Offering highly variant productsrequires sales need time to understand the product structure. The result of beingconfrontedwithhighvariabilityareerrors thatareoftennotnoticedbeforeproduction(Heiskalaetal.,2007).Configurationtechnologiescantacklethischallengebyconstrainttechnologies thatsupportautomated feasibilitychecksofcustomer requirementswhichleadstoshorterleadtime,increasedsales,lessresources,andstableschedule.

Challenge2:AvoidingMassConfusion.Largeandcomplexproductstructure lead tomassconfusion(HuffmanandKahn,1998)whichmakesthecustomerconfusedbytheselectionand finallyunable tomake adecision.PCSscanprovide a search interface thathelps tonarrowdownthenumberofrelevantoptions.Personalizationtechnologiessupportmoreefficientdecisionprocesses(Felfernigetal.,2006).

Challenge3:ComplexityHandlingofNeedsElicitation.Itisdifficulttoidentifytherelevantconfigurationwithout a clear view of the users’ preferences (Falkner, Felfernig, et al.,2011).UserinterfacesinPCSssupportcustomersinfindingtherightconfigurationwhichleadstoincreaseinsaleanddeepunderstandingofmarketrequirements.

Challenge4:KnowledgeIntegration.Businessesarestrengtheningthecommunicationandknowledge sharing with customers. PCSs can be integrated into a larger system todecrease the time to market, reduce order and processing errors, and increaseprofitability.

Challenge 5: Efficient Knowledge Management. The products knowledge may exist in avariety of sources including product data management systems, spreadsheets, legacyenterpriseandCRMsystems,oreven intheheadsofsoon-to-retireemployees(Felfernigetal.2014).PCScentralizeandmanagealltheknowledge.

1.3ProblemstatementOver the previous years, PCS project applications in different kinds of industries isinvestigated through surveys and case studies research. The outcomes reveal benefitsregardingsavedmanhours,costsandetc. indifferentcompanies(Ardissonoetal.2003;Barker et al. 1989; Hvam et al. 2006; Petersen 2007; Forza & Salvador 2006).Furthermore, the results demonstrate the reported challenges during the planning,developmentandmaintenance;whichleadsinreducingtheperformanceandefficiencyorevenmaketheprojecttobeafailure.However,thechallengesrelatedtoutilizingPCSsincompanieshavenotbeenaddressedintheliteraturetothesameextentasthebenefits,as

5

there is a tendency tohighlight the successof the implementations (Haug et al. 2012).However, the challenges should get more attention both from practitioners and theresearchcommunityasthereisaneedforthemostefficientframeworkstoaddresssomeofthemainchallengesinPCSprojects.

Thecurrentliteraturedemonstratesthereportedchallengesduringtheplanning,developmentandmaintenancephasesofPCSprojects;whichresultinreducingthe

performanceandefficiencyofthePCSprojectsorevenmakestheprojecttobeafailure.

1.3.1 ChallengesofproductconfigurationprojectsThere are challenges reported in different studies regarding planning, developing,implementing and maintaining PCS projects. In Table 1-2, the challenges are groupedbased on literature and experiences working with PCS projects. Each of the group isrelated to different phase of PCS projects. The different phases of PCS determined as:Planning, Development, Implementation, and Maintenance (Figure 1-2) (Booch et al.,1999)(Hvametal.,2008).

Figure1-2Theobject-orientedprojectlifecycle,adjustedfrom(Boochetal.,1999)

Barkeretal. (1989)presented thechallengesascross-functionalbusinessneeds,as theimplementation of the PCS enhances changes in the business processes that requiresupportfromtopmanagement.Tiihonenetal.(1996)concludedthatinadequatetoolsandmethods for modelling account for the main challenges of managing the PCS models.Tiihonenetal.(1998a)describedthemainchallengesintermsofmanagingtheknowledgeinlong-termPCSprojectstoensureitiskeptuptodate,whichrequirestoolstomatchtheknowledgeembedded in thePCSof thecompanies’productsandprocesses.ArianoandDagnino’s(1996)performedastudywherethemainchallengesweredescribed intermsof the few numbers of employees who understood the system. This situation causeddifficultieswhen theonlyemployee thathad a fullunderstandingof thesystem left thecompany soon after thedevelopment.Blecker et al. (2004)described thatPCSsdonotsupportthefront-endsufficientlyandemphasizedtheneedforPCSstosupportcustomersintheconfigurationprocessesandtofindanoptimalsolutionthatfulfilstherequirements.


6

Felfernigetal.(2000)explainedthatthecomplexityinthesoftwaredevelopmentofPCSsrequires highly technical knowledge from experts and the knowledge base has to beadaptedcontinuouslybecauseofthechangingcomponentsandconfigurationconstraints.However, the development and the maintenance time for PCSs is short and strict.Aldanondo et al. (2000) explained the challenges regarding the two different kinds ofrequiredknowledgetodevelopthePCS: industrialexpertiseandconfigurationexpertise.ForzaandSalvador(2002)explainedthemainchallengesofimplementingaPCSinasmallmanufacturing company in terms of product modelling. In another study, Forza andSalvador(2002)describedthemainchallengesofimplementingaPCSintermsofpersonalrolechanges,inter-functioncollaborationandsoftwarepersonalization.Forzaetal.(2006)describedthatforhighlycomplexproductsconsistingofaverylargesolutionspace,itmaynotbe economically feasible to implement aPCS.Ardissono et al. (2003) identified themainchallengesexperiencedwithPCSsintermsofincreasedcomplexityofproductsandservicesoffered,whichresults in increasedcomplexityofthesystems,making itdifficultfor theend-user tousePCSdue to a lackof technicalknowledge.Heiskalaetal. (2005)identifiedthechallengesrelatedtoPCSswithafocusonservices.Basedontheliterature,theydescribedchallengeswhichareconfirmed/disconfirmedinrelationtoservicesbasedon two case studies. Heiskala et al. (2007) described the challenges relating to theintroduction and long-term management of PCS based on the previous literatureaddressing PCSs. Hvam et al. (2006) described the challenges in terms of knowledgeacquisition and product modelling for complex products in PCS and communicationdifficultiesbetweendomainandconfigurationexperts.Inanotherstudy,HaugandHvam(2007) expressed that documentation of PCSs is often not maintained after the PCSbecomes operational as the documentation process is too time-consuming. Petersen(2007)foundthemainchallengesofimplementingaPCSinengineeringcompaniestobethe product characteristics, customer relations and long-term projects. Table 1-2summarizes all the related literature to challenges while planning, developing,implementing and maintaining PCS projects. This table reveals the nature of differentchallengestoseparatethetechnicalissuesandmanagementchallenges.

Table1-2ThecategoriesofthechallengesreportedforPCSprojects

RelatedPhasesofPCSprojects

ChallengesinrelationwithmanagementofPCS

Natureofchallenge Maincontributions

Planning Resourceconstraints

Lack of personnel formodelling the PCS and forgathering and providinginformation.

(Aldanondo et al. 2000;Ariano & Dagnino 1996;Barker et al. 1989; Forza &Salvador 2002b; Haug et al.2012; Heiskala et al. 2005;Heiskalaetal.2007)

Planning Product-relatedchallenges

Challenges in the productassortment that arecommonly described incomplexity of the productstructure.

(Petersen 2007a; Aldanondoet al. 2000; Barker et al.1989; Forza & Salvador2002b; Forza & Salvador2002a; Forza et al. 2006;Heiskalaetal.2005;Heiskalaetal.2007)

7

Development IT(technical)challenges

All technical challengesrelated to IT systems (e.g.integration,UIs).

(Tiihonen et al. 1998b;Ardissonoetal.2003;Barkeret al. 1989; Blecker et al.2004; Felfernig et al. 2000;Cipriano Forza & Salvador2002; Heiskala et al. 2007;Tiihonen et al. 1996;Tiihonenetal.1998a)

Development,maintenance

Knowledgeacquisitionchallenges

The process of gatheringknowledge in thedevelopment andmaintenancephases,andtheavailabilityofinformation.

(Aldanondo et al. 2000;Ardissono et al. 2003;Felfernigetal.2000;Haug&Hvam 2007; Heiskala et al.2005; Heiskala et al. 2007;Hvam et al. 2006; Tiihonenet al. 1996; Tiihonen et al.1998b)

Development,maintenance

Productmodelling(knowledgerepresentation)challenges

Challenges related tostructuring the productmodel and reflecting theknowledge to be embeddedinthePCS.

(Aldanondo et al. 2000;Ardissono et al. 2003;Felfernigetal.2000;C.Forza& Salvador 2002; CiprianoForza&Salvador2002;Haug&Hvam2007;Heiskalaetal.2007; Hvam et al. 2006;Tiihonen et al. 1996;Tiihonenetal.1998)

Implementation Organizationalchallenges

Lack of support frommanagement and resistancetochange.

(Ariano and Dagnino, 1996;Barkeretal.,1989;ForzaandSalvador,2002b;Haugetal.,2012a;Heiskalaetal.,2007;Hvam et al.,2006;Tiihonen,Soininen,etal.,1998)

Basedon the literature, it canbe concluded that there arenumerous challenges in thedifferentphasesofthePCSprojects.Thishighlightstheneedforframeworks,inordertosimplify theprocessofPCSprojects.This research suggests frameworks to enableETOcompaniesandpractitionerstoincreasethebenefitsfromusingPCS.

1.4AimandscopeoftheresearchThefocusofthisPhDthesisistosolvesomeofthemainchallengesinthedifferentphasesof PCS projects. PCS projects are divided to 4 different phases listed as: Planning,Development,Implementation,andDocumentation(Boochetal.,1999;Hvametal.,2008).There arechallenges related todifferentphasesofPCSprojects reported in thecurrentliterature(section1.3).While,theliteratureprovidestools,methodsandframework,therearestillundersolvedchallengesreportedfrombothacademiaandindustry.ByaddressingthesechallengesinPCSprojects,man-hours,leadtimesandcostcanbedecreasedduringthe project planning, development aswell as after the implementation inmaintenancephaseandfurtherdevelopments.

BasedonliteratureandduetothecomplexityofPCSprojects,therearerequirementsforstandardwaysofmanagingPCSprojectsalignwiththetoolsandtechniquestoovercomeprojects challenges. By addressing the issues and challenges related to planning,


8

developmentandmaintenanceofPCS,solutionsincludingtools,frameworksandmethodscanbedevelopedtoimprovethecurrentsituation.

ResearchObjective:

ToframeandimprovethemanagementprocessofPCSprojects:planning,development,andmaintenancephases

Inorder toprovide an answer, the researchobjectiveof this thesis isdivided into fiveresearchquestions (RQs) andwillbe elaborated in Section2.Themain aspectsofPCSprojectsarethefollowing:

- Structuredbusinesscasesbeforeplanningtheproject,motivations,benefitsandchallengesinPCSmanagement,

- ScopingPCSandhavingastructuredplanningphase,- ManagingknowledgeinPCS,- Modelling,documentationandmaintenanceofthePCSprojects,- PCSintegrationwithotherITsystems

1.5StructureofthethesisIn Figure 1-3 the structure of the thesis is illustrated.The same colour-coding is usedthroughoutthereportinordertoconnecttherelevanttopicsandRQs.

TheIntroductionchapterprovidesanoverviewofthethesis-whyweneedthisresearchprojectandwhatare thegoalsregarding thementionedchallenges.Basedon literature,this chapter provides a summarized overview on product customization and thechallenges, industries face regardingproductspecifications.Then, thesolution for thesechallenges issuggestedas the ITsmartsystems,whichcan limit thesolutionspaceandautomatically manage the specification process. First, relevant studies are presenteddemonstratinghowPCScansolvethereportedchallengesofproductcustomization.Next,theproblem statementdescribes the challenges available in thewhole life cycleofPCSprojectsfromplanningtodevelopment,andmaintenance.ThePhDthesiscontribution isconcernedtoproposenewsolutionsandframeworktoaddressthereportedchallengesofPCSprojects.

Chapter 2 explains the researchdesign, introduces theparticipants and theproceduresusedintheresearchdesign.Furthermore,thechapterexplaintheresearchobjectivesandresearchquestionsanddifferentstagesoftheresearchwhichisthestructureofthethesis.

ThisisfollowedbytheliteraturereviewinChapter3,whichissplitacrosseightmainsubsections.The firstpartconsistsof a reviewofdifferencesandsimilaritiesofPCsand ITprojects.Thesecondand thirdsection includes the reviewof theavailable literatureonPCS development strategies and reported benefits which lead to have a formalizedBusiness Case (BC) for PCS projects in section 3.4.This section isfollowed by anothersectionwhich isthenextphaseofPCSprojectsanddescribesthe literatureavailableforscopingPCSprojects.The section3.6ofChapter 2provides anoverviewofKnowledge

9

Management(KM)inforITsystemsisprovided.Section3.7discussesandelaboratestheavailable modelling and documentation techniques for PCS projects. Finally, the lastsectionof thischapter includesareviewofstudiesabout the integrationofPCsprojectswithotherITsystemsacrossthesupplychain.

Chapter 4 gives a broad outline of the procedures used for the data analysis in elevenarticlesincludedinthisthesis.ThischapteremphasizesandanswerstheRQsprovidedinChapter2elaborately.Eachofthemainarticlesaresummarizedanddiscussed.

Chapter5concludesanddiscussesthe thesisbysummarizing theresults,discussing thetheoretical contributions of the study, revisiting the research hypotheses, suggestingpedagogical implications, noting the limitations, and suggesting potential avenues forfurtherresearch.

Figure1-3Structureofthethesis


10

11

2RESEARCHDESIGN

2.1ResearchobjectivesandresearchquestionsAs discussed in the previous section, the market is moving towards the customizedproductsandcustomerstendtoorderpersonalizedproducts.Theresultsof theproductcustomization are complexity in product families, confusion in processes, knowledgemanagement challenges and need for integrations and alignments across differentsectionsofthecompanies.OneoftheverifiedsolutionsisasmartITtooltotakeoverthemanualwork and include all theproduct families’knowledge inorder to configure theproducts. The capabilities required for developing and implementing PCS needs to beinvestigatedinordertoaddressthechallengesreportedinSection1.3listedinTable1-2.Theoverallresearchobjectiveofthisresearchcanbebroadlyformulatedasbelow:

ResearchObjective:

To frame and improve the management process of PCS projects: planning,development,andmaintenancephases

ThisresearchobjectiveisverybroadandgivestheopportunitytoformulatepotentialRQsacrossthePCSprojectsphases.ThemanagementofPCSprojectswithinacompanycanbeperformedtaking intoaccountthefourdifferentphasesasshown inFigure1-2,while inthis PhD project, we considered and focus only on three phases. The implementationphasewhichhasafocusonorganizationalchangesatthecompanywhiledevelopingPCSprojectsisskippedinthisresearch.Thereasonistobedeeplyconcentratedonotherthreephasesduringthe limitedtimeduringPhD. Inordertoprovideconcreteanswerstothementioned challenges, the research objective is divided intoRQs,which are elaboratedbelow(seeFigure2-1).


12

Figure2-1ResearchobjectivesforselectedphasesofPCSprojects

The RQs targets: planning, scoping, knowledge management, maintenance anddocumentation,and furtherdevelopments inPCSprojects.Someof theRQssuchasRQ1and2belong to justplanningphase,whilesomeof themsuchasRQ3,4and5canhaveoverlapsamongdifferentphasesofPCSproject.Asanexample inFigure2-1,modellingand documentation could be covered in both development and maintenances phases.Integration with IT systems can take place both in development phase or as furtherdevelopmentsinthemaintenancephase.

RQ.1.HowcanPCSprojectsbeplanned?

EachofRQs(1,2,3,4and5)focusesonthementionedthreedifferentphasesofthePCSmanagementprocess tobe considered inPCSprojects.The firstRQ refers to thephasebeforestartingPCSprojectsortheinitiationandplanningphase.

RQ.1.1.HowcanwemakeastrategytodemonstratedifferentapplicationofPCSprojects?

TheRQ.1canbebrokendown intothreesubquestions. TheRQ.1.1aimsto improvethestrategicaldecisionsforprioritizingPCSprojects.Thisphase isstrategically important inthe companies forprioritizationof thedifferentprojects aswell asprioritizationof theresources time for different tasks. A framework is proposed that aims to provide astructured approach to developing a strategy and guide the development andimplementationprocessesofPCSsinETOcompanies.

RQ.1.2. How can the benefits and challenges from different PCS projects beevaluated?

TheRQ.1.2 ishow todetermine thebenefits, challenges anddeliverablesof theproject.ThisphaseischallengingasthecriteriacoulddeterminethesuccessorfailureofthePCS.DemonstrationdifferentpotentialofPCSprojectsandquantifyingthemresultsinthemostbeneficialprojectselection for thecompany.RQ.1.2addresses theevaluation techniques

13

andframeworksfordeterminingthebenefitsandchallengesfromPCSprojectswhichleadtoaformulatedplanningphase.

Todeterminethebenefits,theresearchcapturesthisopportunitybyquantifyingthelong-term cost savings produced by PCSs and compares it to the cost of development,implementation andmaintenance of the system.The question asked specifically in thepaperisasbelow:

WhatistherealizedROIforthecasecompany’sPCSoverafive-yearperiod?

The other research work asked questions about the motivations and benefits forimplementingPCS.Theresearchadopted inthispaper isbasedonsurvey followedwithinterviewstoanswerthequestions:

WhatarethemainmotivationsindustrialthatcompaniesconsiderforimplementingPCSstosupporttheirdesignandspecificationsprocesses?

How successful are companies in achieving benefits associated with the initialmotivationsdescribedpriortotheimplementationofthePCSs?

Thenextpaper identifiesandcategorizes themainchallengesmanufacturingcompaniesfacewhen implementing and utilizing their PCSs, and then assesses the importance ofthose categories, highlighting the specific challenges that can be found within eachcategory.ThequestionsaskedregardingassessingchallengesinPCSareasfollows.

WhatarethemaincategoriesofchallengesthatmanufacturingcompaniesfacewhenimplementingandutilizingthePCSs?

WhatistheimportanceofeachcategoryofchallengesthatcompaniesfacewhenimplementingandutilizingtheirPCSs?

WhichspecificchallengeswithineachcategorydomanufacturingcompaniesfacewhenimplementingandutilizingthePCSs?

RQ.1.3.HowcanastandardbusinesscaseforPCSprojectsbeframed?

Answering the questions abovewill provide the inputs for BC for PCS projects. RQ.1.3reports thechallengeof theunavailable framework formakingBCs inPCSprojects.TheproposedstructuredBCframeworkcanpavethewayofquantifyingmotivations,benefitsandchallengesregardingtherisksofthePCSproject.TheRQisansweredbyreviewingtheBCsframeworksforITprojectsanddetermineaframeworkforPCSprojects.

RQ.2.HowcanPCSprojectsbescoped?

RQ.2referstothelackofresearchonscopingPCSprojects.Theliteratureisreportingthegap for a structured framework and this research develops the framework byinvestigatingdifferentspreadtoolsandmethodsforscopingPCSprojects.ToanswerRQ.


14

2,allofthestepsavailableintheliteratureforscopingPCSistakenintoaccountandtheirsequenceisdeterminedandvalidatedinthecasecompany.

RQ.3.HowcanweacquireandmanageknowledgeneededforPCSprojects?

RQ.3 targets one of the main challenges related to development, maintenance anddocumentationsofPCSprojects,whichhasbeenreportedinliteratureforseveralyears.Inordertoacquiresufficientlevelofrelevantknowledgewhilemanagingtherequiredlevelof details in PCS projects, a structured framework is needed. There are KnowledgeManagement (KM) frameworks for IT projects available and in this researchworkweadoptthemforPCSprojects.

RQ.4.HowcanthePCSprojectsbemodelledanddocumented?

TomodelanddocumentPCSprojectsisoneofthemainconcernsinPCSprojectsandthisresearch contribute to solve this challenge at the companies. RQ.4 investigates theimportanceofmodellinginPCSprojectsandillustratesthegapintheliteratureforhavinganagileautomateddocumentationframeworkandtool.

RQ.4.1.Howcanweautomatethedocumentation,validationandcommunicationprocessinPCSprojects?

InRQ.4.1,wetooktheopportunitytoinvestigateaboutallavailabletoolsandmethodsindocumentationandmaintenancephaseofPCSprojects.Thehesitationofpractitionerstousedocumentationtoolsisreportedintheliterature.Furthermore,wetrytocontributetotheoryandpracticebyproposinganagileautomateddocumentationframeworkfollowedbyanautomatedITdocumentationsystemfulfillingtherequirementsfordocumentationinthecasecompany.

RQ.4.2.HowimportantisittousethestructuredmodellinganddocumentationtechniquesinPCSprojects?

TheRQ.4.2demonstratestheimportanceofdocumentingandmodellingtheknowledgeofPCS inastructuredway.Toanswer thisquestion,asurvey isconducted toevaluate theimportanceofavailableformalmodellingtechniquesinPCSprojectsandisansweringthefollowingpropositions:

Theuseof a formalmodelling technique (CPMprocedure)will result in improvedcontrolofproductvariants.

Theuseofaformalmodellingtechnique(CPMprocedure)willresultintheincreasedavailabilityofproductknowledgeinorganizations.

Theuseof a formalmodelling technique (CPMprocedure)will result in increasedemployeesatisfaction.

15

RQ.5.HowcantheintegrationswithotherITsystemsleadtoimprovementsintheperformanceofPCSs?

RQ.5 verifies that PCS performance increases when it is in integration with other ITsystems across the supply chain. Thismeans that connecting PCS to other internal orexternal IT systems increases the performance of PCS as it leads to automation in theprocesses. In this thesis, it is shown how to compare the similarities of the designedprojects inanotherdatabaseandconfiguredprojectsby internal integration. Inanotherstudy, the importance of integration across supply chain is illustrated while PCS isintegratedwithsuppliers’PCSexternally.

AnsweringalltheRQsandvalidatingthemisinneedofdesigningtheresearchingeneralandforeachRQ.ThefollowingchaptershowstheresearchmethodandthemainstagesofthePhDproject.

2.2ResearchmethodologyThissectionexplainstheresearchdesignofthePhDprojectingeneral.ThisPhDprojectisbasedontheclassificationofvariousresearchstudies.

2.2.1 BasicscienceresearchVandeVen(2007)proposesfourdifferentformsofengagedscholarshipdependingonthepurpose of the research to examine basic questions of description, explanation, andpredictionorappliedquestionsofdesign,evaluation,oraction intervention.As inFigure2-2,VandeVen(2007)defines these four formsasbasicscienceresearch,collaborativebasicresearch,designresearch,andactionresearch.Inbasicresearch,theresearcherisanoutsiderofthesocialsystembeingexamined,butsolicitsfeedbacksfromstakeholders.Incollaborativebasicresearch,theresearcherandstakeholderssharetheactivitiessuchas:Researchdesign,theorybuilding,problemformulationandproblemsolving.Theresearchteamsareoftencomposedofinsidersandoutsiders.Designscienceresearchisundertakentoexaminethequestionandevaluatethemasanoutsider.Actionresearchprojectstendtobeginbydiagnosingtheparticularproblemorneedofoneparticularclient.


16

Figure2-2Differentformsofengagedscholarship(VandeVen,2015)

ThisPh.D.project isan industrialprojectandabasicscienceresearch inHaldorTopsoeA/S.AccordingtoVandeVen(2007),basicscienceresearchisundertakentodescribeorexplain a socialphenomenonwhere the researcher is adetachedoutsiderof the socialsystembeingexaminedbutsolicitsadviceandfeedbackfromkeystakeholdersandinsideinformantson the research.Moreover, Iwas involved inprojects’planning,developing,implementing and maintaining as well as the overall managerial opportunity. As theproject manager, I was involved in the change process in Haldor Topsoe in order toimplement,testandlaunchtheprojectsrelatedtothePhDperiod.ThePh.D.researchwasexamined and evaluated inHaldor Topsoe A/S alongwith other case studies in otherindustries.Thishelps inunderstandingtheproblem indifferent industrialsettingsandadetailedobservationoflackoftoolsandframeworks.IhadtheoutsiderperspectivewhileevaluatingmydifferentsolutionsinsideTopsoealongwithothercasestudies.Combiningdifferentobservationsandstudiesindifferentcasecompaniesaswellasevaluationoftheproposed frameworks andmethods enrich the validity of research conclusions. At thesame time, the research results are sharedwith the practitioners to be applied in theprocessofplanning,developing,anddocumentingPCSprojects.

2.2.2 DesignresearchmethodologyBased on the previous section, this PhD thesis is conducted as basic science research,whichisdeterminedwheretheresearchislocatedandhowtheresearcherandindustriesareengaged.TheDesignResearchMethodology(DRM)includesdifferentstepsthataimtoformulateandvalidateproposedtheoriesandmethods.

TheDRMframeworkpresentedbyBlessingandChakrabarti(2009)isusedasaresearchframework for thisPhDproject.Themainconceptof theDRM framework is toview the

17

research constant progress with specific objectives. The nature of this researchmethodologyisheuristic,ratherthanalgorithmic.Theprogressfromonestagetoanotheris not linear, while includes many iterations. Figure 2-3 provides a graphicalrepresentation of the DRM framework. The DRM framework is selected as it includesdistinguished stages with specific goals. Moreover, this methodology assists theresearcher to observe the current situationwhilemoving towards the desirable futurestage in an iterative process. The DRM framework also allows the evaluation of thesuggested approach and its application. The findings are assessed and evaluatediteratively,whichensuretheprocessofobtainingmorevalidresults.

TheDRM frameworkconsistsof fourstages:ResearchClarification,DescriptiveStudy I,Prescriptive Study, and Descriptive Study II (Figure 2-3). DRM guides the researchProjectsthroughthesestages:

Inthefirststage,theResearchClarification(RC),thecurrentsituationandthenthefutureaims based on the gaps and drawbacks of the current situation are discussed.Furthermore,theliteraturestudyisperformedtofindouttheavailablesolutionandgapsintheliterature.Atthisstagethepreliminarygoalsandplanningaremadetosupportthefuturesuccesses.

In thesecondphase, theDescriptiveStudy I(DS-I), furtherunderstandingand literatureresearchareprovidedonahigherlevelofdetailsandthegoalofthestudyiselaborated.Thefactors influencingthecurrentsituationareexaminedandsupportedwithempiricaldata.

Inthethirdstage,thePrescriptiveStudy(PS),thescenarioscanbedevelopedbasedonthedataandexperiences.Thesescenariosaddressthechallengesinthecurrentsituation.

InthelaststageoftheDRM,theDescriptiveStudyII(DS-II),theresearchmethodusedtovalidatetheresultinthisthesisarecasestudies.Theoutcomeofthesuggestedscenariosisanalysed.Ifthesolutionisnotpromisingenough,theresearcherreturnstotheDS-Istageto improve the factorsand theassumptions, inorder to re-evaluate thesuggestions forimprovementandconsequentlytheresults.


18

Figure2-3Thedesignresearchmethodology(DRM)adjustedfrom(BlessingandChakrabarti,2009)

2.3ResearchplanThischapterdiscusseseachof thestagesof the thesiswhichhelps toview theresearchconstantprogresswithspecificobjectivesduringthethreeyears.

2.3.1 ResearchstagesBasedonthefirststepofDRM(RC),thefirststepoftheprojectisdefinedtoidentifythechallenges in ETO companies while planning, developing, and documenting or evenfurtherdevelopingPCSprojects.Basedon theavailable literatureand theexperienceofworkinginindustry,therealisticassumptionsbasedonthegapsandproblemsaremade.PartofRCisincludedinthefirstchapterofthethesisformulatingproblemstatement.Inthisstage,aclearanddeepunderstandingof thechallenges in industryandacademia isprovidedalignedwith investigations regardingall theavailable toolsandsolutions.ThemainRQsandsubquestionsareformulated.The initial literaturereview isconductedtoprovide theunderstandingof the importanceofPCS role inproduct customization anduntanglesomeofthepersonalizationchallenges.Theproblemstatement isbasedonthereportedchallenges in thedifferentphasesofPCSprojects.The resultsof thisstagearereflectedinpapersCandDintheChapter4.InPaperC,wearefocusingonmotivationsinindustrialapplicationsofPCSsandweareevaluating the realizedbenefits throughout asurvey.Thisstudyhelps in realizing theexpectationof the industryandhowsuccessfulthesecompaniesareinachievingthesebenefits.PaperDhasacomprehensivereviewon

19

challengesrelatedtoPCSindifferentphasesreportedinthecurrentliterature.Thispaperprovidesaclearunderstandingoftheimpactofthesechallengesthroughoutasurvey.

InthestageDS-I,empiricaldataiscollectedforfurtherunderstandingofchallengesinthePCS projects process from planning to the maintenance phase. Several phases of PCSprojects lifecycle are taken into account for this research project. Empirical data iscollectedfromindustriesviacasestudiesandsurveys.PaperBandIrepresentresultsoftheanalysis.InpaperB,weareanalysingROI inacasecompanyoverafive-yearperiod.CorrelationsbetweenusingthemodellingtechniquesandtheirimpactonPCSprojectsarediscussedinPaperIbasedonthesurvey.

InthePSstageoftheDRMframework,thefocusoftheresearchistoexaminethefactors,whichhave influencesonPCSprojectprocess.Differentframeworksfor improvingthesestagesaredevelopedduringthisresearch.Thefirstframeworkisrelatedtostrategiesfordemonstrating different application from PCS projects, which help in projectsprioritization in Paper A. In Paper E a framework for BCs in PCS application area isdiscussed.Then,PaperFproposesaframeworkforscopingPCSprojectsthroughoutthelifecycle.Paper G andH suggest frameworkspecifically for theknowledgemanagementphaseanddocumentationphaseinPCSprojects.PapersJandKdiscussdifferentstepsofastandard framework to improve the applicability of PCSs in integration with other ITsystems. Insomeof thesepapers, IT toolsarealsodeveloped tovalidate the frameworksuchaspaperHandJandK.

The last stage of DRM framework, PS II, provides the testing and validations for theproposedframeworksandtools.InthisPhDthesis,thevalidationisdonethroughoneormultiplecasestudies.PapersA,B,F,H, JandK focusesononecasecompanyvalidationandpapersEandGvalidatethedatabenefitingfrommultiplecasestudies.Thecasestudyispresented inallpapers isHaldorTopsoeA/S.OtherETOcompaniesarealsoused forproviding further testing and validation for the proposed frameworks which help ingeneralizability of the solution. The other papers including C, D and I are based on asurvey.

2.3.2 CasestudiesCasestudyhasconsistentlybeenoneofthemostpowerfulresearchmethodsinoperationmanagement,particularlyintestingnewtheories(Vossetal.,2002).Casestudiesresearchisthemostwidelyusedqualitativeresearchmethodininformationsystemsresearch,andis well suited to understand the interactions between information technology-relatedinnovationsandorganizationalcontexts(Darkeetal.,1998).Moreover,lackofmaturityinthisresearchfieldmakesthestudysuitableforcasestudies.

The case studymethod could support the objective of this thesiswhile answering the“how” and “why” questions. The case study comprises all-encompassing methods,coveringthe logicofdesign,datacollection approaches, and data analysis techniques(Yin,2013).


20

Inductive case based researches have been used both for hypotheses testing and todevelop new theories.However,when conductingmultiple case studies in this type ofresearch, attention has to be given to the data triangulation as well as observertriangulations(CreswellandClark,2007;Johnsonetal.,2007;Yin,2013).Multiplebenefitscanbegained from triangulationssuchas:complementary insights,whichadd firstly tothe richness and the convergence of observations which enhances the confidence infindings.Forexample, interviewscanbeconductedbytwopersons,withoneresearcherhandling the interview questions, while the other recording notes and observations(Eisenhardt,1989).

2.3.2.1 Casesandunitofmeasurement

The empirical foundation in this research is based on several case studies andsurveys/interviews with the relevant companies. In this study, different case studiesconsideredwhilecollaboratingwithotherresearchersinthearea.Thevastmajorityofthecase companies areproviding chemical andmechanicalproducts and services.Anotherimportantfactoristhatallcompanieshaveastronginterestinresearchanddevelopment,especially within the area of PCS projects. All of them already use PCS in sales andproductionprocess.Finally,allofthecasesaresimilarinreportingthechallengesrelatedto planning, developing, implementing and maintaining PCS project. All of the casecompaniesarediscussedelaborately ineachof thespecificpapers thatareprovided inAppendixchapter.InTable2-1,alistofthecasecompaniesusedinthecasestudiespapersisprovided.Thefirstcolumnisthenumberofcasecompanytoensuretheanonymity.Thesecond columnprovides the information about the typeof industries.Thenext columnshows the papers that used this case company as the empirical foundation. The lastcolumn shows the relevant RQs discussed in section 2.1 and the connection betweendifferentcasestudies,papersandRQsareprovided.

Theunitof analysis isnormally thePCSprojects as theRQs arebasedon thedifferentphases of the projects. Testing PCS projects in different industrial setting help inevaluatingthesolutionacrossthecompanies.

Table2-1Overviewofthecasecompanies

Casecompany Product/Industrialsector

Papers RQ.

I Chemical/Mechanical1,2,9,10,12,13,14,15,16,17,18,19,21,22(A,E,F,G,H,J)

1,2,3,4,5

II Mechanical/Chemical 14,15,19,21(G,H) 3,4

III Mechanical 3,4,19,21,23(B) 1,4,5

IV Mechanical/Chemical 7,11,19,21 1,4

V ITsystems/Services 19,21 4

VI Mechanical/Electrical 21 4

21

2.3.3 SurveysAsmentionedbeforeabout thedata triangulation, theuseofdifferent typesofdataandmethodsprovides richer andstronger evidence to support the findingsof a study (Yin,2013).Thisstudymainlybuildsonqualitativedataandqualitativedatamostlycollectedthrough interviews andobservations.For thequantitativedata,one surveydesigned inorder to have a deep understanding of the benefits and challenges of PCS project andcomparetheresultswiththeliterature.Inthesamesurvey,weaskedquestionsregardingthemodelling techniques in use and the effect of it on the PCS projects process. Thissurveycombinesmailandphoneadministration.

2.3.3.1 Populationandsampling

Thecriteriaforselectingthecompaniesforthestudywasmanufacturingcompaniesthatprovide customized or engineered solutionswith experiences of using configurators tosupporttheirspecificationsprocesses.Brainstormingsessionswereconductedtoidentifyadditional companies such as those involved in consultancies,vendorsof configurators,and collaborating companies. Additional companies were also identified in theinterviewing process, when the respondents were asked for a reference list of othercompanies that might fulfil the criteria to participate in the research. In total, 26companies fulfilling the selection criteria were contacted. The sample companiesrepresentsmall,medium,and largemanufacturingcompaniesprovidingcustomizedandengineeredsolutions.The levelsofcustomizationandcomplexityoftheofferedproductsvaries greatly among the companies, ranging from equipment and components tocompleteplants.Onlyonepersonfromeachcompanywasresponsibleforansweringthesurvey.Thecontactpersonateachcompanywasselectedbasedontheirfamiliaritywiththe PCS projects, irrespective of their formal role at the company, because in-depthknowledge of the configurators was required, which the top-level management at thecompaniesmightnotpossess.

2.3.3.2 Datacollectionanddataanalysis

The data collection is conducted in two phases: (1) a pre-study phase to test thequestionnaire,and(2)the interviewphasetocollectthedataforanalysis.Thepre-studyphase aims to validate the questionnaire and its administration procedure. Therefore,three pilot interviewswere conducted to test the applicability of the questionnaire indifferent typesof industrialconfigurationsettings.Thenextstep involvede-mailing thequestionnaires to the respondentswith a description of the purpose of the study, theinterviewprocedure,andafollowupnotification,andappointmentsweremadeforphoneinterviews.Thephoneinterviewswereconductedasawalkthroughofthequestionnaire.The interviewprocess allows clarification and elaborationof the responses to ensure acorrectandconsistent interpretationofthequestionsandtoensurethatthe interviewergainedacompleteunderstandingofthecompany’ssetting.

Intheanalysisphase,theanswerstothequestionnaireandthe interviewswereenteredintoanMSAccessdatabase,crosscheckedfordataentryerrors,andanalysed.Theanswersfromtheopenquestionswerecodedandgroupedintocategoriesaccordingtothenature


22

oftheanswers.Topreventbiasfromtheinterviews,thedatawerecodedandanalysedbyadifferentpersontotheonewhocollectedthedata.

2.3.4 CommunicationoftheresearchThe followingarticlesprovide thecontribution to this researchproject.Thearticles arecoded alphabetically from A to K, are selected among around 24 articles as the maincontributionsof these threeyearofresearch.The reason fornotusingall thearticles inthesis structure is to make the concept clear, transparent and avoid confusion whilesummarizing.Allthearticlesaresubmittedtoacademic journalsandconferenceswithinthe field of PCS.The selected articles are related specifically to theRQs and discussedelaboratelyinChapter4andthefullarticlesareappendedintheendofthisdissertation.

AsillustratedinTable2-1,thecasestudiesusedindifferentarticlesareclarified.PapersC,DandIaresurveypapersandthentheyarenotrelatedtothecasestudyresearch.

Table2-2Overviewofpublications

RQ.

Num

ber

Jour

nal

/Con

fere

nce

Pape

rsinth

ere

sult

chap

ter

Articles

Plan

ningP

CS(S

trat

egy,b

enef

its,c

halle

nges

,andb

usin

ess

case

s)

1 Journal A

Kristjansdottir,K.,Shafiee,S.,Hvam,L.,&Haug,A.,(2017).Howto Identify Possible Applications of Product ConfigurationSystemsinEngineer-to-OrderCompanies.JournalofIndustrialEngineeringandManagement,V.8,No.3.

2 Conference ___

Kristjansdottir,K.,Shafiee,S.,&Hvam,L.(2016).Developmentand Implementation Strategy for Product ConfigurationSystems in Engineer-to-Order Companies. IEEE InternationalConference on Industrial Engineering and EngineeringManagement(IEEM),Bali,Indonesia.

3 Journal B

Kristjansdottir,K.,Shafiee,S.,Hvam,L.,Myrodia,Anna&Bonev,M. (2017). Economic Value Creation from Using ProductConfiguration Systems – A Case Study. Submitted Journalarticle.

4 Conference ___

Kristjansdottir,K.,Shafiee,S.,Hvam,L.,Bonev,M.,&Myrodia,A., (2016). Quantification of benefits and cost from applyingproductconfigurationsystems.7thInternationalConferenceonMassCustomizationandPersonalizationinCentralEurope(MCP–CE2016),AtNoviSad,Serbia.

5 Conference C

Kristjansdottir, K., Shafiee, S., & Hvam, L. (2016). Industrialapplication of configurators: From motivations to realizedbenefits. 18th International Conference on IndustrialEngineering,Seoul,Korea.

23

RQ.

Num

ber

Jour

nal

/Con

fere

nce

Pape

rsinth

ere

sult

chap

ter

Articles

6 Conference ___

Kristjansdottir, K., Shafiee, S., & Hvam, L. (2015). Utilizingproductconfigurationsystemsforsupportingthecriticalpartsof the engineering processes. 2015 IEEE InternationalConference on Industrial Engineering and EngineeringManagement(IEEM),(pp.1777-1781),Singapore.

7 Conference ___

Kristjansdottir, K., Shafiee, S., Hvam, L., &Bonev,M. (2015).Identification of Profitable Areas to Apply ProductConfigurationSystems inEngineer-To-OrderCompanies,2015WorldConferenceonMassCustomization,Personalization,andCo-Creation,Montreal,Canada.

8 Journal D

Kristjansdottir,K., Shafiee, S.,Hvam,L, and Forza,C. (2017).The main challenges for manufacturing companies inimplementing and utilizing configurators. Submitted Journalarticle.

9 Journal EShafiee,S.,Kristjansdottir,K.,&Hvam,L.(2017).Howtoframebusiness Cases Framework for ProductConfiguration systemProjects.SubmittedJournalarticle.

10 Conference ___

Shafiee, S., Kristjansdottir, K., & Hvam, L. (2016). Businesscases for product configuration systems. 7th internationalconferenceonmasscustomizationandpersonalizationinCentralEurope,NoviSad,Serbia.

11 Conference ___

Myrodia, A., Kristjansdottir, K., Shafiee, S., Hvam, L. (2016).Product configuration systemand its impactonproduct's lifecycle complexity. IEEE International Conference on IndustrialEngineering and Engineering Management (IEEM), Bali,Indonesia.

Scop

ing

PCS 12 Journal F

Shafiee, S.,Hvam, L., & Bonev,M. (2014). Scoping a ProductConfiguration Project for Engineer-to-Order Companies.International Journal of Industrial Engineering andManagement,207-220.

13 Conference ___

Shafiee, S., Hvam, L., & Bonev, M. (2014). How to scope aproductconfigurationproject inanengineeringcompany?6thInternational Conference on Mass Customization andPersonalizationinCentralEurope,NoviSad,Serbia.

know

ledg

em

anag

emen

tin

PCS

14 Journal G

Shafiee,S.,Kristjansdottir,K.,&Hvam,L.,Forza,C.(2017).Howto scope configuration projects and manage the knowledgethey require. In the second roundof revision in InternationalJournalofKnowledgeManagement.

15 Conference ___

Shafiee,S.,Hvam,L.,&Kristjansdottir,K.(2015).Goal-OrientedData Collection Framework in Configuration Projects.2015WorldConferenceonMassCustomization,Personalization,andCo-Creation,Montreal,Canada.


24

RQ.N

umbe

r

Jour

nal

/Con

fere

nce

Pape

rsinth

ere

sult

chap

ter

Articles

Mod

ellin

g,d

ocum

enta

tion,a

ndm

aint

enan

ceo

fPCS

s

16 Journal H

Shafiee, S., Hvam, L., Haug, A., Dam, M., Kristjansdottir, K.(2017)Thedocumentationofproductconfigurationsystems:Aframework and an IT solution. Journal of AdvancedEngineering Informatics, doi:10.1016/j.aei.2017.02.004, V.23,pp.163–175.

17 Conference ___

Shafiee, S., Hvam, L., & Kristjansdottir, K. (2015). An agiledocumentationsystemforhighlyengineered,complexproductconfiguration systems. 22nd EurOMA Conference, Neuchâtel,Switzerland.

18 Conference ___

Shafiee,S.,Kristjansdottir,K.,&Hvam,L.(2015).AlignmentofConfiguration and Documentation for Highly EngineeredComplex Product Configuration Systems: a DemonstrationfromaCaseStudy.IEEE/WIC/ACMInternationalConferenceonWeb Intelligence and Intelligent Agent Technology (WI-IAT)(Vol.3,pp.249-250),Singapore.

19 Conference ___

Shafiee, S., Kristjansdottir, K., & Hvam, L. (2016). Industrialexperience from using the cpm-procedure for developing,implementingandmaintainingproductconfigurationsystems.18th InternationalConferenceon IndustrialEngineering,Seoul,SouthKorea.

20 Journal I

Hvam, L.; Kristjansdottir, K.; Shafiee, S.; Mortensen, N.H.(2017), Product configurators: A study of the impact ofapplying ITformalproductmodelling techniques.In the thirdround of revision in International Journal of ProductionResearch.

21 Conference

Shafiee,S.,Kristjansdottir,K.,Hvam,L.,Felfernig,A.,&Myrodia,A. (2016) Analysis of Visual Representation Techniques forProductConfiguration Systems in IndustrialCompanies. IEEEInternational Conference on Industrial Engineering andEngineeringManagement(IEEM),Bali,Indonesia.

PCSsinin

tegr

atio

nw

itho

therIT

syst

ems

22 Journal J

Shafiee, S., Kristjansdottir, K. & Hvam, L, (2017). AutomaticIdentification of Similarities across Products to Improve theConfigurationProcessinETOCompanies.InternationalJournalofIndustrialEngineeringandManagement,V.8,No.3.

23 Conference

Shafiee, S., Hvam, L., & Kristjansdottir, K. (2015). How toAnalyze and Quantify Similarities between ConfiguredEngineer-To-Order Products by Comparing the HighlightedFeatures Utilizing the Configuration System Abilities. 17thInternational Configuration Workshop (pp. 139-145), Vienna,Austria.

24 Conference K

Kristjansdottir, K., Shafiee, S., Bonev, M., Hvam, L., Bennick,M.H., &Andersen, C.S., (2016). Improved performance andquality of configurators by receiving real-time informationfromsuppliers,ConfigurationWorkshop,Toulouse,France.

25

InordertomakeiteasiertofindtherelevantpapersrelatedtoRQs,Figure2-4illustratesthepapersusedintheresultchapterandcolor-codedtofacilitatethesearchforrelevantRQs.

Figure2-4Linkbetweenpapersandresearchquestions

2.3.5 SchedulingtheresearchInordertohaveaclearunderstandingofthetimingandscheduleplanningduringthePhDproject, the Gantt table in Table 2-3 is considered. The table illustrates the RQs andpublishedpapersplannedover the time.As it isdemonstrated in2014, thePhD focus isRQ2andscoping,whilein2015,RQ1,4,and5areexploredbyconferencepapers.In2016,The PhD focus area is covering almost all the RQs due to the experiment gained byworkingondifferentPCSprojectsanddominantontheliterature.Thefocusin2017istowrapup all the gathered empiricaldata and concentrateonwriting andpublishing theknowledgeduringthepastyears.


26

Table2-3GantttableofPhDschedule

2.4ChaptersummaryThefirstsectionofthisChapterdiscussesthefiveRQswhichwillbeansweredduringthethreeyearsofPhD.ThesefiveRQscoverthreedifferentphasesofaPCSprojectincludingplanning,developinganddocumenting.Theanswers toRQsaim toproposesolution forthechallengesreportedduringthePCSprojectssupplychain.ThisresearchisstructuredbasedonDRMframeworkasitconsistsofdistinctstages;eachofthemsetsspecificgoalsand iterationamongdifferentstages.DRMenables the researcher tosetcleargoalsandcriteriaforachievingandevaluatingthesuccessofthesuggestedsolution.Moreover,thefactorsthatinfluencethecurrentsituationtowardsthefuturedesirablestatearedefinedandtheimpacttheyhaveonthesuccessisassessed.TheDRMframeworkalsoallowstheevaluation of the suggested approach and its application, as it enables an iterativeresearchprocess.Duringtheiterations,thefindingsareassessedandevaluated,toensuretheprocessofobtainingmorevalidresults.Theapplicationevaluationachievedthroughcollaborationswithothercompaniesandpractitioners.Theempirical foundation in thisresearch work is based both on one to multiple case studies. Inductive case basedresearcheshavebeenusedbothforhypothesestestingandtodevelopnewtheories.

27

TheTable2-2 illustrates the listofpublicationduring thePhDwhilemarking themaincontributionsused for this thesis.RQs and thepapers arenumbered,while thepapersusedinthePhDthesisaremarkedalphabetically.Asexplained,thereasonfornotusingallthepapers is tokeep the thesisas asummarized focused report.The timeschedule formanagingandplanningdifferentpublicationinanswerofdifferentRQsisdiscussedasthefinalchapter.


28

3THEORETICALFRAMING

Thischapterestablish themaingroundof theoreticalbackgroundof thePhDstudy.Thechapter introduces the topic of PCS to address the key challenges in ETO companies.Specificquestionsareformulatedtonarrowdownthescopeoftheresearchproject.Thischapterdescribes thebasicconceptsofsomeof the reportedchallenges inPCSprojectswhichleadtothecontributionsinthisPhDthesis.Theliteraturereviewpresentedinthischapter is taken from thearticles in theAppendix.Section3.1discusses thedifferencesbetweenITandPCSprojects.ThesedifferencesclarifytheproblemstatementandwhyweneedtocontributetoPCSareaandseparatePCSprojectsfromITprojects.Sections3.2and3.3explaintheconceptofstrategiesandmotivationforinvestmentinPCSprojectsandtheneedforastructuredframeworkintheliterature.TherequirementsforBCsinPCSprojectisdiscussedinsection3.4.Insection3.5,thescopingofPCSandhowliteraturecontributesto the concept is discussed. The knowledge management challenges and availableliteratures is discussed in section 3.6. Section 3.7 refers to the available modellingtechniques in literature as one of the important topics in PCS projects. This sectiondescribes the importanceofdocumentation inPCSprojects andhowother researcherscontributeso far.The importanceof the integrationsbetweenPCSandother ITsystemsarediscussedbasedontheavailableliteratureinsection3.8.The lastpartofthischaptersummarizesthediscussions.

3.1DifferencesbetweenITandPCSprojectsThereareseveraldifferencesbetweenITprojectsandPCSprojectscomparedtootherITprojectsasthereasonsforhavingaseparateresearchareaforPCS(illustrated inFigure3-1).

29

Figure3-1DifferentphaseofPCSprojectsanddifferencesfromITprojects

Becauseof theknowledgecomplexityandextensions inPCS, it isessential todeterminethescopeof theproject in theearlyphases inorder todetermine theentire flowof theproject.Thisisdonebyidentifyingtherequirements,evaluatingthetimeandbudget,andprioritisingthedifferentproductsandfunctionsdependingonthevarietyandcomplexityof the knowledge, the required tasks and the resources for the project development(Shafieeetal.,2014).ThescopeofITprojectsisdetermineddifferentlybecausenotallofthem necessarily involvewith extensiveproductknowledge. In PCS projects, goals andstakeholders’requirementsfromthefirststepshavetobeclarified;duetothecomplexityof the involvedknowledge, the rangeof stakeholders, and special system requirements(Studeretal.,1998).

TheseconddifferencerelatestothedetailsofeachsteprequiredforPCSscomparedtoITprojects. Because of the differences in the nature of the received knowledge, PCSs areformalised, combined, modelled and communicated differently from IT projects.Consequently, the knowledge modelled in PCSs is extensive and has to be continuallyvalidatedbydomainexperts(BasiliandWeiss,1984).StrongcommunicationbetweentheconfigurationteamanddomainexpertsinPCSsisvital,andspecificmodellingtechniquestend to solve this challenge inPCSs (Forza and Salvador,2002b).Additionally,withoutproper validation, very minor misunderstandings in the knowledge could lead to bigerrorsinthecalculationsandoutputs.

The thirddifference relates to theneed for specific comprehensivedocumentation andmaintenanceoftheknowledgeinPCSs(HaugandHvam,2007).Theknowledgehastobeclear andunderstandable to all stakeholders innon-IT language formany reasons.For


30

example, there is ahigh levelof integrationswithother ITsystems,and theknowledgemustbesharedamongparticipantsinthesupplychain.Inaddition,thefrequentchangesinproductknowledgenecessitatecontinualupdatingandmaintenanceoftheknowledge(Tiihonenetal.1996a;AlexanderFelfernigetal.2000;Friedrichetal.2014).Bycontrast,thedocumentationinITprojectsnormallyisasummarisedexplanationofthecodesandasetofuserstoriesthatarepassedontoanotherITspecialist(CoramandBohner,2005).Most IT projects are not required to work with complicated products or processknowledge,and ITspecialistsdonothave tocommunicatewithpeopleoutsideofthe ITfield forverifying the containedknowledge in the system.Furthermore, theknowledgerequired for IT projects does not have to be updated constantly (Coram and Bohner,2005).

Also,inPCSprojects,thereisahighlevelofintegrationswithotherITsystems(Felfernig,Hotz, et al., 2014), which implies a particular need for IT development, testing, andcollaborations.Mostoften,PCSprojectsbecomemoreexpensiveandtime-consumingthananticipated in thedevelopmentphase,whichmaybeexplainedby thecomplexityof theinvolvedknowledge,therangeofstakeholders,specialsystemrequirements,specificrisksanddifferentcostestimations(Hvametal.2008;Studeretal.1998).

In summary, the phases in IT project management differ from that in PCS projectmanagement. The only step which is as challenging in PCS as in IT projects is theimplementation phase. This phase involves with organizational challenges and changemanagementprincipleswhich is thesame forallkindof ITprojects.However,owing toknowledgecomplexityandvariety instakeholders’range inPCSs,theframeworksforITprojectsarenotsuitableforPCSsbecausetheyfailtoincorporatesufficientstepstocoveralltheneedsinPCSprojects.

3.2StrategiesforPCSProjectsDifferent authorshaveproposed frameworks to guide thedevelopmentof configuratorprojects.Haugetal.(Haugetal.,2012b)havedefinedstrategiesforconfiguratorsinETOcompaniesbyfocusingontheinvolvementofdifferentspecialistsinthedevelopmentandimplementationprocesses.Thesespecialistsaregroupedintoproductexperts,knowledgerepresentation experts and configuration software experts. This study discusses thestrengthsandweaknessesofvariousdevelopmentstrategies in relation to thedifferentexpertise involved.However, itdoesnotdescribehowprojectsareselectedand focusesinstead on the actual development process. Felfernig et al. (Felfernig et al., 2001) alsoproposed a development strategy based on the standard Unified Modelling Language(UML)designlanguagetodevelopandcopewithincreasingcomplexityintheknowledgebase. The three main components of the configuration environment were: knowledgeacquisition, configuration and reconfiguration.The authorsproposeddiagnoses at eachstage (Felfernig et al., 2001). which included an analysis and redesign of businessprocesses, modelling of the product range, selection of configurator software, andmodelling, implementationandmaintenanceof theplan.ForzaandSalvador (ForzaandSalvador, 2007b) provided guidelines for the implementation of configurators, whichincludedbenefitandcostanalyses,planningoftheimplementationprocessesandaligning

31

theexecutionoftheimplementationwithbestpractices.Therefore,Hvametal.(Hvametal.,2008)andForzaandSalvador (ForzaandSalvador,2007b)bothprovidedcompleteframeworks that described the different processes involved in configuration projects.However, inETO companies there is often high product complexity and variety,whichmeansthereisanadditionalneedtocreateastrategytoguidethefuturedevelopmentandimplementationprocesses.

3.3BenefitsfromusingPCSThe focusof this subsectionof literature review ison themotivations and thebenefitsdescribed in the literature inrelationwithPCS.Thissection isthebasicofrealizingandassessingrequirementsforBCs.

Aldanondo et al. (2000) describe how PCSs can be used in industriesproviding highlycustomizedproducts,where thereare iterativestepsresulting in longcycle time,riskofwastedtimeandmoneyifthecustomerrejectsthesolution,riskoftheproposedsolutiontobeunfeasible,andfinallytheinaccuratecostestimation.Ardissonoetal.(2003)proposeaPCSservicestosupportdiversecustomersinopenmarketenvironmentandtobeabletointegratewithsuppliersproviding configurable sub-products.Ariano&Dagnino (1996)presentacasestudybasedonamanufacturingfurniturecompany.Themainobjectivesofthe implementation of the system are described in terms of providing a system thatenables the employees to enter an order in quick and accurate manner, provide themechanism to check thePCS and finally generationofBOM anddrawings.Barker et al.(1989)presentacasewherethe initialpurposeofthePCSwastohelpemployees inthemanufacturingtovalidatethetechnicalcorrectnessbeforeproduction.Fleischanderletal.(1998) present a case where a PCS is applied in the complex domain of telephoneswitchingsystems.Thesystemsupportsnumberofstagesoftheproducts’ lifecyclethatincludessales,engineering,manufacturing,assembleandmaintenance. It isclaimed thatthesystemhasarchivedpositivereturnof investment inthefirstyear.Forza&Salvador(2002)presentacasestudywheretheintroductionofthePCShaspositivelyaffectedthesales,design,engineeringandmanufacturingprocessesat thecompany. ThebenefitsofusingthePCSaredescribedintermsoferrorsgeneratedinthesalesprocessesarealmosteliminatedasaresulttoanautomaticvalidityandcompletenesscheckperformedbythePCS. Furthermore, the time for generating proposal and consequently the man-hoursreducessignificantly.Inanotherstudy,Forza&Salvador(2002)introduceacasecompanyfaced with challenges regarding performance of correctness check of the products’specificationswithout increasing the control cost and reducingproductvariety. In thenext study Forza et al. (2006) present a company that implemented a PCS andimplemented a different product strategy that involved postponement of productdifferentiation. Haug et al. (2011) presents a study where fourteen companies areanalysed inorder toevaluate the impact from implementing aPCSon the lead-time forgenerating quotes and detailedproducts’ specifications.Heatley et al. (1995) present astudywhere PCS rator is used to support operational tasks at a company. Initially thesystemwasimplementedtosupporttheorderingprocess,whereerrorsweregeneratingthatcauseddelays,threatenedtheoverallquality,costandthecustomers’satisfaction.By


32

implementing thePCS, thecorrectnessandcompletenessof theorderswassignificantlyimproved.Heiskalaetal.(2005)performedastudytoanalysethebenefitsandchallengesofMC,configurableproducts,andPCSswithaspecialfocusonservices.Themainbenefitsaredescribedintermsofsuppliersandcustomers.Hvametal.(2004)presentacasestudywhereacompany is facedwithchangedmarketenvironment,and increasedpressuretodeliver inshorter time,with lowercostand improvedoverallperformance.Hvam etal.(2013)presentastudywheretheydescribetheobservedbenefitsfromapplyingPCSs infourindustrialcompanies.Theresultpresentedshowsthatlead-timehasbeenreducedby94-99%, on-time delivery is improved to be 95-100% and resources for making thespecifications have been reduced by 50-95%. Petersen (2007) focus on the benefits inengineeringcompaniesfromimplementingaPCS.Sviokla(1990)presentsthecasewhere,therequireddemandforflexibility,constantnewproductdevelopment,greatnumberofpossibleconfigurationsthecompanywaslackingoverviewresultinginnumberoferrors.Tiihonenetal.(1996)presentthemainmotivationsforimplementingaPCSsdescribedintermsofbeingabletotransferup-to-dateinformationtothesalesunitsandenablethemtouse it intherightways,andby increasingatomizationbyuseofPCSsthenumbersoferrorsshouldbe reduced leading to improvedquality.Trentinetal. (2012)explore theimpact from using PCSs on products quality. Finally, Yu & Skovgaard (1998) present astudyofaPCStoolwheretheaimaretoguaranteethecorrectnessoftheconfigurations,ensuring consistency, handling constrains, overcome limitations with regards tomaintainability and finally to support the use of configuration application in a user-friendlymanners.

Thediscussed literaturedescribesthevariousbenefitsthatcanbeachieved in industrialsettings from implementing aPCS. In fewof the research the initialmotivations for theimplementationofthePCSsareaddressed.

3.4BusinesscasesinPCSandavailabletoolsAlthoughthePCSliteraturedoesnotinadetailedmannerdealwiththedefinitionofBCs,someof theaspects that involveBChavebeendescribed.Basedon literature, themainsteps for BCs for IT projects can be described in terms of: 1) benefit analysis, 2)stakeholder’sanalysis,3) ITrequirements,and4)riskandcostanalysis(Gambles2009;Ashurstetal.2008;Häkkinen&Hilmola2008;McNaughtonetal.2010).TheliteratureofPCS projects reflects these aspects in some way, however there is a need both fromacademiaandpractitionerstocombinethesetoolsandmethodsinaframework.

Benefitanalysis isavitalchallengeforPCSprojectsas itdeterminestherequirementsofthe project and shed light into project scoping (Hvam et al. 2008; Shafiee et al. 2014).There are goalsmentioned forPCSprojects ingeneral (Petersen2007;Ardissono et al.2003),whiledifferentprojectshavedifferentmaingoals.ThegoaloftheprojectnormallydeterminestheexpectedoutputsfromthePCSproject,helpsinmanagingtheknowledge,anddeterminesthestakeholders(Mortensenetal.2011;Heiskalaetal.2007;C.Forza&Salvador2002;Salvador&Forza2007;Felfernigetal.2014).

33

Furthermore, stakeholders’ analysis which means the users’ expectations andrequirements for the system, increases as PCS projects become more successful andpopularamongusers(Barkeretal.,1989).Stakeholders’analysisforPCSprojects isalsonormallymentioned asoneof thedifficulties as the stockholders arevarying andhavedifferent expertise (Forza and Salvador, 2002a). Determining the stakeholders andanalysing their requirements before starting the projects leads to the right decisionswhich leads to time and resource saving (Felfernig,Hotz, et al.,2014;Mortensen et al.,2008;SalvadorandForza,2007;Shafieeetal.,2014).

Processanalysis isoneof themajorstepsbefore initiatingPCSprojects,where thesalesandengineeringprocessareredesignedto increaseefficiencybythehelpofPCSs(Forzaand Salvador, 2002a). This typically involves analysing the current processes andredesigning the future processes. Future process analyses IT architecture andrequirements if needed. Gap analysis is then used tomeasure the performance of thecurrentprocessandsetgoal for the targetperformance.Furthermore, theGAPanalysiscan be used to see how the different scenarios contribute towards the targetedperformance(Hvametal.,2008;TidstamandMalmqvist,2011).

Costandriskanalysisarecarriedouttocomparethedifferentscenarios.Intheliterature,thecostestimationforevaluatingthesavingsfromPCSandfurthersensitivityanalysishasbeenconducted (Kristjansdottiretal.,2016).Differentauthorsare recommending toolssuch as sensitivity analysis for analysing the costs and risks of PCS project before theplanningphase.TheserisksarecategorizedforPCSinliterature(Hvametal.,2008).

3.5ScopinginPCSprojectsAsillustratedinTable3-1theliteratureprovidesinputofpartsofthescoping,howevernoscholarshaveprovidedacompleteframeworkforhowtoscopeaPCS.TherearedifferentstepsforscopingPCS in literaturewhicharethecolumn inTable3-1andthesedifferentsteps are: 1) aims and goals of PCS, 2) identification of stakeholders and theirrequirements,2)objectivesof thePCS, the IT-architecture (inputs,outputs, integrationsetc.),4)productsandproductfeaturestoincludeanda5)projectplan.


34

Table3-1ApproachestoPCSscoping

Concepts

1Ai

ms

and

purp

oses

ofth

ePC

S:PC

SBe

nefit

s

2St

akeh

olde

rs’r

equi

rem

ents

3IT

stru

ctur

e:In

put/

Out

put

Man

agin

gan

dU

I

4Pr

oduc

ts’f

eatu

res: up

datin

gan

dm

aint

enan

ce

5Pr

ojec

tpla

n

RUP

Leve

lof

deta

ils

Mod

ellin

gte

chni

ques

Hvam et al. (2008) X X XX

Forza et al. (2007; 2002) XX X

Salvador et al. (2004) X

Mortensen et al. (2011) X X X

Haug et al. (2007) XX X

X

Ladeby et al. (1998) X

Blecker et al. (2004) X

Tidstam et al. (2010)X

Jannach et al. (2013) X X

Dvir et al. (2003) X

Felfernig et al. (2014; 2000) X X X

Trentin et al, (2011; 2012) X

Tiihonen et al. (1998) XX X

X

Sabin et al. (1998) X X

Jinsong et al. (2004) X

3.6KnowledgemanagementinPCSprojectsThe increased complexity of products increases the number of product features to bemodelled and maintained in the PCS (Ardissono et al. 2003); thus, the configurationknowledge for different parts of the products is often spread among various experts

35

within thecompanies(Hvametal.,2008).Othervaluablesourcesofknowledgearealsoavailable in internal software systems such as Enterprise Resource Planning (ERP)systems,calculationsystemsandspreadsheets(Friedrichetal.,2014).Therefore,earlyinthedevelopmentphaseofaconfigurationmodel,aknowledgeacquisitionandcleansingstageisrequiredtocentralise,modelandformulatetheproductorprocesstheknowledge(Felfernig et al. 2014). Knowledgemanagement (KM) in PCS is one of the most time-consumingtasksforthedomainexperts1andtheconfigurationteam2.KMcanbedefinedasanintegratedprocessincorporatingasetofactivitiestocreate,store,transferandapplyknowledge to a knowledge business value chain (Aurum, et al., 2008). Effective KMfacilitatesthecreationandintegrationofknowledge,minimizesknowledgelossesandfillsknowledgegaps throughout theproject(Lech,2014).AsTable3-2shows,manyauthorshavediscusseddifferentstepsbutnonehaveproposedaframeworkthatincorporatesthestepsinsequence.

Table3-2LiteraturebaseforthemainstepsintheKMinPCS

Author(year)Determiningthescopeof

thePCS

KnowledgeAcquisition

ModellingandKnowledgeValidation

Documentationand

MaintenanceForsytheetal.(1989) ü

Tiihonenetal.(1996) ü

Sabinetal.(1998) ü

Aldanondoetal.(2000) ü

Chaoet(2001) ü

Forzaetal.(2002) ü

Ardissonoetal.(2003) ü

Margoetal.(2003) ü

Tsengetal.(2005) ü

Jinsongetal.(2004) ü ü

1 The experts who provide domain knowledge of the process of performing the task and the data content, as well as

quality assurance, verification support (Barker et al., 1989).2 The team working on configuration projects include knowledge engineers, modellers, developers and project

managers (Hvam et al., 2008).


36

Author(year)Determiningthescopeof

thePCS

KnowledgeAcquisition

ModellingandKnowledgeValidation

Documentationand

Maintenance

Forzaetal.(2007) ü ü ü

Hvametal.(2008) ü ü ü

Mortensenetal.(2008) ü

Haugetal.(2009) ü ü

Yangetal.(2009) ü

Hansenetal.(2012) ü

Felfernigetal.(2014) ü ü

Shafieeetal.(2014) ü

3.7ModellinginPCSprojectsModelling is the product language by which the designer or product developer can elaborate,evaluate and communicate (Andreasen, 1994). The goals of design research are not to developas many different tools and models as possible (Gill, 1990), but to create a coherent set of toolsand models with a universal terminology and theory basis. Product Modelling is a high levelconcept for modelling created in relation to development of a product, defining of the product,or communication about the product (Krause, 1988). Tomiyama defines that “Modelling is aprocess in which observed facts are filtered by a theory” (Tomiyama et al., 1989).

The literature has been discussed in the table has been divided into five main groups in Table3-3. The first group called product architecture representation can be conceptual or visual andcan contain different levels of formalization which is measuring the relevance of the mentionedproduct architecture representation to product configuration field. Formalization simply is thesuccess of the modelling technique while representing the product knowledge. The secondcategory is modelling tool aspects and the method part containing the information about theutilization of specific methods in the field of modelling. The next field is about IT supportinvolved in product modelling and the validation and diagnosis part is implying to the researchworks which using methods for the validation the data used in the product modelling. Modellingfocus is referencing the papers which are doing the modelling from the product aspect or theprocess aspect or even both of them. The case study part is referring to the papers which areconducting the case study research study. The target group is referring to the target group in theindustry who are supposed to use the modelling. If the target group is the domain expert, itmeans that the modelling should be in an understandable language in order to make the

37

communication process simpler. A configuration group within a company can do the modellingin IT language and use it internally inside the configuration team as well.


38

Table3-3literatureresultsformodellingtechniques

Product

architecturerepresentation

Modellingtoolaspects

Modellingfocus CaseStudy Target

Group

Contribution

Conceptual

Visual

LevelofFormalization

Methods

Computersupport

Validationanddiagnosis

Products

Process

Industrialapplication

DomainExperts

Configurationgroup

(Andreasen,1980) ü ü

0 ü ü

(Aldanondoetal.2000) ü ü 5 ü ü ü ü ü

(Soininenetal.,1998) ü ü 5 ü ü

(HaugandHvam,2007) ü ü 5 ü ü ü ü ü ü ü

(Felfernigetal.,2000) ü ü 5 ü ü ü ü ü

(Nageletal.,2011) ü ü 3 ü ü ü ü

(Mortensenetal.,2008) ü ü 5 ü ü ü ü ü ü

(Dahmusetal.,2001) ü ü 1 ü

(ChaoandChen,2001) ü 5 ü ü ü

(MagroandTorasso,2003) ü 5 ü ü ü

(Tseng,H.E.etal.,2005) ü ü 5 ü ü ü ü ü

(Jinsongetal.,2005) ü ü 5 ü ü ü ü ü

(Yangetal.,2009) ü 5 ü ü ü

(Zhangetal.,1996) ü 2 ü ü ü ü

(Zhangetal.,2013) ü 5 ü ü ü ü ü

(Mailharro,1998) ü 5 ü ü ü ü

(Stumptneretal.,1998) ü 5 ü ü ü ü

39

Product

architecturerepresentation

Modellingtoolaspects

Modellingfocus CaseStudy Target

Group

Contribution

Conceptual

Visual

LevelofFormalization

Methods

Computersupport

Validationanddiagnosis

Products

Process

Industrialapplication

DomainExperts

Configurationgroup

(FelfernigandScience,2007) ü 4 ü ü ü ü

(HvamandLadeby,2007) ü ü 5 ü ü ü ü

(Haugetal.,2010) ü ü 5 ü ü ü ü

(ErensandVerhulst,1997)

ü ü 0 ü ü ü ü ü

(Felfernigetal.,2004) ü 5 ü ü ü ü ü

(Felfernigetal.,2013) ü 5 ü ü ü ü ü

(McGuinnessandWright,1998)

ü 5 ü ü ü ü

3.7.1 Documentation, maintenance, and communication in PCSprojects

OneofthemainchallengeswhenusingPCScanbealackofdocumentation,whichcanleadto incomplete and outdated systems that are difficult to understand (Tiihonen et al.,2013).For acompanyusingaPCS, it is thereforecrucial tohaveanefficientsystem fordocumentingthestructure,attributesandconstraintsmodelledwithinthesystem,aswellas to facilitate communication between PCS developers and domain experts.DocumentationisavitalpartofallITprojects,asitisusedforsharingknowledgebetweenpeopleandreducingknowledge loss,whenteammembersbecome inaccessible(Paetschetal.,2003).ThedocumentationofPCSincludesmodelling,maintainingandupdatingtheproductmodel,andstoringallinformationrelatedtotheproducts’attributes,constraintsand rules inside the PCS (Hvam et al., 2008). Product modelling is a method ofrepresenting thestructureandknowledgeof theproducton a relativelyvisual,abstractleveltoensurethattheyareunderstandabletoallpersonsconcerned.

FourbasicrepresentationsformodellingproductfamiliesforPCSareshowninFigure3-2(Duffy and Andreasen, 1995). The real world is the product knowledge available at acompany.Aproductmodelpresentsproductknowledge inastructuredway,whereasaninformationmodel (ITmodel) is a formal, IT-based representationof aproductmodel,


40

which is usually based on Unified Modelling Language (UML) notation (Kruchten andRoyce, 1996). As an IT model often cannot be understood by most domain experts –especiallywhenitincludescomplexconstraintsandrules–theproduct(orphenomenon)modelenablesdomainexpertsandconfigurationengineerstocommunicateinacommonlanguage,thusfacilitatingfutureupdatinganddocumentationchanges(Hvametal.,2008).

Figure3-2FourbasicrepresentationsofproductmodellingforPCS(DuffyandAndreasen,1995).

Previous studies discuss different ways of documenting PCS, which are expressed asproductmodelling techniques (Haug andHvam,2007;Hvam et al., 2008; Jinsong et al.,2005; Yang et al., 2009). There are IT tools available to facilitate the documentationprocess (Haug, 2009).However, drawbacks to these techniques have been reported astheyrequiremanualwork tobeundertaken,evenwhen IT toolsarebeingused.That is,boththemodellingandknowledgemodificationshavetobemaintainedseparatelyinthePCSandinthesystemusedfordocumentation.Accordingly,theknowledgeisduplicated,as the knowledge has to bemaintainedwithin two separated systems (Shafiee, 2015).Thus,maintenanceofthePCSrequiresupdatestobemadetothePCS(ITmodel)andtheproduct model, using both time and resources. Moreover, there is a high risk of thedocumentationof theconfigurationmodelsbeingof lowquality (Tiihonenetal.,2013).Rask(1998)addressesdocumentationandmaintenancerequirementsbyemphasisingtheimportance of traceability, verification and versioning. Application of the DependencyStructureMatrix(DSM)methodcouldprovidenaturalandtransparentdocumentationofthe ‘knowledgebase’(Sunnersjoetal.,2006).A functionalityanalysis forProductModelManagerwascomparedwith typesofsoftwarecommonlyapplied fordocumentingPCS(Haug,2009),whichshowedthatthenewsoftwareofferedsuccessfulPCSdocumentation,although thecommonchallengeofadditionalmanualworkbeingrequiredremained.Assuch,there isstillacriticalneed inthe industryforanefficientautomaticsolutiontobefound fordocumenting theknowledgeofPCS,communicatingwithdomainexpertsandvalidatingproductknowledge.

3.8PCSprojectandintegrationswithotherITsystemsThereareanumberofresearchthathaveexploredthehypothesis“thehigherthedegreeof integration across the supply chain, the better a firm performs” (Frohlich andWestbrook, 2001; Hines et al., 1998; Johnston and Lawrence, 1988; Lee et al., 1997;Metters,1997;Stevens,1989).Ragatzetal.(1997)identifythelinkedinformationsystemsapplicabilityasakeysuccessfactorforintegratingsuppliersintothenewproduct,processorservicedevelopmentprocess.Tallonetal.(TallonandKraemer,1996)pointout thatanypositive impactof IT comes from itsability tocoordinatevalueaddingactivities. Alinkage between integrative IT and supply chain is a key aspect of supply chainintegration.Stroeken(2001)examinesthelinkbetweenITandsupplychaininnovationin

1. Real World 2. Product Model(Phenomenon Model)

3. InformationModel (IT Model)

4. ComputerModel

Domain experts IT and configuration team IT and configuration team

41

sixindustrysectorsinordertoshowtheimportanceofITtodeveloptheprocessorientedstructureofthesupplychainneededfortheintegration(Stroeken,2001).

MukhopadhyayandKekre(MukhopadhyayandKekre,2002)quantifybothstrategicalandoperational impacts for Electronic Integration which leads to efficient procurementprocesses.Thestrategicbenefitsconcerningthesupplierandtheoperationalbenefitsareinrespecttobothparties,orthesuppliersandthecustomers.ItshouldthoughbenotedthattheoperationalbenefitsaregeneratedbyElectronicDataInterchange(EDI)throughreengineeringoftheinternalprocessesofanorganization,unlikestrategicbenefits,whichresultfromchangesinthebuyer-suppliertradingrelationship(MukhopadhyayandKekre,2002).Asupplychainstrategyrecognizesthatintegratedbusinessprocessescreatevalueforthecompanies’customersiftheseprocessesreachbeyondtheboundariesofthefirmbydrawingsuppliersandcustomersintothevaluecreationprocess(Stevens,1989;Tanetal.,1998).Vickeryatal.(2003)explainthislinkageastherelationshipbetweenwhereonevalueactivity isperformedandthecostorperformanceofanother isthen introducedasthe core purpose of supply chain integration as optimizing linkages amongst valueactivities.

Inakoshi et al. (1998) propose a framework for PCS that integrates a constraintsatisfaction problemwith a Case-basedReasoning tool (CBR),where the framework isappliedtoanon-linesalessystem.Thisframeworkcontainsthefollowingsteps:

1. Caseretrieval:similarcasesareretrieved from thecasebase inaccordancewiththesimilaritiesbetweenthecurrentqueryandthecases.

2. Requirement formalization: a well-defined requirement consists of the currentquery and the object function, and it is supplied to a Constraint SatisfactionProblem(CSP)solver.

3. Requirementmodification:Thewell-definedrequirementismodifiedonlyifthereisnoPCSonandtheCSPsolverreturnsnosolutionbacktotheCBRWrapper.

4. Partsdatabase:apartsdatabasethatcontainsthedefinitionofaproductfamily.Itdefinesthetypesofparts,theconstraintsonpartsconnectivity,andotherkindsofrestrictionsontheproducts.

5. CSPsolver:TheCSPsolverreceivesawell-defineduserrequirementandsolvestheproblem.

Based on the current literature in the field, the research highlights the importance ofachieving greater integrations in the supply chain where IT plays a significant role.Furthermore,forcompaniesprovidingcustomizedproducts,thereisaneedforhavingup-to-date information across the supply chains. Therefore, by integrating PCS across thesupplychains,itallowscompaniestofurtherintegratetheflowofinformationandatthesametimesolvesomeofthemainchallengesconcernedwithMCandPCS.However,theimpact from increased integration across the supply chains by enabling interactions ofPCSsacrossthesupplychainshasnotbeenaddressedintheliterature.


42

3.9ChaptersummaryThischapterpresentedanoverviewoftheavailableknowledgeonPCSprojectsdifferentphasesingeneral.Inthefirstsection,differencesandsimilaritiesbetweenITprojectsandPCSprojectswasdiscussedbriefly.TheconclusionisthatPCSprojectsarespecialtypeofIT projects but the available frameworks for IT projects cannot be applicable for PCSprojects.However, IT andPCSprojectsaresimilarenough thatwecanusesomeof themethods and tools or inspired by the available literature for IT projects. Eventually,different phases of PCS projects and available literature is discussed throughout thechapterincluding:theinitiationphaseandidentifyingstrategies,benefitsandBCs,scopingPCSprojects,knowledgemanagementand theknowledgecomplexitychallenges forPCSprojects,Modellinganddocumentingtheknowledge,andintegrationofPCSprojectswithotherITsystemssuchasERPsystem.

Inthenextchapter,theresultsfromresearchispresentedintheformof11papers.

43

4RESEARCHSUMMARY

Theresults frompreviouschaptersclarifies the importanceofPCS inETOcompanies intermsof theprocessautomationandtheproductscustomization.Thebenefits fromPCSare reported in numerous studies. The investigations from literature aligned with theempiricalcasesgivesusacomprehensiveunderstandingofthecorrespondingobjects.PCShavebeenusedinindustryoveradecadewheretheliteraturereportsseveralchallenges.Basedonthosechallengesandthegapinthecurrentliterature,theRQsinthisprojectaredefined.The research objectives of this PhD is to frame and improve themanagementprocessofPCSprojects:planning,development,andmaintenancephases.ThisobjectiveisbrokendownintothefollowingRQs:1)PlanningthePCSproject,Strategies,motivations,benefitsandchallengesinPCSmanagementwhichresultinastructuredBusinesscase,2)Scoping PCS project, 3) Managing knowledge across PCS project, 4) Modelling,documentation andmaintenance of the PCSprojects, 5) PCS integrationswith other ITsystems.Thethesisconsistsof24papersand11ofthemhasbeenconsideredtoanswertheRQs.Section4.1presents thesummaryof theselectedarticles inorder toreply theRQs.TheRQsaresummarizedbelow:

RQ1.HowcanPCSprojectsbeplanned?

RQ.1.1.How canwemake a strategy to demonstrate different application of PCSprojects?RQ.1.2. How can the benefits and challenges from different PCS projects beevaluated?RQ.1.3.HowcanastandardbusinesscaseforPCSprojectsbeframed?


RQ.3HowcanweacquireandmanageknowledgeneededforPCSprojects?



44

RQ.4.1.How canwe automate the documentation, validation and communicationprocessinPCSprojects?RQ.4.2. How important is it to use the structured modelling and documentationtechniquesinPCSprojects?

RQ.5.HowcantheintegrationswithotherITsystemsleadto improvementsintheperformanceofPCSs?

4.1PublicationsoverviewInthischapter,anoverviewoftheachievedresultsfromdifferentpublicationsduringthethreeyearsispresented.AsshowninTable2-2,thisPhDstudyendedupwith24papers,withatotal15conferencepapersand9journalpapers.Thesamecolor-codinginChapter1and2 isutilized toemphasizeontheRQs.All theconferencepapersarepre-reviewedand published in scholarly conference proceedings between 2014 and 2017.Of the 24papers,2of the journalpapersarepublished,2of themareunderrevisionandthe5ofthemaresubmitted.

ThePhDthesisislimitedtoconsiderthemainpublicationstoanswertheRQsduetothehigh number of publications. The other reason for not considering some of thepublications intheresultchapter isthatsomeoftheconferencepaperspresentedasthefirstversion intheconferencesandthentheextendedversionhasbeenpublished intheform of journal paper. Furthermore, 11 articles are chosen for result chapter whichinclude9journalpublications(Table4-1).

PapersAtoEanswerstoRQ.1.PaperAframesastrategyfordifferentapplicationsofPCSforRQ.1.1.PapersB, C andDdiscuss thechallenges,benefitsandmotivations fromPCSprojects throughoutdifferentphasesand refer toRQ.1.2.Paper Eanswers theRQ.1.3 inordertoframeastructuredBCforPCSprojects.PapersEbuildsontheresultsfromPapersA toDwhere thestructureofBCs is formedbasedonall thedata fromROI torisksandchallenges.

Paper Fpropose a framework toscopePCSprojects in theplanningphasebyusing theoutputsfromBC(PaperE).ThispapercontributestofillthegapinRQ.2.

Paper Guses the results fromPaper F toscopePCS inorder todetermine the requiredknowledgeinastructuredway.TheproposedframeworkforKMisananswertoRQ.3.

RQ.4isansweredbyPapersHandI.PaperHanswerstheRQ.4.1regardingtheneedforanautomateddocumentationsystembysuggesting theagile framework fordocumentationandcommunicationpurposeinPCSprojects.PaperHdemonstratesthedevelopmentofITtoolbasedontheproposedframeworkinacasecompanyandreportstheresults.PaperIisasurveypaperandevaluatetheimpactandimportanceofusingastructuredmodellingtechniqueinPCSprojectsinordertoprovidetheanswertoRQ.4.2.

Papers JandKanswertheRQ.5whichevaluatesthe influenceofIT integrationwithPCSaccross supply chain. Paper J propose a framework to assess the similarities between

45

previouslydesignedproductsintheconfigurationprocess.PaperJprovestheapplicationof theproposed frameworkby testing it throughan IT tool in acasecompany.PaperKdemonstratesthe importanceofPCS integration inanexternalsetting inordertoobtainaccurate information in PCS. The paper discusses a case company that benefits fromintegratingtheirownPCStoavendorsPCS.

Table4-1Overviewofpublications

1 RQ.1 A*(Journal)Kristjansdottir, K., Shafiee, S., Hvam, L., (2017). How toIdentify Possible Applications of Product ConfigurationSystems in Engineer-to-Order Companies. Journal ofIndustrialEngineeringandManagement,V.8,No.3.

2 RQ.1 B(Journal)Kristjansdottir, K., Shafiee, S., Hvam, L., Myrodia, Anna &Bonev, M. (2017). Economic Value Creation from UsingProduct Configuration Systems – A Case Study. SubmittedJournalarticle.

3 RQ.1 C(Conference)

Kristjansdottir,K., Shafiee, S., &Hvam,L. (2016). Industrialapplication of configurators: From motivations to realizedbenefits. 18th International Conference on IndustrialEngineering,Seoul,Korea.

4 RQ.1 D(Journal)Kristjansdottir,K.,Shafiee,S.,Hvam,L,andForza,C.(2017).The main challenges for manufacturing companies inimplementingandutilizingconfigurators.Submitted Journalarticle.

5 RQ.1 E(Journal)Shafiee, S., Kristjansdottir, K., & Hvam, L. (2017). How toframebusinessCasesFramework forProductConfigurationsystemProjects.SubmittedJournalarticle.

6 RQ.2 F*(Journal)Shafiee,S.,Hvam,L.,&Bonev,M.(2014).Scoping aProductConfiguration Project for Engineer-to-Order Companies.International Journal of Industrial Engineering andManagement,207-220.

7 RQ.3 G(Journal)Shafiee, S.,Kristjansdottir,K., &Hvam,L., Forza, C. (2017).How to scope configuration projects and manage theknowledge they require. In the second round of revision inInternationalJournalofKnowledgeManagement.

8 RQ.4 H*(Journal)

Shafiee, S., Hvam, L., Haug, A., Dam, M., Kristjansdottir, K.(2017),Thedocumentationofproductconfigurationsystems:A framework and an IT solution. Journal of AdvancedEngineeringInformatics,doi:10.1016/j.aei.2017.02.004,V.23,pp.163–175.

9 RQ.4 I(Journal)

Hvam, L.; Kristjansdottir, K.; Shafiee, S.; Mortensen, N.H.(2017), Product configurators: A study of the impact ofapplyingITformalproductmodellingtechniques.Inthethirdround of revision in International Journal of ProductionResearch.

10 RQ.5 J*(Journal)

Shafiee,S.,Kristjansdottir,K.&Hvam,L., (2017).AutomaticIdentificationofSimilaritiesacrossProducts to Improve theConfiguration Process in ETO Companies. InternationalJournalofIndustrialEngineeringandManagement,V.8,No.3.

11 RQ.5 K(Conference)

Kristjansdottir,K., Shafiee, S.,Bonev,M.,Hvam, L.,Bennick,M.H., & Andersen, C.S., (2016). Improved Performance andQualityofConfiguratorsbyReceivingReal-TimeInformationfromSuppliers,ConfigurationWorkshop,Toulouse,France.


46

4.2PlanningproductconfigurationsystemsThissectioninvestigatesthepossibilitiesofassessmentsfromtheSection3.2,3.3,3.4thatexplainedthebenefitsandchallengesreportedaswellastheimportanceofstrategiesandBCs in PCS projects. Therefore, the research discusses the strategies, benefits andchallengesinPCSprojectsandintroducesthesolutionsforstrategicdecisionsandbenefitsquantifications. A BC framework is proposed by the results from previous studiesregardingbenefitsandchallenges.ThissectionaimstoanswertheRQ1.

4.2.1 Paper A: How to Identify Possible Applications of ProductConfigurationSystemsinEngineer-to-OrderCompanies

Authors:KatrinKristjansdottir,SaraShafiee,LarsHvam

Published in: Journal of Industrial Engineering and Management, V. 8, No.3., earlierversion originally published in: Proceedings of IEEE International Conference onIndustrialEngineeringandEngineeringManagement(IEEM),Bali,Indonesia,2016.

4.2.1.1 Researchobjectiveandresearchquestion

Thecurrent literaturedescribesdifferentdevelopmentstrategies thatcanbeapplied tocompanies thatusePCSs (Felfernigetal.,2001;ForzaandSalvador,2007a;Haugetal.,2012a;Hvam et al., 2008).However, the literature does not explain how to develop astrategy to guide the development and implementation processes. This includesidentifying different applications of PCSs with the aim of getting the most out of theinvestment by avoiding the sub-optimization of individual processes. Furthermore,stakeholdersaremorelikelytohaveacommonunderstandingoftheoverallaimwhenastrategy isguiding theprocess.Thispaperaims tocontribute to the literatureandhelppractitioners by providing a framework that ETO companies can use to maximize thevalueofPCSandminimizerisk.Thisresearch thereforeaims toanswer to the followingRQ:

HowcanETOcompaniesidentifypossibleapplicationsofPCS?

ToanswerstheseRQ,aframeworkisproposedthataimstoprovideastructuredapproachtodevelopingastrategytoguidethedevelopmentandimplementationprocessesofPCSsinETOcompanies.Theproposedframework isbuiltonthe literature inthefieldofPCSsandwasvalidatedthroughacasestudyinarealETOcompany.

4.2.1.2 Researchcontribution

In order to guide the development and implementation strategies of PCS in ETOcompanies, this research proposes a framework that demonstrates alternativeways toapply the PCSs. The following steps must be taken into account when making adevelopmentandimplementationstrategyforPCSsinETOcompanies(Figure4-1):

47

Figure4-1TheproposedframeworktoidentifyapplicationsofPCS.

CasestudyThe company’smain objectives are reducing routinework,decreasing the lead time togenerateproposalsandproductspecifications,increasingthehitrateasaresultofshorterlead times and improving the quality and abilityof the global sales offices to generateproposals. The importance of these objectives can differ from project to project. Forinstance,foraprocessingplantwithavery lowsalesrate,theobjectiveof investing inaconfigurator istoempoweralltheregionalofficesaroundtheworldandextract implicitknowledgefromemployeesinordertomaketheinformationmoreexplicit.Inthisstep,acomplete overview was provided to the stakeholders in order show them why theconfiguratorswerenecessarytoachievetheseobjectives.

Thecompanypreviouslyfocusedondevelopingconfiguratorsthatcouldsupportthesalesprocesses in thedifferentbusinessunits.However, given the increasedmaturityof theconfigurationareaatthecompany,aninvestigationwasinitiatedonhowtheengineeringprocesses could be supported. Sales and technical configurators are not mutuallyexclusive.Theymayevenbeused inconjunctionwitheachother(seeFigure4-2)wheretheoutputofthesalesconfiguratorisusedastheinputforthetechnicalconfigurator.

Thisimpliesamorestandardizedwayofapplyingthetoolsandsoftwarethatareneededtogeneratedifferentproductspecifications.Inthecasestudy,Excel-based in-housetoolswere replaced with configurators. This also allowed for more uniform productspecificationsthatareexpectedto improvetheoverallquality.Usingthesametoolsalsohelped.


48

Figure4-2Theinteractionbetweensalesandtechnicalproductconfigurationsystems(PCS).

Havingan increasednumberofconfiguratorsinoperationindifferentdepartmentsopensupthepossibilitytobasetheoutputontheinputfromindividualconfiguratorsinordertocovercompleteprojectsandplants(seeFigure4-3).Withthissetup,thesalesdepartmentcouldgenerateproposalsbytaking inputdirectlyfromtherelevantdepartments insteadofreceivingtheinputviae-mailandcopyingandpastingitintoaWordfile.Thisimpliesamore standardizedprocedure across departments and amore structured approach forknowledgesharing.

Figure4-3Achievingsynergybygeneratingdocumentsusinginformationfromproductconfiguration

systems(PCS)acrossdepartments.

Thecompanyinthecasestudyusedseveralsoftwareapplicationsforcomplexcalculationsandsimulations.Thesesoftwareapplicationsprovidedcriticalinformationinthesalesandengineeringprocesses.

In order to support a business to the greatest extent possible via configuratordevelopment and implementation, it is necessary to obtain a holistic view of each BU.Therefore, visualizations of how configurators can support the different BUs weregenerated, taking intoaccount theabove-mentioned focusareas.Figure4-4providesanexampleofoneoftheBUswheredifferentdevelopmentconceptswereapplied.

49

Figure4-4Simplifiedoverviewhowthesalesandengineeringprocessescouldbesupportedbyproductconfigurationsystems(PCS)andinteractionwithotherITsystems.

4.2.1.3 Conclusion

PCSs are usually implemented gradually in ETO companies due to highly complexproductsandprocesses,whichhighlightstheimportanceofhavingastrategytoguidethedevelopment and implementation processes. The literature describes differentdevelopment strategies (Felfernig et al., 2001; Forza and Salvador, 2007a;Haug et al.,2012a;Hvam etal.,2008);however,noneof themspecifieshow tocreate a strategy toguidethedevelopmentandimplementationprocessesforPCS.ThisisespeciallyimportantinETOcompanieswhereprojectstypicallyhavehighproductcomplexitythatrequiresthegradualimplementationofPCSs.Therefore,thispaperaimstofillanimportantgapintheliterature,aswellasprovideETOcompanieswithaframeworkthatcanhelpcreateamorestructuredapproachtodevelopmentandtheimplementationofPCSsinordertomaximizetheir investment and minimize risk. Furthermore, this should help companies withstrategicplanningandwhenjustifyingtheirinvestmentsinPCSs,astheycandemonstratethedifferentapplicationareas.Finally, the required resources,benefitsandoutputsaredeterminedandtherequirementsareexpressedbeforestartingtheactualdevelopmentofthePCS.ThisstudyislimitedtoETOcompaniesbecausedevelopmentandimplementationin these companies are more complex. Furthermore, PCS are usually implementedgradually,whichhighlightstheimportanceofhavingastrategy.Therefore,futurestudiesshouldincludetestingonotherkindsofcompanies,aswellasotherETOcompanies.


50

4.2.2 Paper B: Economic Value Creation from Using ProductConfigurationSystems–ACaseStudy

Authors:KatrinKristjansdottir,SaraShafiee,LarsHvam,AnnaMyrodia,MartinBonev

Published in: Submitted Journal article, earlier version originally published in:Proceedingsof7th InternationalConferenceonmassCustomizationandPersonalizationinCentralEurope(MCP-CE2016).UniversityofNoviSad,2014.


ThegapintheliteratureregardingtheReturnOnInvestment(ROI)forPCSsincompaniesisnotduetoa lackof interest,buttochallengesrelatedtoaccessingthenecessarydata.Suchchallenges include(1)confidentiality,(2)a lackofavailability,and(3)theneedfordata tobegathered from internalsystemsat thecompanyandbevalidatedbydifferentstakeholders.This article aims to capture this research opportunity by quantifying thelong-term cost savings from using PCSs and compare it to the cost of development,implementationandmaintenanceofthesystem.TheROIsatoneyearandfiveyearsarepresented.The cost savings factors are calculated based on factors related to resourceconsumption(man-hours),leadtime,thequalityofproducts’specificationsandincreasedsales.These are commonbenefitsof implementingPCSs and are thought tobedirectlylinkedtoactualcostsavings.Forthispurpose,thefollowingRQisformulated:

WhatistherealizedROIforthecasecompany’sPCSoverafive-yearperiod?


To answer the research questions, we first determined whether prior research hadquantifiedthebenefitsandcostsofPCSand,morespecifically,whethertheexpectedROIofPCS isquantified.Next,an in-depthcasestudywasconducted inordertoquantifythebenefitsthataredirectlylinkedtocostsavingsandtheoverallcostoverafive-yearperiod,whichwereusedtocalculatetheROI.Theresultspresented inthepapernotonlyfillanimportantgapintheliteraturebutalsoidentifyabenchmarkofROIwheninvestinginPCSprojects.

TheROIforaPCSwasdiscussedinpreviousstudies(Barkeretal.1989;Fleischanderletal.1998;ForzaandSalvador2002b).Unfortunately,theresultsarenotquantified inthisresearch, and therefore the results cannot be directly compared to those of this study.However,Barkeretal.(1989)statesthatapositiveROIcanbeachieved inthefirstyear,supportingourfindings.

Casestudy

In this study, the significant benefits of implementing and utilising a PCS in a casecompanyareidentified.Bycomparingthedirectcostsavingsduetoreducedman-hourstothedirectcostofdeveloping,implementingandmaintainingthePCS,itcanbeconcludedthatthePCSishighlybeneficialforthecompanywithinthefive-yearperiodanalysed.TherealisedROI for thePCScanbecalculatedbasedon the findings. InTable4-2, themain

51

findingsofthecasestudyaresummarised,andtheROIiscalculatedforbothoneyearandfiveyearsafterimplementation.

Table4-2ROIfromapplyingPCSoverafive-yearperiod

CostsavingandcostfactorsassociatedwiththePCS Amount Unit

Costsavingsbasedonsavedhoursoverfive-yearperiod 31,452,606 €

Costofdevelopment,implementationandmaintenanceoverfive-yearperiod 2,406,260 €

Totalcostsavingsoverfive-yearperiod 29,046,260 €

RealisedROI

ROI(inthefirstyearafterimplementation) 477.76 %

ROI(fiveyearsafterimplementation;2009–2013) 1,207.12 %

Otherbenefitsthatcouldnotbequantified

Leadtimesavings(average) 4.71 Days

IncreaseinsalesoutputthroughPCS 467 %

4.2.2.3 Conclusion

TheresultsofthestudyshowthatahighROI(477.76%afterfirstyearofimplementationand1,207.12%afterfiveyears)canbeachievedbyinvestinginaPCSinamanufacturingcompany.Ourresultsshowthattheamountoftimerequiredtogeneratequotationscanbe significantly reduced; 629,052 man-hours (21.41%) were saved due to theimplementationandutilizationof thePCSovera five-yearperiod.Ourresultsalsoshowthat the lead time for generating quotationswas reduced from 8.00 days to 3.29 days(41.13%)due to thePCS.Thedirectcost factorsweredivided intothreegroups:costofdevelopment, implementationandmaintenance.DevelopmentofthePCSwasperformedovertwoyearsandcost461,760€,and implementationcoststotalled300,000€.Overafive-yearperiod,thecostofmaintenancewasestimatedtobe442,000€.

ThemotivationofthestudywastoanalysetherealisedROIforaPCSinamanufacturingcompanyovera five-yearperiod.TheROI iscalculatedbasedon theactualcostsavings,which can be traced to the man-hours saved when generating specifications for theproduct and the cost of development, implementation andmaintenance. This researchaims to fill a gap in the literature regarding theROI of PCSs and the link between thebenefitsofPCSs anddirect costsavings.Thisgapexistsdue todifficulties in retrievingdata,suchasalackofavailabilityandvalidation.ThefindingsofthisresearchshowthatapositiveROIcanbeachievedwithinoneyearafterimplementationofthePCS.TheROIis477.76%oneyearafter implementationand1,207.12% fiveyearsafter implementation.This indicates that a PCS is highly profitable when implemented to support thespecification processes of manufacturing companies. This paper provides empiricalevidencebasedonsinglecasestudies,which imposesome limitationson thestudy.Thefindings are derived from an in-depth analysis at the company, which was based onhistoricaldata,interviewsandworkshops.


52

4.2.3 Paper C: Industrial application of configurators: Frommotivationstorealizedbenefits

Authors:KatrinKristjansdottir,SaraShafiee,LarsHvam

Published in:Published in:Proceedingsof18th InternationalConferenceon IndustrialEngineering,Seoul,Korea.


The literature describes the various benefits that can be achieved from utilizing PCSs.However,themotivationbehindtheimplementationandhowsuccessfulcompaniesareinachievingthebenefitsthatcanberelatedtotheinitialmotivationshasnotbeenaddressedto great extent in the literature. Besides, the majority of the literature describes themotivations and the realized benefits based on single case company, which makes itdifficulttogeneralize.Thispaperaimstocapturethatresearchopportunitybyanalysingthe actual motivations for implementing PCSs and how successful companies are inachieving the benefits in relation with the initial motivations. In addition, the resultsprovide insight in to the main challenges companies face in the process of providingcustomizedproducts andhow those challenges canbe addressedby implementationofPCSs.AlignedwiththefocusofthestudythefollowingRQsweredeveloped:

Whatare themainmotivations industrialcompaniesconsidering for implementingPCSstosupporttheirdesignandspecificationsprocesses?

How successful are companies in achieving benefits associated with the initialmotivationsdescribedpriortotheimplementationofthePCSs?

Theresearchmethodadopted in thispaper isbasedonsurvey followedwith interviewswith22companies,whichallprovidecustomizedorengineeredproductsandusePCSstosupportthegenerationofthespecificationsinthesalesanddesignprocesses.Thesurveyconsistedbothofopenandclosedquestions,tocaptureboththequalitativenatureofthemainmotivationsandtoquantifytowhatdegreecompaniesagreewithachievingthemainbenefitsdescribed in the literature.Basedon theanswersgathered from thecompaniesthemotivationsweregroupedintosevencategorizes.


Sevenmaincategorizewere identifiedbasedonthemotivationsgivenbythecompanies.Those categories are 1) general competitiveness (27% of companies), 2) knowledgemanagement (36%of companies),3) efficiency in the sales andorderprocess (45%ofcompanies),4)efficiencyintheproductionprocess(27%ofcompanies),5)accuracyoftheproducts’ specifications (41% of companies), 6)management of products variants andcomplexity(23%ofcompanies)andfinally7)othermotivationsthatcouldnotbegroupedwiththeothers(24%ofcompanies).

In the firstmotivation categorygeneral competiveness, sevenbenefitswere grouped. Inthiscategorysignificantdifferenceof thecompanies thatexpressed amotivation in this

53

categorycanbeseenas79%onaverageagreedwiththosebenefitswhileforcompaniesnotexpressingamotivationgroupedintothecategory53%agreed.

The second motivation category knowledge management three benefits were grouped.However, not significant difference can be found between companies expressing amotivation inthiscategoryandtheonesnotexpressingamotivation inthiscategory,asthe number of the companies agreeing to the benefits on average are 67% and 66%consequently.

Thethirdmotivationcategoryefficiencyinthesalesandorderprocessesfourbenefitsweregrouped.However,aninterestingfindingisthatonaverage90%ofthecompanies,whichdidnotexpressamotivationinthiscategoryagreedwiththosebenefitswhile78%ofthecompaniesexpressingamotivation inthecategoryagreedonaverage.Therefore,higherpercentagesof companiesnot expressing amotivation grouped in this category agreedwithachievingtheassociatedbenefits.

The fourthmotivation category efficiency in theproductionprocesses twobenefitsweregrouped,whichare (15) reductionofcost in relationwithconstructionandproduction,and(16)reductionofcostinrelationwithproductionandprocurementofmaterials.Outofthosetwobenefits77%ofthecompaniesagreedwithreductionofcostinrelationwithconstruction and production being a benefitwhile 46% of the companies agreedwithreductionofcost inrelationwithproductionandprocurementofmaterials. In termsofcompanies thatexpressedamotivationgrouped into thiscategoryasignificantdifferentwas found.Onaverage from thecompaniesexpressingachallenge in thiscategory83%agreedwiththisbeingarealizedbenefit,whileonly53%ofcompaniesnotexpressingamotivationinthecategoryagreedonaveragewiththisbenefit.

The fifthmotivationcategoryaccuracyof theproducts’specifications threebenefitsweregrouped.Intermsofcompaniesthatexpressedamotivationgroupedintothiscategoryasignificantdifferentwasfoundas71%onaverageagreedthatthosethreebenefitswererealizedfromusingthesystemwhile64%ofcompaniesnotexpressingamotivationinthecategoryagreedonaverage.

The sixth motivation category management of products variants and complexity threebenefitsweregrouped.An interesting finding is thatonaverage70%of thecompanies,whichdidnotexpressamotivationinthiscategoryagreedwiththosebenefitswhileonly50% of the companies expressing a motivation in the category agreed on average.Therefore,higherpercentagesofcompaniesnotexpressing amotivationgrouped in thecategoryagreedwithachievingtheassociatedbenefits.

4.2.3.3 Conclusion

ThefirstRQaimstoidentifythemainmotivationsfortheimplementationofthePCSs.Themainmotivationsweregrouped intosevencategories,whichare to1) improvegeneralcompetitiveness, 2) knowledge management, 3) efficiency in the sales and orderprocesses, 4) efficiency in the production processes, 5) accuracy of the products’specifications,6)managementofproductsvariants and complexity and finally7)other


54

motivations.ThesecondRQaimedtoexpresshowsuccessfulcompanieswereinachievingthe benefits associated with the motivations prior to the implementation. For themotivationcategorizes,generalcompetiveness,efficiency in theproductionprocessandaccuracyof theproducts’specifications,companies thatexpressedamotivationgroupedintothesecategoriesagreedtogreaterextentwiththeassociatedbenefitsbeingrealizedin their companies.Thatmeans that companies that have plans from the beginning toachieve those goals aremore likely to accomplish them. For themotivation categoriesefficiency in the sales and order processes and management or product variants thecompanies that expressed a motivation grouped in to these categories, agreed to lessextent than thecompaniesnotexpressingamotivation into thecategory that thiswasarealizedbenefit.Finally, for themotivation categoryknown asknowledgemanagementnow significant difference could be determined between the companies expressing amotivationinthecategoryandnotintermsofthetowhatdegreecompaniesagreedoftheassociatedbenefitsbeingrealizedbenefitsassociatedwiththeusageofthePCSs.

55

4.2.4 PaperD:Themainchallenges formanufacturingcompanies inimplementingandutilizingconfigurators

Authors:KatrinKristjansdottir,SaraShafiee,LarsHvam,CiprianoForza

Publishedin:SubmittedJournalarticle.


ThechallengesrelatingtoPCSsthathavebeendiscussedintheliteraturearemostlybasedon single case studies, and the importance of the challenges is not usually assessed interms of the overall project’s success. It is therefore difficult to generalize the mainchallengesmanufacturingcompaniesfacewhen implementingandutilizingtheirPCSsortheimportanceoftheindividualchallengesonthesuccessofthePCSs.Thisresearchstudyaims to capture this research opportunity. Therefore, this paper first identifies andcategorizes themainchallengesmanufacturingcompanies facewhen implementingandutilizingtheirPCSs,andthenassessestheimportanceofthosecategories,highlightingthespecificchallengesthatcanbefoundwithineachcategory.ThisresearchthereforeaimstoanswertothefollowingRQs:

WhatarethemaincategoriesofchallengesthatmanufacturingcompaniesfacewhenimplementingandutilizingthePCSs?

WhatistheimportanceofeachcategoryofchallengesthatcompaniesfacewhenimplementingandutilizingtheirPCSs?

WhichspecificchallengeswithineachcategorydomanufacturingcompaniesfacewhenimplementingandutilizingthePCSs?

Toanswer theRQs,a literaturestudyandsurveycomprisingopenandclosedquestionswasadoptedastheresearchmethodin22companies.TheopenquestionswereusedforRQs1and3, to identifyadditionalcategoriesofchallenges thatarenotdescribed intheliteratureandtoidentifythespecificchallengesbelongingtoeachofthosecategories.TheclosedquestionswereusedforRQ2,whichaskedrespondentstoratethe importanceofsixcategoriesofchallengesidentifiedintheliteratureonafive-pointscale.


Based on the first RQ, six main categorieswere identified from the literature review:challengesrelatedtoIT(technical),productmodelling,organisationalchallenges,resourceconstraints,product-relatedchallenges,andchallengesrelated toknowledgeacquisition.ThesecondRQwasconcernedwiththeimportanceofthecategoriesrepresentingthemainchallengeswhenimplementingandutilisingPCSs.Intermsofimportance,IT challenges were the most commonly reported categories of challengesexperiencedat companiesbut theywere ratedashaving the lowest importance.Finally,thethirdRQaimedtoprovidemorein-depthknowledgeaboutthespecificchallenges thatmanufacturing companies facewhen implementing and utilisingtheirPCSsforeachofthecategories.


56

CaseStudy

Thefirstpartoftheresearchexploresthemainchallengesmanufacturingcompaniesfacewhen implementingandutilisingconfigurators.Theanswersprovidedbythecompaniesaregroupedintomaincategoriesidentifiedintheliteratureaccordingtotheirnature.Theresultspresentedinthissectionareaimedtoverifythemaincategoriesidentified,(2)toascertainwhetheradditionalcategorieswouldbefound,andfinally(3)highlightspecificchallenges described within each of the categories. In Table 4-3, the percentages ofcompaniesexpressingachallengegrouped intothedifferentcategoriesofchallengesareshown.

Table4-3Percentagesofcompaniesexpressing(oneormore)challengesgroupedintothecategories

Challengesexpressedbythecompaniesgroupedintocertaincategory %Companiesexpressingachallenge

1.ITchallenges 36.36%

2.Productmodelling 40.91%

3.Organisational 68.18%

4.Resourceconstraints 22.73%

5.Product-related 22.73%

6.Knowledgeacquisition 59.09%

Thesecondpartof theresearch focusedon identifying the importanceof thechallengesexperiencedwhenimplementingandmanagingconfigurators.

Table 4-4 presents the importance of the main categories of challenges that weremeasuredonfive-pointscalerangingfrom1(notimportant)to5(veryhighimportance).Furthermore, the resultsaresummarised into threegroups ranging fromnot important(1),lowimportance(2-3)andhighimportance(4-5).

57

Table4-4Theimportanceofthemaincategoriesofchallenges

Quantitative(RQ.2)Not

important Low High

NotimportantVeryLow

Low

High

VeryHigh

1.ITchallenges 9.09% 54.55% 36.36%9.09% 31.82% 22.73% 18.18% 18.18%

2.Productmodelling 9.09% 40.91% 50.00%9.08% 22.79% 18.18% 36.36% 13.64%

3.Organisational 13.64% 36.36% 50.00%13.64% 13.64% 22.73% 36.36% 13.64%

4.Resourceconstraints 18.18% 36.36% 45.45%18.18% 13.64% 22.73% 31.82% 13.64%

5.Product-related 22.73% 50.00% 27.27%22.73% 31.82% 18.18% 18.18% 9.09%

6.Knowledgeacquisition 18.18% 31.82% 50.00%18.18% 18.18% 13.64% 36.36% 13.64%

IT (technical) challengeswas ranked fourth out of the six in the qualitative part of thestudy(section1.3).However,9.09%ofthecompaniesdidnotconsiderthisanimportantchallenge,54.55%stated that thechallengehasa low impactand36.36%placedahighlevel of importance on the challenges faced in this category. Thus,while IT challengescause problems with implementing the configurators, the majority (54.55%) of thecompaniesconsideredtheseproblemstobeoflowimportance.

Product modelling challenges were expressed by 40.91% of the companies in thequalitative part of the study, placing it the third most common challenge. As for ITchallenges,9.09%ofthecompaniesdidnotconsiderthesechallengesimportant.However,50.00% of the companies placed high importance on this category, and 41% of thecompanies’ratedthiscategorywithlowimportance.

Mostof therespondents(68.18%)expressedorganisationalchallenges inthequalitativepartofthestudy.Inthequantitativepartofthestudy,13.64%ofthecompaniesratedthiscategory as not important, while 36.36% considered it to have low importance and50.00% high importance. Indications were found that when organisational challengeswere rated with high importance, other challenges also became more significant,especiallychallengesrelatedtoresources.

In terms of challenges related to resources constraints, 22.73% of the companiesmentioned these challenges in the qualitative part of the study,making them the leastmentionedcategoryalongwithproduct-relatedchallenges.However, in thequantitativepart of the study, 18.18% of the companies rated these challenges as not important,36.36%placedlowimportanceonthiscategoryand45.45%consideredthesechallengeshighly important.The companies that rated thesechallengeswithhigh importance alsorated organisational challenges with high importance, and a tendency was found forresource constraints to affect the other challenges, thus raising its importance. This


58

indicates that organisational and resources-related challenges have an impact on otherchallenges.Thus,eventhough itwasnotmentionedasoneofthemainchallenges inthequalitativepart,thistypeofchallengehasahigherimpactasanunderlyingchallenge.

Product-related challenges,whichwerementioned by 22.73% of the companies in thequalitativepart,were the leastmentionedchallenges.Alignedwith those findings,mostcompanies see this challenge as an unimportant challenge in relation tomanaging theconfigurators;50.00%ofthecompaniesratedthisasachallengewithlowimportanceandonly 27.27% considered it highly important, making this category ranked by fewestcompaniesasimportant.

Finally, challenges related to knowledge acquisition (59.09%) were the second mostmentioned in thequalitativepartafterorganisationalchallenges.Alignedwith that, thiscategorywasratedwithhighimportanceby50.00%ofthecompaniesandlowimportanceby31.82%ofthecompanies.

4.2.4.3 Conclusion

The empirical findings presented in this study confirm the challenges that have beenpresented in previous literature and add greater insight. However, the sample of 22companiesused in this research limits thegeneralisabilityof the results.Futurestudiesshould thereforeuse a largersamplesize.Furthermore, the importanceof thedifferentsubcategories should be researched further. Nevertheless, the findings are thought torepresent some of the challenges manufacturing companies face and provide greaterinsight into how the companies experience these challenges when implementing andutilising theirPCSs.This researchstudyshould raise theawarenessofothercompaniesregardingthechallengesfacedwhenimplementingandmanagingPCSs.

59

4.2.5 PaperE:How to framebusinessCasesFramework forProductConfigurationsystemProjects

Authors:SaraShafiee,KatrinKristjansdottir,LarsHvam,AndersHaug

Published in: Submitted Journal article, earlier version originally published in:Proceedingsof7th InternationalConferenceonmassCustomizationandPersonalizationinCentralEurope(MCP-CE2016).UniversityofNoviSad,2014.


PCS projects are a particularly complex type of IT system (Salvador and Forza, 2004),which canbe clarified in termsof (1) adiverse setofprocess elements (e.g.machines,operations),(2)ahighvarietyofcomponentpartsandassemblies,and(3)alargenumberof constraints and rules (Zhang and Rodrigues, 2010). Some studies have introduceddifferentstepsthatcanbeusedtodefinetheBCsforPCS,suchasstakeholders’analysisandprocessevaluations(Felfernig,Hotz,etal.,2014).However,existingliteraturefailstoprovideasystematicframeworktoguidethedefinitionofBCsforPCSprojects.Toaddressthisissue,thispaperaimstoanswerthefollowingresearchquestion:

How can we frame business cases to guide PCS projects?

Toanswerthisquestion,wedevelopaspecificBCframeworkforPCSprojects,whichwetest in three PCS projects at two engineering companies. The framework aims to helppractitioners evaluate their investment in different PCS projectswhile prioritising andevaluating the projects in terms of the goals, scenarios and cost benefits before theproject’sinitiation.

This study comprises a literature review, framework development and validation. Theliterature reviewpositionsexisting research in thescientificdiscussion toascertain theneedforourresearch.ThefirstphaseoftheresearchaimstofindasuitableBCfoundationfrom those available for IT projects, and the second step investigates the possibledifferences between IT and PCS projects, which highlight the need for specific BCframeworksforPCSprojects.ThethirdstepdiscussesthespecificrequirementsandtoolsforBCsinPCSprojectsandthelackofsuitableframeworks.Inthefourthstep,weconsiderBCframeworksthathavebeendevelopedforITprojectsingeneralandascertainthetoolsand requirements that are available for PCS projects. After making the final decisionsaboutthesequenceofstepsandproposedtoolsfortheframework,the laststep involvesconductingmultiple case studies tovalidate the usabilityof the framework indifferentcircumstances.


TherearemultipleframeworksforBCsforITprojects,inwhichthereareoverlapsintheelements included.Someof theauthors focuson thestepsofBCdevelopmentat ahighlevelofabstraction(Ashurstetal.,2008),whilesomeareverydetailed(McNaughtonetal.,2010).Basedon the literature, itcanbeconcluded that themainelements forBCs in IT


60

projectscanbedescribed in termsof1)benefitanalysis,2)stakeholder’sanalysis,3) ITrequirements, and4) risk and cost analysis.However, asmentioned, these frameworkshavelimitedusefulnesstosupportBCdefinitioninPCSprojectsbecauseofthedifferencesbetweengeneralITprojectsandPCSprojects.

AlthoughthePCSliteraturedoesnotinadetailedmannerdealwiththedefinitionofBCs,some of the aspects that involve BCs have been described. Benefit analysis is a vitalchallengeforPCSprojectsasitdeterminestherequirementsoftheprojectandshedlightinto project scoping (Hvam et al. 2008; Shafiee et al. 2014). Stakeholders’ analysis, theusers’ expectations and requirements for the system increases as PCS projects becomemoresuccessfulandpopularamongusers(Barkeretal.,1989).Processanalysis isoneofthemajorstepsbeforeinitiatingPCSprojects,wherethesalesandengineeringprocessareredesignedtoincreaseefficiencybythehelpofPCSs(ForzaandSalvador,2002b).Costandriskanalysisarecarriedouttocomparethedifferentscenarios.Inthe literature,thecostestimation forevaluating thesavings fromPCSand furthersensitivityanalysishasbeenconducted(Kristjansdottiretal.,2016).

Frameworkdevelopment

The suggested framework sequence in this paper is based on currently available ITframeworks.However,owingtothegeneralsimilaritiesanddifferencesbetweenPCSandITsystems,theframeworkisdevelopedbasedondifferenttoolsandmethodsavailableforPCSs’BCs.Theframeworkisimprovedinaniterativeprocessusingacasecompanywhilebenefitingfromtheexperiencesandknowledgeofthepractitionersandacademics.

Basedonthediscussionsinliterature,themainBCstepsavailableforITprojects(benefitanalysis,stakeholder’sanalysis,ITrequirements,andriskandcostanalysis)andtoolsforPCSprojects reveal that it ispossible tokeep the sequence and core of the frameworkfrom IT projects.However, the tools andmethods aremore specified for PCS projects.Furthermore,basedontheliterature,PCSprojectscallforastrongprocessevaluationandreportchallengesformoreaccuratecostanalysis.Inmostofthecases,theITstructureandplatformforPCSprojects isfixedanddecidedwhentheconceptofPCSisfoundedinthecompanyandthereisnoneedtodiscusstheITarchitectureineachBCeverytime.Hence,basedon literature,discussionsand initialtesting,theITrequirementstep issubstitutedwiththeprocessandgapanalysiswhichisalsoincludingtheITarchitecturediscussionsifnecessary. Because of the emphasis on inaccurate cost estimation in PCS projects,sensitivityanalysisstepis introduced.ThemainstepsdecidedforBCsframework inPCSprojectsare:

1. Benefit analysis2. Stakeholders’ analysis3. Process analysis, scenario making, and gap analysis4. Scenarios evaluation:

· Cost-benefit analysis· Sensitivity analysis· Risk Analysis

61

Step1:Benefitanalysis

The literature emphasizes on the various benefits gained by using PCSs in differentorganizational settings. Based on the commonly described benefits, the goals of theimplementation have to be aligned with the current strategy and difficulties at thecompany. Identifying the goals and the desired benefits to be gained from theimplementation of the PCS is highly important as it will provide guidelines for thefollowingsteps.

Step2:Stakeholderanalysis

ForbothPCSandITprojects,ingeneral,thecategorizationofrequirementscanbedividedinto two types of requirements: functional and non-functional. Demands for a bettercommunication with stakeholders specifically in PCS projects leads to illustraterequirements through use case diagrams (Shafiee et al., 2014). Furthermore, therequirements have to prioritize based on their importance. The MoSCoW rule can bebeneficialwhen prioritizing the stakeholders' requirements (Bittner, 2002):Must have(Mo),Shouldhave(S),Couldhave(Co),Wanttohave(W).

Step3:Processanalysis,scenariomaking,andgapanalysis

Thespecificationprocessat thecompany isanalysed inorder togetanoverviewof themost important activities, their sequences, and connections, list up the personsresponsible for the different activities, information flows and the processes’inputs/outputs.UnderstandingthecurrentprocessesisafundamentalsteptodesignhowthefutureprocesseswouldlooklikewiththesupportofPCS.

Step4:Scenarioevaluation

Thelaststepoftheframeworkisconcernedwithevaluatingtheproposedscenariosbasedon(Hvametal.2008;Shafieeetal.2014;Kristjansdottiretal.2016):

· Cost-benefit analysis· Sensitivity analysis· Risk analysis

BasedonthediscussionsaboutthecomplexityandunpredictedcostsofPCSprojects,costandriskanalysisforBCdiscussedtobeachallengeasitisaroughestimationandneedsmoreattentionfromacademia(Shafieeetal.,2014).ThefinancialbenefitsforPCSprojectsshould be clear from the beginning and cost evaluation is one of the most importantpurposesofdoingBCs.Cost-benefitanalysisarecarriedouttocomparetheexpectedcostsandbenefits for thedifferentscenariosand ingeneral is a financialmethod tocomparedifferent results from a variety of actions (Haddix et al., 2003).ReturnOn Investment(ROI) iscommonlyusedasacost-benefitratio,which isaperformancemeasureusedtoevaluate theefficiencyof anumberofdifferent investment (PhillipsandPhillips,2010)andithasbeenusedinPCSprojectstodeterminetheprofitabilityoftheseprojects.


62

Inordertotaketheuncertaintyorchanges indifferentparameters intotheaccountandincrease the accuracy of the cost analysis, sensitivity analysis is conducted. Sensitivityanalysis is concerned with representing the certainty which can be apportioned todifferentsourcesofuncertaintyinitsoutput(Saltelli,2002).

Casestudies

The proposed framework for BCs of PCS projects was tested in two engineeringcompanies.Workshopswere conducted to introduce the proposed framework and thetoolssuggested inthe individualstepstoallthestakeholders.Finally,feedbackmeetingswere held as semi-structured interviews to collect knowledge about the team’ssatisfactionwith the new framework. The interview questions aim to obtain a generalevaluationofthebenefitsandchallengesassociatedwiththeframework’sperformance.

Benefitanalysis

The goals of the project should reflect the benefits of implementing the PCS from thestakeholders’ viewpoint. The main benefits identified in this step are concerned withreducing the time needed formeetingswith experts and to clear the task assignmentbeforefurtherdecisionsaremade.Thedeterminedgoalsdifferforthecompaniesbecausetheyreflectthecompanies’currentoperationalchallenges.

Stakeholders’analysis

The benefits of using these methods are (1) a full understanding of stakeholders’requirements and (2) improved communication and task delegation, which results insavingresourcesandtime.Thestakeholdersreportedthemainobstacleinthisstepastheunfamiliaritywiththeintroducedtools.liststhemainresultsfromthecasesinthisphase.

Processanalysis

A common understanding of the current processes provided learning points for thestakeholders. In all cases, the numbers of redesign loops were noticed due to aninsufficient flowof information in thevariousstepsof theprocesses. Furthermore, thegapanalysisprovidedaneffectiveoverviewofthefuturestateatthecompanies.Trainingsessionswerepreparedtoensureemployeesknewhowtousethenewmethods;however,stakeholders considered this a time-consuming process.The gap analysis also helps tocommunicate theneed for implementing thePCSs inallcasesand thereby increases thestakeholders’commitmenttotheproject.

ForCase1,theexpectedtimesavingscreatedbyautomatingtheprocesswillnotcoverthecost of man-hours saved because the quantity of plants sold every year is too low.Therefore,thesavingsarecalculatedbasedonsellingonemoreplantperyear.However,Cases 2 and 3 will save on man-hours because of the higher quantity of products orprocesses sold each year.The cost is calculated as theproject cost,which includes thedevelopment and implementation and the yearly running cost, such as licenses andmaintenanceactivities.Theriskofsystemavoidancefromusershighlightedtheneedforgoodmanagementtochangeemployees’mind-sets.Thesolutionwastoinvolveallfuture

63

system users from the beginning of the project to create a feeling of ownership andcommitment.Anotherriskrelatestothebenefitrealizationoftheprojectandtrustintheaccuracyandstabilityofthecalculations.Thesolutionwastopresentvisualizationsofallthedataandformulasinthesystemtothedomainexpertsandtoinvolvethemintestingthesystem.

4.2.5.3 Conclusion

This paper proposed a framework for developing BCs for PCS projects, based on theavailable literatureforITandPCSprojects.Thesuggestedframeworkandthesuggestedtoolsshouldhelpthewholeteamtofocusonandprioritisethegoals,specificstakeholders’requirements,analysisanddesignofthecurrentandfutureprocesseswherePCSisused,andtheevaluationoffuturescenariosbasedoncostbenefitanalysis,andsensitivityandriskanalysis.

The framework tested in three PCS projects at two engineering companies proved theapplicationoftheframeworkindifferentprojectsandindifferentcompanies.Tovalidatetheframework,wehaveclarifiedtheapplicationoftheframeworkandthedifferentstepsinvolved.Allthreeprojectsaimedtoreducethecomplexityofthecurrentprocessesandachieve economic benefits from implementing a PCS. The results from testing theframeworkandobservationsofthecasestudiesshowtheinterestamongthewholeteaminusingsuggestedBC frameworkaswellas itschallengesat thecompany. ItalsoshedslightonunclearpointsinearlyphasesofthePCSprojectssuchastheexpectationfromthesystem.


64

4.3ScopingofproductconfigurationsystemsThissectioninvestigatesthepossibilitiesofassessmentsoftheSection3.5thatrefinedtheimportanceofscoping inPCSs.Therefore, thissectionaddresses thescopingofPCSsbyintroducingaframework.ThissectionaimstoanswertheRQ2.

4.3.1 PaperF:ScopingaProductConfigurationProjectforEngineer-to-OrderCompanies

Authors:SaraShafiee,LarsHvam,MartinBonev

Published in: International Journalof IndustrialEngineering andManagement (IJIEM),pp. 207-220, earlier version originally published in: Proceedings of 6th InternationalConferenceonmassCustomizationandPersonalizationinCentralEurope(MCP-CE2014).UniversityofNoviSad,2014.


InordertostartaPCSproject,there isaneedfordeeperunderstandingofhowtoscopethese kind of projects. This paper reflects upon the available literature and refines itwithin the ETO companies and in the context of process plants and machineryapplications. The PCS deals with broad range of knowledge and seeks for high levelmanagementskillsandplanningtoaligndifferentresourceswithdifferentexpertise.TheresearchgroupexperiencesrevealthelackoffocusonPCSevenfromthefirststepstothefinalrelease.Thislackoffocusandplanningleadstobothpoorperformance,increasesoftimeandresourcesconsumptioninallphaseofthePCSs.Actinguponthischallenge,thispaper suggests a framework for scoping PCSs in companies with complex and highlyengineered products. This framework is based on a general and well-establishedframework for scoping IT-systems and on specific tools for managing PCSs. The mainresultswillbediscussedinthefollowingsubsections.


BasedonliteratureinthefieldofPCSandthebasediterativelytestingtheframework,thefollowingstepsareproposedforscopingPCSprojects:

1. AimsandpurposeforthePCSandoverallprocessflow2. Theidentificationofstakeholdersandtheirrequirements3. IT-architecture incl. flow in thePCS,UI, input,output, integrations,andthemain

functionalityofthePCS4. ProductsandproductfeaturestoincludeinthePCS,includinglevelofdetail5. A project plan including resources, time table, modelling approach, test and

development,systemmaintenance,etc.AimsandpurposefortheCSandoverallprocessflow

ThispartwilldiscusstheoverallaimandpurposeofimplementingPCSsasthevisionforspecific PCSs projects, which can differ across different projects. However, there are

65

generalaimsandpurposeswhichareusually thesame,even though the importancecanvary,whichare:leadtimeandresourcesreduction,automaticdocumentgeneration,fewererrors and higher quality, higher independency and empowering sales resources,knowledgemanagement anddocumentation, standardization andmodularizationof theproductrangeandetc.

Stakeholders’identificationandtheirrequirements

Thesubsequentstepsoftheprojectshouldbebasedonthestakeholders’needs.Thereisaneed to analyse the stakeholders in the very first steps of the project to support theprioritization of the individual requirements (Tiihonen et al., 1996a). One of the firstmethod to illustrate thestakeholders’requirements isTO-BEprocessand theotheroneforthispurposecouldbeusecasediagramsfromRationalUnifiedProcess(RUP)method(Kruchten, 2007). Flowcharts demonstrate the workflow and task floating betweendifferentstakeholdersby the focuson theprocess.Usecasediagrams toolsexpress therequirements and the actors involved in the project, while the same use case can beutilizedinsystemanalysis,design,implementationandtesting.

IT-architecture

IT architecture addresses the technical aspects and structure of the PCS. there aretechniquestohandletheproductorprocessmodelsforPCSespeciallywhenmanagingbigsizedcomplexprojectssuchasrecommendationsystems(Tiihonenetal.2014).

Productsandproductfeatures

Havingagooddescriptionofproductsandprocess features leads toanefficient levelofdetails in PCS which reflects on some of the technical and business aspects such as:avoidingcomplexityinPCS,knowledgemanagementanddocumentation,savingtimeandresources,efficientcommunicationwithdomainexperts.

Projectplanandmodellingapproaches

The RUP method is a popular iterative software development process which iscustomizableforspecificorganizationsorprojects(Kruchten,2007).ThismethodandthetoolsintroducedsuchasmodellingtoolsorcomponentbaseddevelopmentcanbeusedforPCSprojects(Figure4-5).


66

Figure4-5UnifiedProcess(Kruchten,2007)

Oneofthemajorsteps istofindanefficientstructured“modelling”tool,suchasProductvariantmaster(PVM) (Hvametal.,2008) inorder tounderstand theproducthierarchyand facilitate the communication process. CRC stands for “Class, Responsibility andCollaboration” and these cards are used to define the classes and hierarchy and otherdetailed information to be added to the PVM (Hvam et al., 2008). Documentation andmaintenanceishighlydependingonmodellingandthetoolsusedinthisstepsuchasPVM.This tools can be used for communication, clarification and documentation andmaintenanceofthesystem.

“Component-baseddevelopment”isabouttosplittheprojectintosmallercomponentsanddeliveritinsmallversions.Acomponentbasedstrategycanbeveryhelpfulespeciallyincomplicatedandhighlyengineeredprojects(Felfernigetal.2014)andhelptodeliverandtest the system iteratively.Method recommended tobeused in this step areweightingtables,whereeachrequirementisratedagainstseveralspecific,weightedprojectsuccesscriteriaandascoreiscomputedtorankthepriorityoftherequirements(Wiegers,1999).“Testing”isincludedintheiterationtemplate,anditisascriticalforPCSsasitisforotherITprojects.For “projectplanning”, it ispossible tohave twodifferentplans:1.The firstplanisacoarseprojectplanwithstartandenddatesofallphasesanditerations.InFigure4-6thegeneraliterationprocessspecificallyforPCSprojectsisillustrated.

67

Figure4-6ThegeneraliterationprocesstemplateforPCSs

Casestudy

Theframeworkdevelopment isbasedon literatureand iterativetestingandexperiencesfrom industry where the authors have been involved in more than 20 PCS projects.Thereafter, the proposed framework has been applied in a ETO industry to assess itsfunctionality and possible challenges in a real project from the planning to the releasephase.The case company is an international company specialized in the production ofheterogeneouscatalystsandinthedesignofprocessplantsbasedoncatalyticprocesses.

Step1:AimsandpurposeforthePCSs

Themainaimsoftheprojectdefinebasedonthechallengesoftheselectedcase(aprocesschemical plant) regarding complexity, empowering the local offices, knowledgedocumentationandneededresources.Inordertodescribefuturescenarios,itisnecessarytohave a comprehensiveoverviewof the current situation. Salespersons are currentlyusingexcelsheetsandacomplexhomemadecalculationsystemasthemainfoundationforthe creation of technical proposals.Based on the current processes different scenariosweremadewherethePCSisused,thechosenscenariofortheprojectshowninFigure4-7.


68

Figure4-7TO-BEprocessexample

Step2:Stakeholders’identificationandtheirrequirements

Forthisspecificproject,thestakeholdersarethesalesstaff,costestimators,processandchemical engineers, product developers, marketing staff, regional offices and generalmanagerswithdifferentrequirementsandinputsfortheproject.Figure4-8illustratesanexample for use case diagram used in this project to demonstrate the stakeholdersrequirements.

69

Figure4-8Usecasediagramexample

Step3:IT-architecture

Examplesof themain functionalitiesof thePCSare:capacitydimensioningof theentireplant, cost estimation for the machineries, calculations of engineering hours, energyconsumption and etc. From IT aspects, different plugins and integration to the maincalculationandsimulationsystematthecompanyarerequiredto introduceafasterror-freesolutionforthestakeholders.

Step4:Productsandproductfeatures

Figure4-9showsthereverseengineeringprocessusedforfindingtheneededinputsandmanaging the level of details. As an example, stakeholders asked for price calculationsheets, which are a combination of all component prices according to their internalselectionsandsizes, theengineeringhoursbasedon theplantcomplexities,consultancyhours,transportationexpensesdependingonthesizeandtype,insuranceanddestination.


70

Figure4-9Themodelsubjectlayerandintegrationwithothersystems

Projectplanandmodellingapproaches

Concerningthecomplications inprojectandthe importanceofhavinganuptodatePCS,documentingofthesysteminitiatedinearlyphasesoftheproject;whileconsideringPVMasthemodellinganddocumentationtool(Figure4-10).

Figure4-10TheinitialPVMstructureexample

In order to break the project to smaller pieces and to prioritize the products andfunctionalities to be included in the PCSs, it is recommended to use weighting tables(Wiegers, 1999) as it provides structured approach. In Table 4-5, the calculations ofweightingtableareassessedusingtheexplanationsfromWeigers(1999).

71

Table4-5PriorityTableexamplebasedonWeigers(1999)

Feature RelativeBenefit

RelativePenalty

TotalValue

Value%

RelativeCost

Cost%

RelativeRisk

Risk% Priority

Product1 5 5 15 21.1 2 12.5 5 31.3 0.116Product2 9 9 27 38.0 5 31.3 2 12.5 0.209Product3 5 2 12 17.0 3 18.7 3 18.7 0.151Product4 4 1 9 12.6 4 25 5 31.3 0.049Product5 2 4 8 11.3 2 12.5 1 6.25 0.361

Totals 25 21 71 100 16 100 16 100 --

4.3.1.3 Conclusion

This paper clarifies that having a standard framework for implementing PCSs has aremarkable effectondecisionmaking in the earlyphasesof theproject.The suggestedframeworkforscopingaPCShasbeentestedinacasecompany.Inthecasecompany,theframeworkprovedtobeusefulfortheprojectteaminsupportinganearlyclarificationofthePCSs.ThedevelopedframeworkformedasolidbasisforthesubsequentPCSsprojectandhelpedtofocusandgivepriorityonlytoneededpartsofthePCS.Howeveradditionalresearch is required regarding the maintenance and testing stages. The case studymoreover revealedsome challenges in identifyingandprioritizing thestakeholdersandtheir requirements,which is a field that needsmore research in the future. Finding asolution for the documentation and maintenance part of the PCS also need furtherresearch. Furthermore, the suggested framework needs to be tested in a number ofcompaniestofurthervalidateit,andtotestiftheframe-workcouldbeusedalsoinotherkind of companies than only ETO companies with complex and highly engineeredproducts.


72

4.4KnowledgemanagementThissection investigatesthepossibilitiesofassessmentsoftheSection3.6whichrefinedthe importance of knowledge management in PCS projects. Therefore, this sectionaddresses the scoping of PCSs in terms of the complexities in determining the neededknowledge, the process of knowledge acquisition, validation and maintenance. ThissectionaimstoanswertheRQ3.

4.4.1 PaperG:How toscopeconfigurationprojectsandmanage theknowledgetheyrequire

Authors:SaraShafiee,KatrinKristjansdottir,LarsHvam,CiprianoForza

Published in: In the second round of review in journal, earlier version originallypublished in:ProceedingsofWorldConferenceonMassCustomization,Personalization,andCo-Creation,Montreal,Canada,2015.


OneofthemainandfirstchallengesinPCSprojectsfromtheplanningtothemaintenancephase ishow tomanageandhandle themassiveamountofknowledgeneeded in thesetypes of projects. Knowledge Management (KM) in PCSs is one of the most time-consuming tasks for both domain expert and the configuration team. The increasedcomplexity of products increases the number of products features to bemodelled andmaintained in the PCS (Ardissono et al., 2003). The configuration knowledge is oftenspreadwithin the companies among various experts in different parts of the products(Hvametal.,2008). Othervaluablesourcesofknowledgearealsoavailable in internalsoftwaresystemssuchasERPsystems,calculationsystemsandspreadsheetsdocuments(Friedrich et al., 2014). Therefore, early in the development phase of a configurationmodel, aknowledgeacquisitionandknowledgecleansingstage isrequired tocentralize,modeland formulate theproductorprocessknowledge(Felfernig,Hotz,etal.,2014). InPCSs,theKMincludesthelifecycleofknowledgefromthetimeofacquisition(Tiihonenetal.,1996a)tomodelling,validating,testing(Hansenetal.,2012),andfinallyupdatinganddocumentingphase(Haug,2009).Infact,itisveryimportantthattherequiredknowledgeforthePCSisavailablefortheprojectteam(Turley,2007).

This study presents a framework, in which the KM process can scope the projectknowledgeandmanagethelargeamountsofcomplexknowledgeofPCS.Thequestiontobeansweredwillbe:

Whatare themost importantavailable steps related toKM inPCSprojects,andhowtoformulateaframeworkforPCSprojects.

BasedonthechallengesinKMduringPCSs,aliteraturestudytofindaKMframeworkforPCSprojectsperformed,whichrevealsagapintheliterature.Inthenextphase,adetailedliteraturestudywasconductedtofindKMframeworksforITprojectsingeneral.ThenextstepoftheliteraturereviewaimedtoidentifytoolsandmethodsthathavebeenproposedfortheindividualphasesoftheKMinPCSprojects.Afterwards,thedevelopedframework

73

was testedandvalidated inan iterativeprocess inonecasestudy.Thiscasestudywasused to improve the quality and usability of the proposed framework and rectify thedrawbacksbeforetestingitonmultiplecasestudies.


TheliteraturereviewshowstheavailableframeworksforITprojectsandatthesametimereveals the differences between IT and PCS projects. Different authors discuss thedifferentstep/sbutthereisasignificantlackofaframeworktorelateandsequencealltherequired steps. PCS relates to a great deal of knowledge that represents complexityrelations among components or modules, such as configuration rules and assemblyconstraints(Jinsongetal.,2005).Anumberof frameworksareavailable in literatureonKM for ITprojects (Table4-6).Different frameworksdiffer from 3phases/actions to 6phases/actions based on the level of abstraction. Even though these frameworks usedifferenttermstoaddressvariousactions/phasesoftheKMinITprojects,theypresentanumber of similarities. There are several differences between an IT projects and PCSprojects.The scope of PCS projects compared to other IT projects is one of them.Theseconddifference is related to thedetailsof each step for ITprojects compared toPCSprojectswhichmakesustodevelopanewframeworkforPCSprojects.

Table4-6KMframeworksforITprojects.

Authors Actions/phases included in KM framework

BasiliandWeiss(1984)

1. Establish the goals of knowledge selection, 2. develop a list of questions of interest, 3. establishknowledge categories, 4. design and test knowledge collection form, 5. collect and validate gatheredknowledge, 6. analyse the knowledge.

KuczaandKomi-Sirviö(2001)

1. Identify need for knowledge, 2. share existing knowledge, 3. create new knowledge, 4. collect andstore knowledge, 5. update knowledge.

Komi-Sirvioetal.(2002)

1. Define scope and requirements for knowledge capturing, 2. acquire knowledge, 3. package knowledge.

Rodríguezetal.(2004)

1. Identify knowledge sources and resources, 2. identify kind of knowledge, 3. identify knowledge flows(graphical modelling techniques), 4. identify faults in the knowledge flow (analyse knowledge).

Reichetal.(2012) 1. Knowledge stock (relevant domain knowledge of the IT team, the business team and the governanceteam), 2. enable the environment (combination of the technological and social aspects of a project thatfacilitate knowledge practices), 3. knowledge practices (actions taken to map and share knowledge withinand between the IT, business and governance teams in an IT-enabled business project).

Lech(2014) 1. Identification (determine knowledge sources and resources), 2. acquisition/creation, 3.transfer/dissemination, 4. storage/capture, 5. use/application.

McGinnisandHuang(2007)

1. Socialisation (scoping and deliverables), 2. Externalisation (formalise the knowledge to be explicit), 3.Combination (knowledge clarification and team communication), 4. Internalisation (new deliverables,improved documentation, improved training, and process refinements).


AsthePCSgrowsandgetsmoresuccessful,theusers’expectationsincreaseaswellastherequirements for the system (Barker et al., 1989); where further development andchangesareofgreat importanceforPCSs.Thathighlightstheneedofhavingan iterativeframeworkfortheKMasillustratedinFigure4-11.Inthefirststepthescopeofthesystemisdetermined,secondlytheknowledgeacquisitioniscarriedout,thirdlytheknowledgeisstructured by use of special modelling techniques and validated and the last step is


74

concernedwithdocumentationandmaintenance.The individualstepsof the frameworkareillustratedinFigure4-11alongwithrelationsbetweenthesteps.

Figure4-11TheproposedframeworkforknowledgecollectioninPCSs

Step1:ScopingofthePCS

Project goals aredeterminedby identifyingstakeholders' functional andnon-functionalrequirements.Thisstepaimsatincreasingtheunderstandingoftheprojectbyidentifyingthemainstakeholders’requirements(BasiliandWeiss,1984).BasedontheRUPmethods,the stakeholders and their requirements can bedrawn up by using process flowcharts(ComptonandJansen,1990)andbyutilizingtheusecasediagramsbasedonRUPmethods(Kruchten,2007)(Figure4-12).

Figure4-12.Anexampleofausecasediagram(Hvametal.,2008).

Asdiscussedearlier inPaperA, for largeandcomplicatedproject thecomponentbasedapproach isveryhelpful(Briand,2003).Methodrecommendedtobeused inthisstep isweightingtable,whereeachofthecomponents isratedagainstseveralspecificweightedprojectsuccesscriteriaand a score iscomputed to rank thepriorityof thecomponents(Wiegers,1999).

Step2:Knowledgeacquisition

75

OnemethodofknowledgeclusteringorcategorizationinPCSsistodeterminetheoutputknowledgeaccording to thestakeholders’requirementsandsubcategorize themstepbystep.Forexample,regardingtheoutputsrelatedtointegrations,alltheneededinputsandresources should be determined and categorized as integration category. Listing thesourcesandresourcesoftheknowledgecreatesvaluesincategorizingtheknowledgeandhelpsindelegatingthetaskstodifferentresources(Tiihonenetal.,1996a).

Step3:Modellingandknowledgevalidation

Therearevarietymodellingtechniqueswhichcanbeusedformodelling,communicationand further documentations in this step.Recommendedmodelling techniques are PVMandCRCcards.

Step4:Documentationandmaintenanceoftheknowledge

This step is of great importance as keeping the knowledge up-to-date is a key successfactorinPCSprojects.Modellingtechniquesaretobeusedasdocumentationtoolsbesidethe taskofcommunicationandvalidation.Thereare researchworkswhere theproductmodelling process is supporting by using software supports to integrate the differentmodellingtechniques(PVMandCRC)(HaugandHvam,2007)(Shafiee,2015).

4.4.1.3 Conclusion

Due to the differences between IT and PCSs, the available frameworks for KM in ITprojectsarenotsolvingtheKMchallengesforPCSs.TheproposedframeworkarrangetheKMstepsproposed in literatureforPCSsbyusingavailableKMframeworks in literaturefor ITprojectsas abase.Therefore,basedon literature,KMprocess forPCSs isdividedinto four steps (modelling and validation, knowledge acquisition, project scoping,documentationandmaintenance).

ThesuggestedframeworkhasbeentestedintwoindustrialETOcompaniesonfourPCSsprojects intotal.Theconfigurationteams involvedinthedevelopmentandtestingoftheframeworkhaveexpressedwillingnesstousetheframeworkinfutureprojectsasitsavedtimeandresourcesintheproject.Thecrosscasecomparisonshowedthattheframeworkbehaved differently in company A and in companyB.This could be traced back to thedifferences in the culture present in company A and Company B as previous studiesillustratesresultsandjob-orientedcultureshavepositiveeffectsonemployeeintentionintheKMprocesswhereasatightlycontrolledculturehasnegativeeffects(ChangandLin,2015). Inorder tovalidate the framework,Version 1andVersion 2of thesameprojectwere compared in order to eliminate the threat of possible not-controlled influencingfactors.

Theabilityoftheframeworktocopeupwithhighlyengineered,complexproductsinETOcompanies indicates that it could also be used in PCSs,which involve less complexity.However, the necessity of applying such a structured framework in smaller projects isquestionable and needs further testing. There are some limitations on frameworkregardingthesuggestedtoolsandtechniques.Thefutureworksshouldspecificallydevote


76

to findmore efficient and simpler tools or techniques to be used in each steps of theframework.

77

4.5ModellinganddocumentationThis section investigates the possibilities of assessments of the Section 3.7 thatemphasized the importance of modelling and documentation in PCSs. Therefore, thissectionaddressesthemodellinganddocumentationofPCS intermsofvisualknowledgemanagement techniques and proposes an automatic solution for documenting theknowledge in PCS projects; which leads to better communication and validation ofknowledgeandsystem.ThissectionaimstoanswertheRQ4.

4.5.1 PaperH:Thedocumentationofproductconfigurationsystems:AframeworkandanITsolution

Authors:SaraShafiee,LarsHvam,AndersHaug,MichaelDam,KatrinKristjansdottir

Published in: International Journal of Advanced Engineering informatics,doi:10.1016/j.aei.2017.02.004,V.23,pp.163–175.earlierversionoriginallypublished in:Proceedings of 22nd Euroma conference, Neuchatel, Switzerland. Industrial versionoriginallypublishedin:IEEE/WIC/ACMInternationalConferenceonWebIntelligenceandIntelligentAgentTechnology(WI-IAT),Singapore.


PCSdocumentationincludesmodelling,maintainingandupdatingtheproductmodel,andstoringallinformationrelatedtotheproducts’attributes,constraintsandrulesinsidethePCS.PCSdocumentationrequiresthecontinuous involvementandcommunicationofthedomainexperts.OneofthemainchallengesofusingPCSsisthatthemodelscanbepoorlydocumented,incomplete,difficulttounderstandandoutdated.Forindustrialcompanies,itis thereforecrucial tohaveanefficientsystem fordocumenting theattributesandrulesmodelled insidePCSs tounderstand the consequencesof changing theknowledgebaseandtoenablediscussionsbetweenPCSdevelopersanddomainexperts.

Thispaperpresentsanapproach(frameworkandITdocumentationsystem)foranagiledocumentation process for the complete PCS lifecycle in order to validation of initialmodelling,maintenanceandfurtherdevelopments inthesystem.Theaim isthereforetopropose a structured documentation process which automatically generates thedocumentationasrequiredforcommunicationwithdomainexperts.Thisprocessshouldensure that no knowledge is duplicated and all the maintenance and updates arepreservedinoneplace.

To support the framework, we develop an IT system that is capable of retrievinginformation available in thePCS andpresenting it in a structure that corresponds to aproductmodel,whichcanbeusedtodocumentthesystem.TheframeworkpresentstheknowledgeandchangesinthePCS(informationmodel)andgeneratesthedocumentationautomaticallybasedon theknowledge stored in thePCS.Alignedwith the focus of thestudy,thefollowingresearchquestionshavebeendeveloped:

How do we best frame the documentation framework for PCS projects usingagile documentation methods?


78

How can we develop an IT system to support the developed agile documentationframework of PCSs automatically?

To answer the research questions, in the first phase of the research, we studied theavailable literature on product modelling, documentation methods and agiledocumentationmethods. For this study, the product (phenomenon)models,which arealso used for documentation, are the Product Variant Master (PVM) and ClassResponsibilityCollaborationCard(CRC)cards(Hvam,etal.,2008).WechosethePVMandCRCcardsmethodsbecausetheresearchteamandthecasecompanyarefamiliarwithandhaveexperienceusingboth.Afterdevelopingtheframework,weexaminedthe literatureand investigatedallmethodologiesand requirements foran ITdocumenting toolwith aspecial focusonPCSprojects.Figure4-13 illustrates thecompleteprocessofdevelopingthe framework and the IT documentation system and explains the research steps andmethods supporting differentphases used.To evaluate the functionality,we tested theproposed framework and thedocumentation system.The testing aimed todiscover (1)whether thestructure,attributesandconstraintscouldbe extracted from acommercialPCS,(2)howtoextracttheknowledgefromPCSand(3)whetherthemodelsmadebasedon this solution were sufficient for domain experts and the configuration team at acompany.

Figure4-13MethodologyfordevelopingtheframeworkandtheIT-baseddocumentationsystem.


The literature review retrieve information needed to determine the methodology fordeveloping the framework and the IT documentation system. The paper focuses onmodelling and documentation in PCSs and the agile documentation principles fordeveloping the framework. Then, the research is devoted to the requirements for

Availablemodellinganddocumentation

techniques

Steps Methods

AgiledocumentationframeworkforIT

projects

-Specificandgeneral

requirementsforanITdocumentation

toolforPCSs

-Experience-Literature

-Literature-Analytical

thinking


-Analyticalthinking


RQ2.DeveloptheITsystembasedonthe

frameworkandrequirements,testandvalidatetheframework

andITsystem

-AppliedCasestudy

RQ1.Developtheframework

-ProductmodelandstructureinPCSsandpossibilityof

reusingthesestructures

-Analyticalthinking

-Experience- Literature

Methods

Methods

Steps

Steps

-TechnicaldetailsfordevelopingITdocumentation

system

-Analyticalthinking

-Experience- Literature

Frameworkdevelopmentrequirements

ITdocumentationsystemdevelopmentrequirements

79

developing an IT documentation system, the structure inside PCS to support therequirement for generating the documentation system and the technical details forgeneratingthisITtool.


This study presents a framework for agile documentation of PCSs, where the ITdocumentation of the product model supports the framework. The main idea of thisframeworkisbasedontheapproachofanagilelifecyclemodel(Ambler,2002).Thefocusistovisualizethecreationofatemporarymodelversionandmakeitpermanentwhenitisprovedtobeviableandtrustworthyasadocumentationsystem.

Theframeworkdemonstratesthat,first,theinitialproductmodelshouldbebuiltbasedonspecificmodelling techniques (in thisstudy,we recommendusingPVMandCRCcards).Thisphaseisthetemporaryphaseofdocumentation.Basedontheinitialmodelling(PVMandCRCcards),themodelisprogrammedintothePCS,andmodelssimilartothePVMandCRCcardsaregeneratedfrom thePCSdirectlyby thedocumentationsystem.For futureversionsofthePCS,updateswillbemadeinsidethePCSandtheupdatedproductmodelswill be generated directly from the PCSwithout spending time on it. This permanentmodel will then be stored inside the PCS, and the maintenance and updating will beconductedinthePCS.Figure4-14demonstratestheprocessesfortheagiledocumentationofthePCS.

Figure4-14TheprocessofgeneratingdocumentationfromthePCS(thephotoswillbeshowninthecasestudy).

Thedocumentationmodelhasbeengeneratedbasedonthestructure insidethePCSandtheconfigurationengineeronlyhastouploadhisPCSmodeltothedocumentationsystem.The programmers provide the team with the ability to control their models, such asshow/hidedifferentpartsorprovidinguserswithdescriptions.Themodel isshownasatree structure inside the documentation system. The models that are uploaded to thedocumentationsystemarestoredinthehistoryasdifferentversions.Domainexpertscanaccessthesystemdependingontheirrightsmanagedbytheadministratorofthesystem.


80

Theycanseethetreestructureandnavigatebetweenvariantsandconstraintsandaskforchangestoanypartofthemodelbysendingachangerequest.

DevelopmentoftheITdocumentationsystem

Figure4-15showsacommercialPCSwithstructuresthatcorrespondtothePVMandCRCcardswhichwasfoundbyevaluatingthemainsuccessfulcommercialPCSsonthemarket.Allofthecomponentsandsub-components,includingalloftheattributes,constraintsandrules,are listedinatreestructure.Thegoalistotransferandtranslatealloftherelevantknowledge into the documentation system and generate PVMs and CRC cardsautomatically.

Figure4-15TherelationsbetweentheproductmodelandcommonstructureincommercialPCSs.PVMandCRCCardsstructure(a&b)(Mortensen,etal.,2008);traditionalstructureavailableincommercial

PCSs(c).

Theplatformforthedocumentationsystemisacontentmanagementsystem(inthiscaseMicrosoftSharePoint).Theoriginaldocumentation formatgenerated fromacommercialPCS model (in this case Tacton) is based on xml format; by renaming the model fileextension to .xml it can be read/ parsed. Figure 4-16 shows the interactions betweenSharePointandtheUIRenderer,whichtogetherarecalledthedocumentationsystem.Forintegration,weusedtheSharePointJavaScriptObjectModel(JSOM).ThisJavaScriptAPIisused for receiving lists and list-items from SharePoint and creating requests back in

81

SharePoint. The following is an example of how a request is registered to the requesttrackinglist:

var clientContext = new SP.ClientContext(‘/sites/20’);

var oList = clientContext.get_web().get_lists().getById('{7A6EFEDF-5D65-4241-A6E0-E84CF2F3312D}');var itemCreateInfo = new SP.ListItemCreationInformation();var oListItem = oList.addItem(itemCreateInfo);oListItem.set_item('Title', $('#component_name').text());oListItem.set_item('Body', $('#rfctext').text());oListItem.set_item('Model', 'testtest2.dk');oListItem.update();clientContext.load(oListItem);clientContext.executeQueryAsync(function() { alert('Item created'); $('#rfctext').text(‘‘)}, function(sender, args) { alert('Request failed. ' + args.get_message() + '\n' + args.get_stackTrace());});

Figure4-16InteractionsbetweentheContentManagementSystemandtheUIRenderer(documentationsystem)

Casestudy

Theproposed approach (framework and the ITdocumentation system)were testedonPCSsprojects to assess the functionality.The companyhas fiveoperationalPCSswhichsupportthesalesandengineeringprocessestovaryingextents;itthereforeenablesustotestthedocumentationindifferentenvironmentsatthesamecompanybutwithdifferentlevelsofcomplexity.Thesystemwastestedduringthedevelopmentandoperationofallfiveprojects.TheexamplesinallpartsofthecasearerelatedtoProject2.Eachworkshoplastedbetween30minutesand1hour.

Theinitialphaseinvolvedproductmodellingtodevelopanincreasedunderstandingoftheoverallproduct, the individualcomponentsand the relationbetween thesecomponentsusingPVMandCRCcards.ThePCSmodelcontainsalloftheknowledgerequiredforthedocumentation includingthetreestructureoftheproductandallthevariantsandrules.


82

Figure4-17illustratesthetreestructureofthePVMandtheCRCcardsasgeneratedbythedocumentationsystembasedonthePCSmodel.

Figure4-17KnowledgegenerationinthePCSandPVM/CRCstructureinsidetheITdocumentationsystem.

4.5.1.3 Conclusion

Wesuggestaframeworkbasedontheagilemodelandpreviousliterature,throughwhicha temporarymodelcanbecreated.Subsequently, thePVMandCRCcardsaregeneratedfrom the PCS, based on the structures, attributes and constraintswithin the PCS. ThispracticalapproachhasbeendevelopedasanITdocumentationsystemthatsupportstheframework’sprinciples.Theproposed ITdocumentationsystemwas implementedusingthexml(standard)andJavaScriptformatatthecasecompany.

Structuredinterviews,conductedinthetestingphasetomeasuretheusabilityofthenewITdocumentationsystem, indicated that theproposedmethodwasuseful in facilitatingthe understanding and debugging of the system. However, the approachmight not beapplicabletoallkindsofcommercialPCS(duetothedifferingstructuresofPCS).Variouschallengesmightarise,dependingon themind-setsandculturesatdifferentcompanies.Wesuggest,therefore,thatfutureresearchshouldtestthesystemusingdifferentprojects,companies,typesofsoftwareandplatforms.

83

4.5.2 Paper I: Product configurators: A study of the impact ofapplyingITformalproductmodellingtechniques.

Authors:LarsHvam,KatrinKristjansdottir,SaraShafiee,NielsHenrikMortensen,

Publishedin:InthesecondroundofreviewinJournal,2017.


This research opportunity is captured in this study; the impact of using a structuredmodellingtechnique,ortheCPMprocedure,iscomparedtolessformalornonemodellingtechniques. The impact is analysed in terms of improved control of product variants,increased availability of the product knowledge in the organizations and employeesatisfaction. The study is limited to experiences of using the CPM procedure and doesthereforenotstudy thequalityof theCPMprocedurerelative toothersuggested formalmodellingtechniques.TheresearchfocusisthereforeontheexperiencesofusingtheCPMprocedure relative to using less formal modelling techniques—i.e. structured bill ofmaterials(Jiaoetal.2000;Hegge1992)ornon-formalmodellingtechnique.Alignedwiththefocusoftheresearch,thefollowingpropositionshavebeendeveloped:

Proposition1:Theuseofaformalmodellingtechnique(CPMprocedure)willresultinimprovedcontrolofproductvariants.

Proposition2:Theuseofaformalmodellingtechnique(CPMprocedure)willresultintheincreasedavailabilityofproductknowledgeinorganizations.

Proposition3:Theuseofaformalmodellingtechnique(CPMprocedure)willresultinincreasedemployeesatisfaction.


Theresearchrevealsthatin18companies,sixcompaniesusedtheCPMprocedure,sixuseothermodelling techniques(e.g.structuredbillofmaterials)and theremainingsixusednon-formal modelling techniques. The research has illustrated that formal modellingtechniques (CPM procedure) are used in the larger companies and in PCSswhere thesystemincludelargernumbersofrules,attributesandintegrations.Furthermore,threeofthe six respondents from the companies using the CPM procedure claimed that theystartedtousethemoreformalmodellingtechniquesasthenumberofPCSsandthePCSsgrewand involvedmorepeople.This indicates that inorder for largercompanies tobesuccessful in managing a setup where there are several PCSs in operation with highcomplexityandnumerousemployees involved(oftengeographicallydispersed),aformalmodellingtechniqueisrequired.ThecompaniesusingtheCPMprocedureasamodellingtechnique claim to be more in control of their product knowledge and their productvariants than thecompaniesusing less formalmodelling techniques.Thismaybepartlydue toan increasedability to involvedomainexperts inthemodellingprocessand thusensuring that the right decisions are beingmade regardingwhich product variants toincludeinthePCSs.Furthermore,amoreformalmodellingtechniquemakesitpossibletokeeptrackoftheproductvariants,featuresandrulesimplementedinthePCS.


84

The impactofusingtheCPMprocedurewasmeasuredintermsofacompany’sabilitytodocument and share product knowledge, the ability to reduce the number of productvariants in thecompanyand the levelofemployeesatisfactionamong those involved inconfiguratorprojects.Therespondentsratedtheimpactonthecategoriesaccordingtothe5pointLikertscalefrom1to5,where1meanstheystronglydisagreewiththestatementand 5means they strongly agreewith the statement. In two cases a non-answerwasprovided,whichwasmarkedwithNA.TheresultsarelistedinTable4-7.

Table 4-7 Comparison of companies using the CPM procedure and less formal or no modellingtechniquesaccordingtotheirabilitytodocumentproductknowledge,controlproductvariants,qualityandemployeesatisfaction.

CompanyID

Impr

oved

docu

men

tatio

nof

know

ledg

e

Impr

oved

avai

labi

lity

of

prod

uct

know

ledg

e

Redu

ctio

nofit

em

num

bers

Incr

ease

dus

eof

stan

dardp

arts

Impr

ovedq

ualit

yofp

rodu

cts

Incr

ease

dem

ploy

ee

satis

fact

ion

CompaniesusingtheCPMprocedure

Average 4.7 4.7 4.0 4.7 4.4 4.6

Companiesusinglessformalmodellingtechnique

Average 4.3 4.5 2.5 4.3 4.2 3.8

Companiesusingnon-formalmodellingtechnique

Average 3.7 3.8 2.2 4.0 3.8 3.3

ThecompaniesnotusingtheCPMprocedureseemstohavelessdocumentationandaccesstotheirproductknowledge.However,thedifferencesbetweenthethreegroupsinrelationtodocumentationandtheaccessibilityofproductknowledgearenotverysignificant.Thiscould be related to the fact that the companies not using the CPM procedure haverelativelysmallconfigurators,wheretheystillcanmanagethecomplexity.

The reduction of product variants (item numbers) refers to the ability to eliminateunnecessaryproductvariantsfromtheproductassortment.Theabilitytokeepdownthenumberofproductvariantsisclaimedtobeanimportantenablerforreducingcomplexityandthuskeepingdowncostsinthecompany(Hvam2008;Lindemann,MaurerandBraun2008).ThecompaniesusingtheCPMprocedureclaimtohaveabetterabilitytoreducethenumber of product variants than the other companies in the research, which may berelatedtoanincreasedabilitytodocumentandgainaccesstoproductknowledge.

Finally,thecompaniesthatusedtheCPMprocedurereportthattheiremployeesaremoresatisfied with their working situation. This may be related to the increased ability todocumentandgetaccesstoproductknowledge,whichmakesiteasierfortheemployeestocontroltheproductknowledge implemented intheconfiguratorsandtocommunicate

85

about the product knowledge with people from product development, sales andproduction,forexample,andfinallydocumentalltheknowledgeusingtheCPMprocedure.

4.5.2.3 Conclusion

Thestudyhasrevealedan importantcorrelationbetween theuseofa formalmodellingtechnique(inthiscasetheCPMprocedure)andthesizeandcomplexityofthePCSsaswellastheabilitytocontroltheproductknowledgeandtheproductsvariants.However,theseresultsraisefurtherquestionssuchaswhichspecificfeaturesofthemodellingtechniqueslead to an increased control of the product knowledge; and what is the correlationbetween the use of a formal modelling technique and the capability to successfullyimplementaPCS.Itisalsohighlyrelevanttocompareinmoredetailtheimpactsofusingdifferent formal modelling techniques. Finally, it would be beneficial to expand thenumberofcompaniesincludedintheresearch.


86

4.6PCSsinintegrationwithotherITsystemsThissectioninvestigatesthepossibilitiesofassessmentsoftheSection3.8thatrefinedtheimportanceof integrationofPCSprojectswithother ITsystems.Therefore, thissectionaddressestheintegrationofPCSsystemswithInternalandexternalITsystemsacrossthesupplychain;whichleadstobetterknowledgemanagementprocessandtasksautomationintheorganization.ThissectionaimstoanswertheRQ5.

4.6.1 PaperJ:AutomaticIdentificationofSimilaritiesacrossProductstoImprovetheConfigurationProcessinETOCompanies.

Authors:SaraShafiee,LarsHvam,KatrinKristjansdottir

Publishedin:InternationalJournalofIndustrialEngineeringandManagement,V.8,No.3.,earlier version originally published in: 17th International ConfigurationWorkshop (pp.139-145),Vienna,Austria.


Ithasbeenobservedthatinmanyfirmsthereuseandgeneralizationofpastexperiences(oftencalled"lessonslearned")isbecomingakeyfactorfortheimprovement,intimeandin quality, of operational processes (Ruet and Geneste, 2002). Further, most ETOcompanies are using PCSs to simplify their specification processes (Hvam et al., 2008)whenthereisnoframeworkavailabletoconnectthePCSwithpreviousknowledge.Basedontheliterature,PCSsfaceaprobleminensuringthemostsatisfactoryconfigurationsforuser requirementswhileobserving thedefinitionof theproduct family (Inakoshi et al.,2001).TheideaistomakeaconnectionbetweenpreviousknowledgeandthePCSwhengeneratingquotations in thePCSandcompare itwith thepreviouslydesignedandsoldprojects from different perspectives. This means that if there is a high percentage ofsimilaritybetweenthecurrentprojectandapreviousproject,thepreviousprojectcanbereused,andthusthecostsandresourcesrequiredtomeettheproductspecificationwillbesignificantly reduced (i.e. costs in the sales, engineering and production phases). Thisresearchthereforeaimstoanswerthefollowingresearchquestion(RQ):

Howcanwe frameandcreateaclusteringdatabase forcomparing thesimilaritiesfrompreviouslydesignedproductsintheconfigurationprocess?


Thispreviousresearchworks inSection3areusedasan inspiration fordeveloping theframework, clustering the data for the comparison purpose, creating the database andintegrating itwith thePCS.There isnodetaileddiscussiononhow tomake adatabasefromtheERPsystemandhowtoclusterandconstrainvariantstoaccesstheolddata.Inthispapertheaimistosuggestaframeworkinordertocreateastandarddatabaseforthiscomparison,wherethecurrentlyavailabletoolsandmethodscanbeutilized.Basedontheliterature, a four-steps framework has been developed, with the following individualsteps:

87

1.IdentifydescribingproductvariablewithinthePCS;2.RetrievespecificationsonpreviouslydesignedproductsfromtheERPsystem;3.Identifyamethodtocompareproductsbasedondescribingvariables;and4.SetupthedatabasewiththepreviousproductsandintegrateitwiththePCS.

First, in order to guide the development and the implementation processes, clearobjectives have to be defined. The next step is concernedwith identifying applicationareastosupportboththesalesandengineeringprocesses.Inthisstep,thelevelsofdetailand completeness that the PCSs has to fulfil are determined.There is often a trade-offbetween the level of detail and completeness, which is the number ofcomponents/modules included in the system. Sales PCSs often include a variety ofproductswithahigh levelofdetail,whereastechnicalPCSshaveagreater levelofdetailbutlowercompletenessasaresultofbeingdesignedtosupportspecificproducts.

Casestudy

TheproposedframeworkwastestedinanETOcompanybyfindingthesuitableclusteringsystem, developing the comparison software and completing the integration of thissoftwareandthePCS.Theteamforthispilotproject includedfourresearchersfromtheuniversityand twoemployeesknowledgeable inERPsystemsmanagement.Figure4-18illustrates theprocessofdelivering thecomparison tool to thecasecompanyover fourmonths.

Figure4-18Structureandinformationflowofthecomparisonsysteminthecasecompany

Inthefirststep,theproductfeaturestobecomparedmustbeselected.DifferentmethodssuchasPVMwereused,andthedesirablefeatureswereselected(Hvametal.,2008).Inthesecondstep,allthevariablesanddatawereretrievedfromtheERPsystemusedatthecompany.Basedon themain customized featuresdetermined in theprevious step,onespecific component with different variables was selected, and with help from the ITdepartment itwaspossibletoretrievethecostdocumentsfromtheERPsystem intoMSExcelspreadsheets.Thefirstobjectivehereistoselectthebestsetofclusteringvariablesleading to an optimized product grouping.Hence, the k-means procedurewas run foreverycombinationof thevariables.Eachonebelonged toadifferentExcelsheet. In thiscase, therewere four sheets for each cluster: x-y, x-z, y-z and x-y-z. In order to assesswhich,onewould leadtotheoptimalclustering,theaverageSilhouetteIndex(SI)forall


88

the analyseswas stored. A higher SImeans accurate clustering.The next stepwas thecalculation of the distance between the new and old products based on the Euclideandistance.Thiswascalculatedforallcombinationsofthevariables,thatis,threevariables(x, y, z) and six possibilities (xyz, xy, xz, yz, x, y, z). The idea is that a small distancebetween the new product and a previous productwould indicate a high similarity. InFigure5,theformulaappearscomplicated,buttheprincipleissimplebasedonEuclideandistancemeasurement.TheITsystemwasdevelopedbasedontheproposedframeworkandwastestedinthecasecompanywiththeavailablePCSandwiththerealproducts.Thesimpleuser interface isshown inFigure4-19aftertheExcelsheet isgeneratedfromthePCS.TheExcelsheetisintegratedwithandreceivesalltheinputsfromthePCS.

Figure4-19Finaluserinterfaceofthecomparisonsystem

4.6.1.3 Conclusion

In thisresearch,weproposea framework forcreatingadatabase for thecomparisonofnewprojectsagainstonespreviouslycompleted,whichallowsthecomparisontobedonein a standardized way while using the available methods and tools. Moreover, theproposed framework is tested in a case ETO company to determine whether theframework ispractical inareal-lifesituation.Thedatabasewascoded inaseparatedMSExcelasthepilotprojectusingtheminimumresourcesatthecasecompany.TheITsystemshowedtheabilitytoclusterandcomparetheknowledgeontheproductsandthusprovedtheconcept.ThesystemdevelopedwasintegratedtoPCSsystemtoretrieve informationfrompreviousprojectsasaninputtotheproposal.

89

4.6.2 PaperK:Improvedperformanceandqualityofconfiguratorsbyreceivingreal-timeinformationfromsuppliers

Authors:KatrinKristjansdottir,SaraShafiee,MartinBonev,LarsHvam,MortenH.Bennic,ChristianS.Andersen

Publishedin:Publishedin:18thConfigurationWorkshop,Toulouse,France.


The recent advancement of cyber-physical systems has enabled a closer integration ofsupply chains relationships (Petnga and Austin, 2016), allowing for efficient ways ofinformation management across multiple organizations. However, to make such an e-businessenvironmentpossible,theestablishedknowledgebaseneedstoaccountforhighdegree of tailoring and dependency from suppliers (Klein, 2007). A technical approachthatenablesreal-timeinformationsharingacrossthesupplychainbyintegratingPCSshasbeenproposed(Ardissonoetal.,2003).However,the impactfromreceiving informationdirectlyfromsuppliersintheconfigurationprocesseshasnotbeenaddressedinpreviousliterature.

This paper aims to capture that research opportunity by analyzing: the overall impactfrom establishing the supplier integration, the gained benefits and themotivations forfutureinvestments,especiallyinETOcompanies,andthechallengescompaniesarefacedwithintheprocess.Alignedwiththefocusoftheresearch,thefollowingpropositionshavebeendeveloped.

Propositions1:ByintegratingPCSsacrossthesupplychain,thecomplexityintermsof business rules, tables, parts and values, of the PCSsmodel can be reduced andconsequentlythemodellinganddevelopmenteffortofthesystemwillbereduced.

Propositions 2: By integrating PCSs across the supply chain, the quality of theproductspecifications intermsof increasedaccuracy–asthe information includedareoptimized,moredetailedandup-to-date.

Propositions3:Themoredetailedspecificationsfromthesuppliermakeitpossibletoimprovetheoveralldesigns,whichleadtodecreasedcostbothforthecomponentinfocusandforrelatedResearchcontribution.


Based on the current literature in the field, the research highlights the importance ofachieving greater integrations in the supply chain where IT plays a significant role.Furthermore,forcompaniesprovidingcustomizedproducts,thereisaneedforhavingup-to-date information across the supply chains.Therefore,by integratingPCSs across thesupplychains,itallowscompaniestofurtherintegratetheflowofinformationandatthesametimesolvesomeofthemainchallengesconcernedwithMCandPCSs.However,theimpact from increased integration across the supply chains by enabling interactions ofPCSsacrossthesupplychainshasnotbeenaddressedintheliterature.Thedataexchange


90

between the case company and the supplier is done via .XML files. The case companysends20designparameters(suchasmin/maxtorque,thereductioninthegearbox,gearfactors),whicharedefinedinthepreviousstepsoftheconfigurationprocess.Therequestistofindadesignwithintheseparameters,wherethesupplier’sPCS,basedontheirlogicandbusiness rules, findallpossibledesignsolutions,whichcanbearound100 and theprices for thedesigns. It ishighlyunlikely that thesupplierPCSwillnotbeable to findfeasiblesolution.However, if thatsituationdoescomeupeitherparametershave tobechangedintheconfigurationatthecasecompanyorthesupplierhastobecontacted.Thedesignsolutionsaresortedaccordingtoprices(fromlowesttohighest)andsentbackonan.XMLformatviathewebAPIwebservices.Forthisspecificproduct,thepricesaremostimportant and therefore the cheapest solution is automatically selected by the casecompany’sPCS.

Casestudy

ThetechnicalsetupinvolvesallowingthePCSsatbothcompaniestointeract(business-to-business communication) in order to retrieve real-time and accurate PCS from thesupplier,whichcanbeused in theoverallconfigurationat thecompany. InFigure4-20,thesetupofthesupplierintegrationisdemonstrated.

Figure4-20Thetechnicalsetupatthecasecompany:thesupplierintegrationviaAPIwebservices

Thiscanbetracedtoimprovetheaccuracyofthesuppliers’data,optimallevelofdetails,andup-to-dateinformationwhichpositivelyaffectsthecostoftheoveralldesign.Withthissetup the supplier does not have to revile the actual logic behind the designs and thepricingstrategyasthesupplier’sconfiguratoristreatedasablackboxintheconfigurationprocess.Furthermore,thecomplexityoftheconfigurationmodelscanbereducedandthetimeconsumingtaskofmodellingandmaintenancearedelegatedtothesupplier.

4.6.2.3 Conclusion

Thepresentpaper analyses the impact fromhaving integratedPCS in the supply chainnetworkinanETOcompany.TheapproachsuggeststheinvolvementofPCSthatretrieveaccurate sub-product information in real-time from suppliers during the customizationprocess.Threepropositionsweredeveloped toanalyse the impact from integratingPCSacross the supply chains to retrieve more accurate and detailed information in theconfigurationprocesses

91

Themodellinganddevelopmenteffortprovedtobereducedatthecasecompanyastheyare not responsible for modelling the supplier’s product information. Thereby themodelling and maintenance effort is moved to the supplier. The findings support thispropositionsasthecomplexity,whichisdefinedinnumbersofbusinessrules,tables,partsandvaluesisreducedtoalmostzero.Thisalsoeffectsthedevelopmenttimeofthesystemwhichisreducedfrom8+daysto2andthespecialisttimespentonthedevelopmenthasbeen reduced from 8+ to zero.The quality of the configuratorsmodel in this article isdefined in terms of improved accuracy as the information retrieved via the supplierintegrationareoptimized,moredetailedandup-to-date.Thefindingssupportthisasoverdimensioning of different parts is not required as a result to improve quality of thespecifications.Theresultindicatethatthecompanycansaveupto20%ofmaterialcostasa result to immediate and in-direct savings gained from offer sizing of surroundingsystems.The resultsbasedon thisstudy indicate thatsignificantbenefitscanbegainedfrom increased supply chains integrations inETO companieswhere integratedPCS aredistributedacrossthesupplychain.


92

4.7ChaptersummaryThis chapter presents the results from the PhD project taken from different publishedarticles based on different RQs and providing the validation of the contribution. Thesuggested methods and frameworks are developed based on the literature andexperiencesgainedfromworkingwithPCSprojects.Theproposedsolutionsaretestedinseveral case studies which are the evidence of the research project evaluation. Asdiscussed,researchisdividedintofivemainRQs.

Tobeginwith,chapter4.2 is theanswer toRQ.1 and the literature reviewandsurveyshelpedindeterminingthebenefitsandchallengesreportedfromacademiaandindustryinorder to clarify the RQs. In the first step, the framework formaking strategies beforestarting theplanning issuggested.Business casesare reported asoneof the importantphases of planning PCSprojects and a framework for developingBCs is suggested andvalidatedbytestinginmultiplecasecompanies.

In chapter 4.3 where RQ.2 is answered, the scoping criteria and framework for PCSprojectsincludingalltheneededfuturestepsareproposed.

InRQ.3andchapter4.4,thechallengeofKMishighlightedandtheframeworkisproposed.Theframeworkistestedandvalidatedinmultiplecasecompanieswhileusingtheoutputsfromthepreviousphasesinplanningandscopingstages.

In RQ.4 and chapter 4.5, a framework, an automatic IT solution for modelling anddocumentingPCSprojects isproposed. Inanotherpublication, asurveystudyevaluatestheimportanceofusingmodellingtechniquesinindustries.

ThelastRQregardingdifferentintegrationsalignedwithPCSisansweredbyproposingITsolutions fordoing the internal and external integrations (chapter4.6).The firstpaperproposes a framework for clustering the similarities of previous and the new orderedprojects throughoutdevelopinganExcelsheetand integrate itwithPCS.The lastpaperdiscussesanITsolutiontointegratethePCSinsidethecompanywiththevenders’PCSsasanexternalITsystem.

93


94

5CONCLUSIONS

5.1ContributiontotheoryThis section gathers the formulated RQs and the results of the research. Each RQ isansweredbydifferentpapersandevaluated indifferentcasecompanies. In thissection,the answers to themainRQs are summarized including their sub questions developedthroughoutthethesis.

Theoverallresearchobjectiveofthisthesisis:

ResearchObjective:

To frame and improve the management process of PCS projects: planning,development,andmaintenancephases

Inordertoprovideananswertotheobjective,theresearchobjective isdivided intofiveRQs,whichareansweredbelow.TheprocessofPCSprojectswithinanorganizationcanbeperformedtakingintoaccountthefollowingareas:

- Structuredbusinesscasesbeforeplanningtheproject,motivations,benefitsandchallengesinPCSmanagement,

- ScopingPCSandhavingastructuredplanningphase,- ManagingknowledgeinPCS,- Modelling,documentationandmaintenanceofthePCSprojects,- PCSintegrationwithotherITsystems

Figure 5-1 illustrates different phases of PCS projects and PhD thesis contributionfollowingFigure2-1.

95

Figure5-1IndividualphasesofPCSprojectsandthesiscontribution

EachofthemainRQs(1,2,3,4and5)focusesonmentionedfivedifferentaspectstobeconsideredinPCSprojects.TheRQsandtheanswersarepresentedbelow.

RQ.1.HowcanPCSprojectsbeplanned?

RQ.1referstoastandardmanagementprocesstomotivatetheorganizationsforinvestinginPCSprojectsbyconsidering thebenefitsandchallenges.The answer to thisquestionwillbe aBC framework forquantifyingbenefits, evaluating challenges, and consideringcritical parts and risks in PCS projects. This helps organizations to be aware of thechallengesintheprocesswhilebenefitingfromPCSprojectsinordertoprioritizeprojects.

RQ.1.1.HowcanwemakeastrategytodemonstratedifferentapplicationofPCSprojects?

TheRQ.1canbebrokendownintotwosubquestions.ToanswertheRQ.1.1,areviewonliterature is conducted related to different applications of PCS projects. Moreover, aframework issuggested to facilitate thestrategicaldecisionsand it is tested in thecasecompany (chapter 4.2). The main contribution of this answer can be summarized asfollows:

1.FrameworkdevelopmentforstrategicaldecisionstodemonstratedifferentapplicationsandprioritizingPCSprojects(PaperA):AframeworkisproposedwiththeaimtoprovideastructuredapproachtodevelopastrategytoguidethedevelopmentandimplementationprocessesofPCSsinETOcompanies


96

2.Empirical case study validation (PaperA):The proposed strategy framework of PCSprojectswas tested in oneETO company.The results from testing the framework andobservationsof thecasestudy facilitate thedecisionmakingprocessdue to theanalysismadeinthisstep.ThecompanywasmanagedtosolvetheconfusionchallengeofselectingandprioritizingdifferentprojectsandstrategicallyusedtheoutputsofpreviousprojectstosavetimeandresourcesintheITarchitectureoftheirfuturePCSprojects.

RQ.1.2. How can the benefits and challenges from different PCS projects beevaluated?

The answer to RQ.1.2 includes a review of key publications and the investigation ofrelevant subjects from literature related to benefits and challenges in conducting PCSprojectsmainlyinETOcompanies(Chapter3.4).Moreover,theestablishedunderstandingwastestedthroughouttheempiricalinvestigation(Chapter4.2).Themaincontributionofthisanswercanbesummarizedasfollows:

1.Indepthlong-termquantificationofcostsavingsfromapplyingPCSs(PaperB):Inthispaper, we calculate the costs from PCSs and compare it to the cost of development,implementation andmaintenance of the system.This study is conducted in a five-yearperiodandinasinglecasecompany.

2.Surveyanalysisbasedon theapplicationofPCS frommotivations torealizedbenefits(Paper C): The motivations as a response to organizational needs are addressed andanalysedbasedonasurvey.Thequestionsinthesurveyconcernedwhatmotivatedthesecompanies to invest on PCS projects. The result from the companies indicates howsuccessfulthesecompaniesaretoachievethebenefitsfromapplyingPCS.

3.Surveyanalysisofchallenges from implementingandusingPCSbasedonsurveyandinterviews(PaperD):ThisthesisestablishedadetailedunderstandingofPCSchallengesinorganizations as a result ofusingPCSbasedon literature, interviews and surveys.Theopen questions were used to identify additional categories of challenges that are notdescribedintheliteratureandtoidentifythespecificchallengesbelongedtoeachofthosecategories.Then,theclosedquestionswereaskedtoratetheimportanceofsixcategoriesofchallengesidentifiedintheliteratureonafive-pointscale.Theanswersfromcompanieshelped to categorize the challengeswhich help the companies to be aware of possibleinvestments inthefuturetoavoidthechallenges.Thesechallengescouldbe listedasthepossiblerisksintheBCs.

RQ.1.3.HowcanastandardbusinesscaseforPCSprojectsbeframed?

The answer to RQ.1.3 includes a review of key publications and the investigation ofrelevant subjects from literature related to available BCs for IT projects and availablesteps forBCs inPCS (Chapter3.4).Moreover, the establishedunderstandingwas testedthroughout the empirical investigation (Chapter 4.2). The main contribution of thisanswercanbesummarizedasfollows:

97

1.FrameworkdevelopmentforBCsinPCSprojects(PaperE):Basedonaliteraturereviewofkeysubjects,thisthesisestablishedadetailedunderstandingofgeneralapplicationofBCsinPCSprojects(Chapter3.4).InspiredbytheavailableframeworksforITprojects,theavailablestepsandtoolsinPCSprojects’literaturearelisted(Chapter3.1).Combiningtheliteraturewithexperiences leads tothe frameworkdevelopment forBCsneeded forPCSprojects. The framework for making BCs for PCS projects was tested in multiple casestudies(Chapter4.2).

2.Empiricalcasestudiesvalidation (PaperE):Theproposed framework forBCsofPCSprojectswastestedintwoengineeringcompanies.TheresultsfromtestingtheframeworkandobservationsofthecasestudiesshowtheinterestamongthewholeteaminusingthesuggestedBCframeworkaswellasitschallengesatthecompany.


RQ.2referstoastandardframeworktoscopePCSprojectsandmakeasuccessoutoftheinvestment.Thisframeworkmakestheorganizationsawareofthedecisionstheyhavetomake on the way and standardize the planning, development, implementation anddocumentation process based on RUP methods (Chapter 3.5). This question aims atreducing the needed time, resources and confusions during PCS projects. The maincontributionofthisanswercanbesummarizedasfollows:

1.Frameworkdevelopment forscoping thePCSprojects(PaperF): Inorder tocombinethe literaturereviewofavailabletoolswithexperience,thisthesisestablishedadetailedframeworkandsequencesofstepsforscopingPCSprojects.Thesestepsevaluatetheaimsandbenefitsforthespecificproject,introducetoolsforstakeholders’analysisaswellasITarchitecture and define the level of details. Moreover, it introduces the managementplanningfromRUPtohelpindeterminingdifferentphasesoftheprojects.

2.Empiricallyvalidatedsignificance(PaperF):Theframeworkwasvalidatedinonecasecompany and the detailed example of the tested project was provided to show theapplicationoftheframework.Thetestingresultsshowedtheimpactoftheframeworkondifferentphasesoftheprojectwhilereducingtheneededtimeandresource.

RQ.3.HowcanweacquireandmanageknowledgeneededforPCSprojects?

The answer to RQ.3 requires a deep review of literature on the differences betweengeneralITandPCSprojects(Chapter3.1).TheliteraturereviewinvestigatestheavailableliteratureonITprojectsandPCSprojectsregardingKM.TheinvestigationsdescribedKMframeworksonITprojectsandtheneedofKMframeworksforPCSprojects(Chapter3.6).The research aims at reducing the efforts and cost for PCS projects by structuring theknowledge management process. The main contribution of this answer can besummarizedasfollows:

1.FrameworkdevelopmentforknowledgemanagementinPCSprojects(PaperG):Alignedwith a detailed literature review, this thesis established a structured framework and


98

sequencesofsteps forKM inPCSprojects.ThispaperdemonstratesdifferentphasesofPCS projects and how all these phases need documentation and management of theknowledge.Theframeworkmentionsthescopingneededtoplantheprojectinahighlevelofabstractionbeforeinitiation.ThescopingframeworkusedinthisphasewastheresultsfromRQ.2astheoutputofPaperF.AfterstartingthePCSprojects,thetoolsandmethodare introduced for knowledge acquisition, modelling and documentation phase. Inknowledge acquisition phase, the importance of receiving right level of details for theknowledge in next steps is discussed. Modelling phase clarifies the impact of a clearcommunication regarding knowledge with the stakeholders; and the importance ofknowledge validation on the quality of PCS projects. In the last phase, the knowledgemaintenanceanddocumentationforupdatesandfurtherdevelopmentsinPCSprojectaredescribed.

2.Multiplecasestudiesempiricalevidences(PaperG):The frameworkwasvalidated intwocasecompaniesonfourdifferentPCSprojects.Theresearchpresentstheanalysesoftheapplicationof the frameworkand thecrosscasecomparison indifferentcompanies’settings. The observations indicate that KM framework leads to improvements in themanagementprocessofPCSwhilereducingtheneededtimeandeffortforgatheringandmodellingknowledge.


Theanswer toRQ.4 requires adeep reviewof literatureandunderstandingofdifferentmodelling techniques, communications tool, and documentation systems (Chapter 3.7).TheliteraturereviewillustratestheneedforformalmodellinganddocumentationtoolsinordertosavetimeandresourcesintheprocessofPCSprojects.Thisquestionincludestwosubquestions,whichwillbesummarizedbelow.TheanswertoRQ4.1outlinestheresultsof the research on an agile framework for documenting PCS projects and IT tooldevelopment (Chapter 3.7).The second area of the research focuses on survey resultsabout the applicationof formalmodelling techniques and theirbenefits inPCSprojectssettings.

RQ.4.1.Howcanweautomatethedocumentation,validationandcommunicationprocessinPCSprojects?

InRQ.4.1, the researchaimsat introducing amoreefficient framework fordocumentingPCSprojects,whileitissupportedwithanITtooltoprovetheapplicationoftheproposedframework.Theoutputsarelistedasfollows:

1.Frameworkdevelopment formodelling,documenting and communicatingduringPCsprojects (Paper H): aligned with a detailed literature review, this thesis established astructuredanagileframeworkfordevelopinganautomateddocumentationandmodellingsolutioninPCSprojects.

99

2.Developingan ITTool(PaperH):Using the frameworkas theautomaticsolutionandlistinguptherequirementsfordocumentingITandPCSprojectsfrom literature leadstoan ITsolution.The ITsolutionproved theapplicationof the frameworkwhich indicatedthesatisfactionatthecompanybasedonthereviews.

3. Multiple case studies empirical evidence (Paper H): The framework is validated bydevelopingtheITsystemandtheITsystemisvalidatedbytestingdifferentPCSprojects.Thefeedbacksformthestakeholdersaregatheredbyinterviewsacrossthecompany.Theusabilityofthesystemisprovedwhichdemonstratesthesavingintimeandresourcesinthe case company aligned with a high quality in communication and validation ofknowledge.Thedocumentationisdoneinminimumamountoftimeusingthisautomateddocumentation system. Meanwhile, the utilization is efficient and easy for thestakeholders.

RQ.4.2.HowimportantisittousethestructuredmodellinganddocumentationtechniquesinPCSprojects?

TheRQ.4.2demonstrates the importanceofdocumenting andmodelling theknowledgefromPCSprojects in a structuredway.The literature emphasizeson the importanceofusing formal modelling techniques for communication, validation and maintenancepurposes (Chapter3.7).Thispaper aims to show the realdata from several companiesillustrating thebenefitsofusingmodelling techniquesonPCSprojectsdevelopmentandperformance(Chapter4.5).Themaincontributionof thisanswercanbesummarizedasfollow:

1. Survey and interview analysis based on the benefits of using modelling anddocumentationtechniques:theresearchprovedthecorrelationbetweenthehighdegreeofusingvisualmodellingtechniqueandhighqualityofdevelopmentandcommunication.Theresultsarebasedonanswers from industrialmanufacturingcompanies.Theresultsindicatethatcompaniesusingtheformalmodellingprocedurestendtohaveanimprovedabilitytoreducethenumberofproductvariantsinthecompany,improveddocumentationoftheirproductknowledgeandhigheremployeesatisfaction.

RQ.5.HowcantheintegrationswithotherITsystemsleadtoimprovementsintheperformanceofPCSs?

RQ.5 verifies that PCS performance increase when it integrateswith other IT systems(Chapter3.8)andhighlightstheimportanceofachievinggreaterintegrationsinthesupplychainwhere ITplaysasignificantrole.The literaturestudiesshow theefficiencyofPCSsystemswhen theyarealignedwithother ITsystemssuchasERPsystems,simulationsandcalculationsystems,CADsystemsandevenexternalITsystemssuchasvenders’PCssystems.Themaincontributionofthisanswercanbesummarizedasfollows:

1. Framework development for analysing the similarities between configured products(Paper J):Literature introduces a framework for comparing theprevious soldproductsand thenewconfiguredorders throughPCS.Thisresearchsuggestsa frameworkwhere


100

product features from the company’s PCS are clustered in order to compare withpreviouslymade products by retrieving information from the proposed database. Thisframeworkandtheproposeddatabasearetestedinonecasecompany.

2. IT tool development and case study validation (Paper J): The IT tool developed tosupport thesuggested frameworkandaimedatclusteringandassessing thesimilaritiesfor a specific product at the case company. The IT tool is a proof of concept anddemonstrates the possibility for these kinds of automations in industries based onconfiguredproducts fromPCSprojects.The case companysatisfaction from this IT toolverifiesthepotentialofanalysingthesimilaritiesamongprojectsforotherproductandinanautomatedway.

3. In-depthanalysisof integrating thePCSacross thesupplychain to receive real timeinformation from suppliers (Paper K): The result presented in this study are reportedfromanETOcasecompanyused theproposedframeworkandsolution foraperiodandmeasuredthebenefitswhichcanbeachievedbyintegratingPCSsacrossthesupplychains.Thebenefitsreportedfromthecasestudyarequantifyingbasedonthesavedmanhoursfor receiving the knowledge from different venders and qualifying based on the shortresponsetimeanderrorfreedocumentssenttothecustomers.

4.ITtooldevelopmentandcasestudyvalidation(PaperK):TheIttoolisdevelopedbasedon the frameworkbyusing .XML formatvia thewebAPIwebservices.Theresults fromusing this automated solution shows a lot of benefits at the case company listed as:complexityreduction,qualityofthespecificationprocess,andcostsavings.

101

5.2ContributiontoPracticeDrawingontheanswerstotheRQs,thissectiondiscussesthepracticalsignificanceofthisPhDstudyinTable5-1.

This PhD study focuses on the case company Haldor Topsoe A/S for stablishing,developing,andmaintainingPCSprojects.TheobservedchallengesinpracticeisbasedonrealexperiencesworkingwithPCSprojectsinrealorganizationalsettings.Combiningtherealexperiencesandchallengeswithacademicresultsleadstocontributiontotheoryandpractice.

Table5-1PhDstudy’scontributiontopracticeandtheoreticalchallenges

PracticalChallenges TheoreticalChallenges ContributionofthePhD

RQ1. How toidentify thepossiblelong-termand short-termbenefitsfrom PCSproject?Howto have awiseselectionandprioritizationof PCSprojects?How tomakea businesscase for PCsprojects?

There are research works ongeneral benefits of PCSprojects. The researchers tryto showbenefitsor introducetools andmethods on how toselect and prioritize the PCSprojects in industries to getthemostbenefitsoutof them(Chapter1.2).However, thereis no framework to developBCsandguideidentificationofthe future projects which arecost and time beneficial(Chapter3.4).

There is a need for surveys,interviews and in depthstudies in order to prove thesuccess and benefits of PCSproject.

There is further need for aframework and template forBCsinPcsprojects.

First of all, by literature reviews andexperiences from different project, aframework for prioritizing andmakingstrategic planning for PCS projects isdeveloped.Theresultswereastrategicplan reported to the case company forthefutureprojects(Chapter4.2).

Secondly, illustrating the results andsuccessratesinPCSprojectsfromotherindustries is one of the mainmotivations to invest and use PCSprojects. The awareness of thechallengesforallthecompaniesleadstobetter planning and risk assessment.Furthermore, the challenges reportedfrom surveys warn all the industriesabout the bottlenecks on the way(Chapter4.2).

Byreviewingalltheavailable literaturefor IT projects in general and PCSprojects specifically (Chapter 3.1), wefound trends for framework and weadded the available tools to theframeworks. The results lead to a BCsframeworkwhichmakes the selection,planning and prioritization efficient atthecasecompanies(Chapter4.2).

RQ2. How toscope andmanagePCS?

Thereisavailableliteratureondifferent steps of managingand planningdifferent phasesof PCS projects (Chapter 3.5).There is no framework toguidethepractitionersstepby

By reviewing the available steps forplanning, implementing, developing,andmaintainingPCSprojects,thisstudysuggests a framework for scoping PCSprojectfromplanningtodocumentationphase(Chapter4.3).


102


step through different stagesof planning, implementation,development, andmaintenance.

A framework is required tocontain available tools andmethodstointroduceasimpleefficient process for planning,implementation, developing,andmaintainingPCSprojects

In this framework, the stakeholders’requirements are analysed in the firststepwhichisreportedasachallengeinPCS projects. We aim at analysing thecurrentprocessof tasksandproposingthe futureprocess,benefiting from theavailable tools. Determining the ITstructure and number of needed ITdevelopments is one of the confusionsfor all practitioners. This frameworkproposed the methods to prioritizedifferent components in one specificPCS project. The study aims to clarifyhow to plan the project iteratively aswellashowtomanagethemaintenanceanddocumentationofthesysteminthefuture.

Most of the inputs for this phase isbased on the outputs from BCs in thepreviousRQ.

RQ.3.How tohave anefficientframeworkfor acquiringthecomplicatedknowledgefor PCSproject?

Previous studies addresshowcomplicated is to gather andgroup the needed knowledgeforPCSprojects.Somestudiessuggest tools andmethods tosimplifytheKMprocessinPCSprojects (Chapter 3.6). Otherstudies just report thechallenges regarding KM andcomplexitiesaboutknowledgeforPCSprojects.

There is no framework toscope and standardize thesteps in KM process in asequential manner. Having astandard framework forgathering and prioritizingknowledge inPCSprojectwilllead to saved time andresources both from businessandPCSteam.

ThisstudyclarifiestheconceptofKMinIT project and the available proposedframeworks for IT projects (Chapter3.1,3.6).

The conclusion of literature study is aframework inspired byKM frameworkfor IT projects and available steps forPCS. The framework facilitates themanaging of the knowledge from theplanning to the maintenance phase ofPCSprojects(Chapter4.4).

The application of the framework indifferent companies and projectsdemonstratesthesatisfactionratefrompractitioners due to an efficient KMprocedureinPCSproject.

TheScopingofthePCSprojectsandtheoutputsfrompreviousRQisneededforKMprocessinthisstage.

RQ.4. Howeto model theproduct,documentand validate

Further developments andmaintenance of productknowledge explains theattention of the literature tothe modelling and

Thestudysuggestsanagile frameworkfor automatic documentation andmodelling of PCS projects. Theframework isdesigned inawaytoskipany over-writing and duplication of

103


theknowledgeinPCSprojects?

communication techniques(Chapter3.7).ThePCSprojectshould be updated based onthe products changes at thecompany. The productknowledge has to becommunicated and validatedall the time. Therefore, thecompanies need an efficientway of modelling anddocumentingtheknowledge.

Thereisaneedforresearchonhow to do documentation inthe most efficient way butthereisaneedforanITtooltoautomate the wholedocumentation processwithout aneed for redundantworkandresources.

knowledge in order to have an agileknowledge documentation (Chapter4.5).

The framework results in an ITdocumentationsystemwhichautomatethe whole process of modelling,communication, validation anddocumentationoftheknowledge insidePCS.

Theresults from theusabilityof the ITdocumentation system demonstratesthe satisfaction and improvements aswellascostsavingsinthedevelopment,testingandmaintenanceprocess.

The documentation and maintenanceplus communication and validation ofknowledgearetwoimportantphasesoftheRQ.3.

RQ.5 How toget the mostbenefits outof PCSprojects byintegratingthem toother ITsystems?

The literature refers to theefficiencyofPCsprojectwhileintegrating with other ITsystem such as ERP or CADsystems(Chapter3.8).

There are researchersfocusing on IT aspects ofdevelopment while efficientmanagement of connectingthese IT systems isunderestimate.

This study presentsframeworks in order tosimplify the integration of ITsystem with PCS projects,without any focus on ITtechnicalchallenges.

The study reviews the availableliterature of possible integrationswithIT systems inorder to retrieve theoldknowledgeofproducts.Thiswillleadtoaframeworkforclusteringtheproductsto assess the similarities of the oldproducts and the configured products(Chapter4.6).

An IT tool isdeveloped to support theframeworkandtheresultsfromusingitin the case company show savedman-hoursbecauseoftheautomations.

The relations in supply chain andvenders’managementisoneofthemainchallenges in order to receive the realtime response and costs for thecustomers. The research took thisopportunity to propose a structuredway align with the IT solutionsuggestions, which will help thecompany to integrate their own PCSwiththeonesfromcustomers.Thiswillhelp the company even to comparedifferent suppliers in terms of quality,price and etc. and select the mostbeneficial vender in terms of theircustomers’preference(Chapter4.6).


104

5.3ResearchlimitationsTheunderlyingmethodological implicationsof thisresearchshouldbeacknowledgedaspresentedbelow:

Generalizability of the research: This research was undertaken as an applied researchproject within the Operations Management research group with the collaboration ofHaldor Topsoe A/S. Different factors limited the generalizability of the studies in thisthesis including:stakeholders’ requirements indifferent industrialsettings,productandprojects types, and finally industry types used as case studies in the research work.Otherwise, Haldor Topsoe A/S and other case companies used in this study are wellknown as ETO companies producing highly engineered complex products and processplants.ThecasestudiestypesallowtheresearchgrouptofacethemostcomplicatedtypesofPCSprojectsandrepeat the in-depth long-termtestingofthedevelopedmethodsandtools.Meanwhile,thestudyofonecasecompanygivestheteamtheopportunitytohavehands inpracticedirectlyand ITdevelopments toassess the research in realwork in along-termperiod.Furthermore,moreETO case companiesmake the resultsgeneral forETOcompanies.

5.4OpportunitiesforfuturestudiesThis PhD project resulted in several frameworks and methods which could ariseopportunitiesforthefutureresearch.

1. Referring to the all RQs, in order to motivate and assess the PCS benefits from 1)planning, 2) scoping, 3) KM process, 4) documentation and modelling, 5) effectiveintegrationwithotherITsystems,severalframeworksandmethodicalinstruction(theorycontribution)areintroduced.ThesesuggestedframeworksandtoolsaretestedinlimitednumbersofETOcasecompanies.ETOcompaniesrepresentcomplicatedhighlyengineeredproducts and serviceswhich could cover other less complex products. Further testingcouldgeneralizethesuggestedapproachesor improvethemtobestable inall industrialsettingsandforallkindsofPCSprojects.Thisresearchisanin-depthcasestudyreportinorder toevaluate thechallenges in industriesandcontribute to the theoryandpractice.The further research could focus to implement and test all reported challenges andsolutionsinawiderangeofstakeholdersandtypesofindustries.

2.The secondopportunity for future studies is the stepsof the frameworkwhichhavereceivedlessattentionfromacademiaandreportedasabigchallengeinpractice.Oneofthese challenges is the organization influence on user acceptance or rejection of PCSprojects.TheorganizationalchallengessuchaschangemanagementisreportedasoneofthemainrisksofthePCSprojectsbasedonliteratureandthisPhDstudy.

3.Theotherwell-knownaspectoftheorganizationalchallenges istheriskmanagement.AssessingtherisksofPCSprojectsandtheirinfluence inPCSprojectsfailureandfindingthepropersolutionscouldbebeneficialbothforthe industryandacademia.ElaboratingonriskmanagementinPCSprojectscouldbebeneficialbothinBCandplanningphaseandinimplementationphase.

105

4.ThethirdresearchopportunityistheintegrationbetweenPCSandotherITsystemsinindustries such as simulation tools, CAD systems or ERP systems. Benefits could beachieved ifPCScouldcommunicatewithother ITsystems.Morespecifically, IntegrationwiththeERPsystemtoreceivecustomerinformation,create2Dand3DfiguresintheCADsystembasedon the configuredknowledge,or simulate complicated calculations in thesimulationsandcalculationsystems.Somequestionsregardingintegrationsareansweredin this PhD thesis and the available literature,while there are stillmany unansweredquestionsaboutthepossibilitiesandsolutionsforpotential integrations.TherearegreatinterestsandpotentialsinindustriesfornewframeworksandITsolutionforintegrationpurposes.

5.OneoftheimportantphasesofPCSprojects,whicharenotdiscussedinthisthesisduetothetimerestrictions, isthe implementationphase.Thisphase isvery interestingbothforacademiaand industriesandhavenotyetattracteda lotofattentionfromacademia.Thisphasecontainsthemanagerialproficiencytolaunchthesystem,testandvalidatethesystemandenforcetheuserstoreplaceitwiththeircurrentwaysofworking.


106

5.5ConcludingremarksIn Conclusion, the frameworks,methods and IT solutions tested in the case companiesindicate the efficacy in planning, developing, documenting andmaintenance of the PCSprojects.Ingeneral,thetimeandresourcesneededforeachofthephasesarereducedduetotheproposedsolutions.Thesesolutions includingtheframeworksandITautomationsleadtocostsavingsaswellasqualitativebenefits.Therefore,inordertoanswerthemainobjective of the thesis from both the theoretical and practical perspectives the maincontributioncanbelistedas:

1. Literature reviews and survey investigations: The literature knowledge helps theresearchertofindthegapsandchallengesintheindustry.Thesurveysconductedduringthethesisimprovetheunderstandingoftheindustriesopiniononbenefits,challengesandtheimpactoftheformalmodellingtechniquesontheirdailywork.Thesurveysillustratethe influences of academic results as well as gaps in the literature and also providesindustrieswithknowledge fromother industries.Both literature and surveyswere thefoundation forthePhDprojectwhile the importanceof theexperiencesofworkingwithPCSprojectsinpracticecannotbeunderestimated.

2.Frameworksdevelopment:Developingdifferent frameworks is the resultof reportedchallengesindifferentphasesofPCSprojects.Theseframeworksaimatstandardizingandformulizing different phases of PCS projects in order to have an efficient projectmanagementandsavedman-hours.ThedevelopedframeworksincludetoolsandmethodsalignedwithITsolutions.

3.ITtoolssupport:TheITtoolsaredevelopedtosupporttheframeworksandprovetheapplicationoftheframework.ThevalueoftheITtoolfor industriesistheautomationoftheworkingprocesseswhileprovidinganeasyefficientsolution.TheseITtoolsaretestedincasestudies toassess thesatisfaction rateamonguserscompared to theold routinesolutions.

107

6REFERENCES

Aldanondo, M., Rouge, S. and Ve´ ron, M. (2000), “Expert configurator for concurrentengineering :Caméléononsoftwareandmodel”,JournalofIntelligentManufacturing,Vol.11No.2,pp.127–134.

Ambler, S. (2002),AgileModeling,Guide to theUnifiedProcessFeaturingUML, JavaandDesign Patterns, Guide to the Unified Process featuring UML, Java and DesignPatterns,availableat:https://doi.org/10.1007/1-85233-856-3_25.

Andreasen, M.M. (1980), “Methods for Synthesis, a System’s Approach”, Diss. LundsUniversityofTechnology.

Andreasen, M.M. (1994), “Modelling—The Language of the Designer”, Journal ofEngineeringDesign,Vol.5No.2,pp.103–115.

Ardissono,L.,Felfernig,A.,Friedrich,G.,Goy,A.,Jannach,D.,Petrone,G.,Schafer,R.,etal.(2003),“AFrameworkfortheDevelopmentofPersonalized,DistributedWeb-BasedConfigurationSystems”,AIMagazine,Vol.24No.3,p.93.

Ariano, M. and Dagnino, A. (1996), “An intelligent order entry and dynamic bill ofmaterials system formanufacturing customized furniture”, Computers & ElectricalEngineering,Vol.22No.1,pp.45–60.

Ashurst,C.,Doherty,N.andPeppard,J.(2008),“Improvingthe impactofITdevelopmentprojects:thebenefitsrealizationcapabilitymodel.”,EuropeanJournalofInformationSystem,Vol.17No.4,pp.352–370.

Barker, V.E., O’Connor, D.E., Bachant, J. and Soloway, E. (1989), “Expert systems forconfigurationatDigital:XCONandbeyond”,CommunicationsoftheACM,Vol.32No.3,pp.298–318.

Basili, V.R. and Weiss, D.M. (1984), “A Methodology for Collecting Valid SoftwareEngineeringData”,IEEETransactionsonSoftwareEngineering,No.6,pp.728–738.


108

Bittner,K.(2002),UseCaseModeling,Addison-WesleyLongmanPublishingCo.,Inc.

Blecker, T., Abdelkafi, N., Kreutler, G. and Friedrich, G. (2004), “Product configurationsystems:stateoftheart,conceptualizationandextensions”,ProceedingsoftheEightMaghrebianConferenceonSoftwareEngineering(MCSEAI2004),pp.25–36.

Blessing,L.T.M.andChakrabarti,A.(2009),DRM,aDesignResearchMethodology,Springer,availableat:https://doi.org/10.1007/978-1-84882-587-1.

Booch,G.,Rumbaugh,J.andJacobson,I.(1999),TheUnifiedModelingLanguageUserGuide,Reading,UK:AddisonWesley,Addison-Wesley.

Briand,L.C. (2003), “Softwaredocumentation:howmuch isenough?”,SeventhEuropeanConferenceonSoftwareMaintenanceandReengineering,2003.Proceedings.,pp.3–5.

Chang, C.L. and Lin, T.-C. (2015), “The role of organizational culture in the knowledgemanagementprocess”,JournalofKnowledgeManagement,Vol.19No.3,pp.433–455.

Chao,P.Y.andChen,T.Te.(2001),“Analysisofassemblythroughproductconfiguration”,ComputersinIndustry,Vol.44No.2,pp.189–203.

Compton, P. and Jansen, R. (1990), “A philosophical basis for knowledge acquisition”,KnowledgeAcquisition,Vol.2No.3,pp.241–258.

Coram, M. and Bohner, S. (2005), “The Impact of Agile Methods on Software ProjectManagement”,12thIEEEInternationalConferenceandWorkshopsontheEngineeringofComputer-BasedSystems(ECBS’05),pp.363–370.

Creswell,J.andClark,V.L..(2007),DesigningandConductingMixedMethodsResearch,2nded.,SagePublications,Inc.,ThousandOaks,CA.

Dahmus,J.B.,Gonzalez-Zugasti,J.P.andOtto,K.N.(2001),“Modularproductarchitecture”,DesignStudies,Vol.22No.5,pp.409–424.

Darke, P., Shanks, G. and Broadbent, M. (1998), “Successfully completing case studyresearch:combiningrigour,relevanceandpragmatism”,InformationSystemsJournal,Vol.8No.4,pp.273–289.

Duffy, A. and Andreasen, M. (1995), “Enhancing the evolution of design science”,ProceedingsofICED95,Praha.

Dvir, D., Raz, T. and Shenhar, A.J. (2003), “An empirical analysis of the relationshipbetween project planning and project success”, International Journal of ProjectManagement,Vol.21No.2,pp.89–95.

Eisenhardt, K. (1989), “Building theories from case study”, Academy of ManagementReview,Vol.14No.4,pp.532–550.

ElMaraghy, H.A. (2007), “Reconfigurable process plans for responsive manufacturingsystems”,DigitalEnterpriseTechnology,Springer,pp.35–44.

ElMaraghy, H.A. (2009), “Changing and evolving products and systems–models andenablers”,ChangeableandReconfigurableManufacturingSystems,Springer,pp.25–45.

109

ElMaraghy,H.,Schuh,G.,ElMaraghy,W.,Piller,F.,Schönsleben,P.,Tseng,M.andBernard,a. (2013), “Productvarietymanagement”,CIRPAnnals -ManufacturingTechnology,Vol.62No.2,pp.629–652.

Erens, F. and Verhulst, K. (1997), “Architectures for product families”, Computers inIndustry,Vol.33No.2–3,pp.165–178.

Falkner, A., Felfernig, A. and Haag, A. (2011), “Recommendation Technologies forConfigurableProducts”,AIMagazine,Vol.32No.3,pp.99–108.

Falkner,A.,Haselböck,A., Schenner,G. and Schreiner,H. (2011), “Modeling and solvingtechnical product configuration problems”, Artificial Intelligence for EngineeringDesign,AnalysisandManufacturing,Vol.25No.2,pp.115–129.

Felfernig, A., Bagley, C., Tiihonen, J., Wortley, L. and Hotz, L. (2014), “Benefits ofconfiguration systems”,Knowledge-Based Configuration: FromResearch toBusinessCases,Elsevier,pp.29–33.

Felfernig,A.,Friedrich,G.andJannach,D.(2001),“Conceptualmodelingforconfigurationofmass-customizableproducts”,Artificial Intelligence inEngineering,Vol.15No.2,pp.165–176.

Felfernig, A., Friedrich, G., Jannach, D. and Stumptner, M. (2004), “Consistency-baseddiagnosisofconfigurationknowledgebases”,ArtificialIntelligence,Vol.152No.2,pp.213–234.

Felfernig,A.,Friedrich,G.,Jannach,D.andZanker,M.(2006),“AnIntegratedEnvironmentfortheDevelopmentofKnowledge-BasedRecommenderApplications”,InternationalJournalofElectronicCommerce,Vol.11No.2,pp.11–34.

Felfernig,A.,Friedrich,G.E.andJannach,D.(2000),“UMLasdomainspecificlanguagefortheconstructionofknowledge-basedconfigurationsystems”,InternationalJournalofSoftwareEngineeringandKnowledgeEngineering,Vol.10No.4,pp.449–469.

Felfernig,A.,Hotz,L.,Bagley,C. andTiihonen, J. (2014),Knowledge-BasedConfigurationFromResearch toBusinessCases,Knowledge-BasedConfiguration,MorganKaufman,availableat:https://doi.org/10.1016/B978-0-12-415817-7.00029-3.

Felfernig, A., Reiterer, S., Stettinger, M., Reinfrank, F., Jeran, M. and Ninaus, G. (2013),“Recommendersystems forconfigurationknowledgeengineering”,CEURWorkshopProceedings,Vol.1128,pp.51–54.

Felfernig, A. and Science, C. (2007), “Standardized Configuration KnowledgeRepresentationsasTechnologicalFoundationforMassCustomization”,pp.31–54.

Fleischanderl,G., Friedrich,G.E.,Haselböck,A., Schreiner,H. and Stumptner,M. (1998),“Configuring largesystemsusinggenerativeconstraintsatisfaction”,IEEEIntelligentSystems,Vol.13No.4,pp.59–68.

Forsythe, D.E. and Buchanan, B.G. (1989), “Knowledge acquisition for expert systems:somepitfallsandsuggestions”, IEEETransactionsonSystems,Man,andCybernetics,Vol.19No.3,pp.435–442.

Forza,C.andSalvador,F.(2002a),“Productconfigurationandinter-firmcoordination:Aninnovativesolution from asmallmanufacturingenterprise”,Computers in Industry,


110

Vol.49No.1,pp.37–46.

Forza, C. and Salvador, F. (2002b), “Managing for variety in the order acquisition andfulfilmentprocess:Thecontributionofproductconfigurationsystems”,InternationalJournalofProductionEconomics,Vol.76No.1,pp.87–98.

Forza, C. and Salvador, F. (2007a), Product Information Management for MassCustomization: Connecting Customer, Front-Office and Back-Office for Fast andEfficientCustomization,PalgraveMacmillan,NewYork.

Forza, C. and Salvador, F. (2007b), Product Information Management for MassCustomization,PalgraveMacmillan,NewYork.

Forza,C.,Trentin,A.andSalvador,F.(2006),“Supportingproductconfigurationandformpostponement by grouping components into kits: the case of MarelliMotori”,InternationalJournalofMassCustomisation,Vol.1No.4,pp.427–444.

Friedrich,G., Jannach,D.,Stumptner,M.andZanker,M.(2014), “KnowledgeEngineeringforConfigurationSystems”,inFelfernig,A.,Hotz,L.,Bagley,C.andTiihonen,J.(Eds.),Knowledge-BasedConfiguration:FromResearchtoBusinessCases,MorganKaufmann,pp.139–155.

Frohlich,M. andWestbrook,R.T. (2001), “Arcsof integration: an international studyofsupplychainstrategies”, JournalofOperationsManagement,Vol.19No.2,pp.185–200.

Gambles, I. (2009),Making the Business Case: Proposals That Succeed for Projects ThatWork,Books.Google.Com,GowerPublishing,Ltd.

Gill, H. (1990), “Adoption of Design Science by Industry — Why so Slow?”, Journal ofEngineeringDesign,Taylor&Francis,Vol.1No.3,p.289.

GonçalvesFilho,E.V. andMarçola, J.A. (1996), “Umapropostademodelagemda listademateriais”,Gestão&Produção,Vol.3No.2,pp.156–172.

Haddix, A.C., Teutsch, S.M. and Corso, P.S. (2003), Prevention Effectiveness: A Guide toDecisionAnalysisandEconomicEvaluation,OxfordUniversityPress.

Hansen,C.L.,Mortensen,N.H.,Hvam,L.andHarlou,U.(2012),“CalculationofComplexityCosts–AnApproachforRationalizingaProductProgram”,ProceedingsofNordDesignConference2012,p.8.

Haug, A. (2009), “A software system to support the development and maintenance ofcomplexproductconfigurators”,TheInternationalJournalofAdvancedManufacturingTechnology,Vol.49No.1–4,pp.393–406.

Haug,A.andHvam,L.(2007),“Themodellingtechniquesofadocumentationsystemthatsupports the development and maintenance of product configuration systems”,InternationalJournalofMassCustomisation,Vol.2No.1–2,pp.1–18.

Haug,A.,Hvam,L.andMortensen,N.H.(2010),“A layouttechniqueforclassdiagramstobeused inproductconfigurationprojects”,Computers inIndustry,ElsevierB.V.,Vol.61No.5,pp.409–418.

Haug,A.,Hvam,L.andMortensen,N.H.(2011),“The impactofproductconfiguratorson

111

leadtimes inengineering-orientedcompanies”,ArtificialIntelligenceforEngineeringDesign,AnalysisandManufacturing,Vol.25No.2,pp.197–206.

Haug, A.,Hvam, L. andMortensen,N.H. (2012a), “Definition and evaluation of productconfiguratordevelopmentstrategies”,ComputersinIndustry,ElsevierB.V.,Vol.63No.5,pp.471–481.

Haug,A.,Hvam, L. andMortensen,N.H. (2012b), “Definition and evaluation of productconfiguratordevelopmentstrategies”,ComputersinIndustry,ElsevierB.V.,Vol.63No.5,pp.471–481.

Haug, A., Mortensen, N.H. and Hvam, L. (2009), “A classification of strategies for thedevelopmentofproductconfigurators”,5thWorldConferenceonMassCustomization&PersonalizationMCPC2009,p.21.

Heatley,J.,Agarwal,R.andTanniru,M.(1995),“Anevaluationofaninnovativeinformationtechnology-thecaseofcarrierexpert”,JournalofStrategicInformationSystems,Vol.4No.3,pp.255–277.

Hegge,H.M.H.(1992),“Agenericbill-of-materialprocessorusingindirectidentificationofproducts”,ProductionPlanning&Control,Vol.3No.3,pp.336–342.

Heiskala, M., Paloheimo, K. and Tiihonen, J. (2005), “Mass customization of services:benefits and challenges of configurable services”, Frontiers of E-Business Research(FeBR2005),pp.206–221.

Heiskala,M.,Tiihonen, J.,Paloheimo,K.-S.andSoininen,T. (2007), “MassCustomizationwithConfigurableProductsandConfigurators:AReviewofBenefitsandChallenges”,IdeaGroupIncMassCustomizationInformationSystemsinBusiness.Hersey,IgiGlobalLondon,No.C,pp.75–106.

Hines,P.,Rich,N.,Bicheno,J.,Brunt,D.,Taylor,D.,Butterworth,C.andSullivan,J.(1998),“ValueStreamManagement”,InternationalJournalofLogisticsManagement,The,Vol.9No.1,pp.25–42.

Von Hippel, E. (2001), “User toolkits for innovation”, Journal of Product InnovationManagement.

Huffman, C. and Kahn, B.E. (1998), “Variety for sale: Mass customization or massconfusion?”,JournalofRetailing,Vol.74No.4,pp.491–513.

Hvam, L., Haug, A., Mortensen, N.H. and Thuesen, C. (2013), “Observed benefits fromproduct configuration systems”, International Journal of Industrial Engineering:Theory,ApplicationsandPractice.

Hvam,L.andLadeby,K.(2007),“Anapproachforthedevelopmentofvisualconfigurationsystems”,Computers&IndustrialEngineering,Vol.53No.3,pp.401–419.

Hvam,L.,Malis,M.,Hansen,B.andRiis,J.(2004),“Reengineeringofthequotationprocess:applicationofknowledgebasedsystems”,BusinessProcessManagementJournal,Vol.10No.2,pp.200–213.

Hvam, L., Mortensen, N.H. and Riis, J. (2008), Product Customization, Springer, BerlinHeidelberg.


112

Hvam,L.,Pape,S.andNielsen,M.K.(2006),“Improvingthequotationprocesswithproductconfiguration”,ComputersinIndustry,Vol.57No.7,pp.607–621.

Häkkinen, L. and Hilmola, O.-P. (2008), “ERP evaluation during the shakedown phase:lessonsfromanafter-salesdivision”,InfoSystemsJ,Vol.18,pp.73–100.

Inakoshi,H.,Okamoto,S.,Ohta,Y.andYugami,N.(2001),“EffectiveDecisionSupportforProduct Configuration by Using CBR”, Proceedings of the Fourth InternationalConferenceonCase-BasedReasoning(ICCBR),p.8.

Jannach, D. and Zanker, M. (2013), “Modeling and Solving Distributed ConfigurationProblems: A CSP-Based Approach”, IEEE Transactions on Knowledge and DataEngineering,Vol.25No.3,pp.603–618.

Jinsong, Z., Qifu, W., Li, W. and Yifang, Z. (2005), “Configuration-oriented productmodelling and knowledge management for made-to-order manufacturingenterprises”,InternationalJournalofAdvancedManufacturingTechnology,Vol.25No.1–2,pp.41–52.

Johnson,R.B.,Onwuegbuzie,A.J.andTurner,L.A. (2007), “Toward aDefinitionofMixedMethodsResearch”,JournalofMixedMethodsResearch,Vol.1No.2,pp.112–133.

Johnston,R. andLawrence,P.R. (1988), “BeyondVertical Integration --TheRiseof theValue-Adding Partnership”, edited by Thompson,G.Harvard Business Review, Sage,London,Vol.66No.4,p.94.

Klein,R.(2007),“CustomizationandrealtimeinformationaccessinintegratedeBusinesssupply chain relationships”, Journal of Operations Management, Vol. 25 No. 6, pp.1366–1381.

Kotler,P.(1989),“FromMassMarketingtoMassCustomization”,PlanningReview,Vol.15No.5,pp.10–47.

Krause, F.-L. (1988), “Knowledge integrated product modelling for design andManufacture”,TheSecondToyotaConference(1988),pp.179–218.

Kristjansdottir, K., Shafiee, S. and Hvam, L. (2016), “Utilizing product configurationsystems for supporting the critical parts of the engineering processes”, IEEEInternationalConferenceonIndustrialEngineeringandEngineeringManagement,pp.1777–1781.

Kruchten, P. (2007), The Rational Unified Process, an Introduction, PhysicianSportsmedicine,3rded.,Addison-WesleyProfessional.

Kruchten,P.andRoyce,W.(1996),“Arationaldevelopmentprocess”,CrossTalk,Vol.9No.July,pp.11–16.

Kumar,D., Chen,W. and Simpson, T.W. (2009), “Amarket-driven approach to productfamilydesign”,InternationalJournalofProductionResearch,Taylor&Francis,Vol.47No.1,pp.71–104.

Ladeby, K. andOddsson, G. (1998), “Reference frame for Product Configuration”, 2011WorldConferenceonMassCustomization,Personalization,andCo-Creation.,availableat:http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Reference+Frame

113

+for+Product+Configuration#0.

Lee,H.L., Padmanabhan, V. andWhang, S. (1997), “InformationDistortion in a SupplyChain:TheBullwhipEffect”,ManagementScience,Vol.43No.4,pp.546–558.

Lindemann,U.,Maurer,M.andBraun,T. (2008),StructuralComplexityManagement:AnApproachfortheFieldofProductDesign,Springer.

Magro,D.andTorasso,P.(2003),“Decompositionstrategiesforconfigurationproblems”,AiEdam,Vol.17No.1,pp.51–73.

Mailharro,D.(1998),“Aclassificationandconstraint-basedframeworkforconfiguration”,Artificial Intelligence forEngineeringDesign,AnalysisandManufacturing(AIEDAM),Vol.12No.4,pp.383–397.

McGuinness, D.L. and Wright, J.R. (1998), “Conceptual modelling for configuration: Adescription logic-based approach”, Artificial Intelligence for Engineering Design,AnalysisandManufacturing,Vol.12No.4,pp.333–344.

McNaughton,B.,Ray,P.andLewis,L.(2010),“DesigninganevaluationframeworkforITservicemanagement”,Information&Management,Vol.47No.4,pp.219–225.

Metters,R.(1997),“Quantifyingthebullwhipeffectinsupplychains”,JournalofOperationsManagement,Vol.15No.2,pp.89–100.

Mittal,S.andFrayman,F.(1989),“TowardsaGenericModelofConfiguratonTasks.”,Ijcai,pp.1395–1401.

Mortensen,N.H.,Harlou,U.,Hvam,L.andHaug,A.(2008),“Improvingdecisionmakinginthe early phases of configuration projects”, International Journal of IndustrialEngineering,Vol.15No.2,pp.185–19.

Mukhopadhyay,T.andKekre,S.(2002),“StrategicandOperationalBenefitsofElectronicIntegrationinB2BProcurementProcesses”,ManagementScience,Vol.48No.10,pp.1301–1313.

Nagel, R.L., Hutcheson, R., McAdams, D. a. and Stone, R. (2011), “Process and eventmodelling for conceptual design”, Journal of EngineeringDesign,Vol. 22No. 3, pp.145–164.

Paetsch, F., Eberlein, A. and Maurer, F. (2003), “Requirements engineering and agilesoftware development”, WET ICE 2003. Proceedings. Twelfth IEEE InternationalWorkshops on Enabling Technologies: Infrastructure for Collaborative Enterprises,2003.,IEEEComput.Soc,pp.308–313.

Petersen, T.D. (2007), “Product Configuration in ETO Companies”, Mass CustomizationInformationSystemsinBusiness,pp.59–76.

Petnga,L.andAustin,M.(2016),“Anontologicalframeworkforknowledgemodelinganddecision support in cyber-physical systems”, Advanced Engineering Informatics,ElsevierLtd,Vol.30No.1,pp.77–94.

Phillips, P.P. and Phillips, J.J. (2010), “Return on investment”, Handbook of ImprovingPerformance in the Workplace, Selecting and Implementing PerformanceInterventions,pp.823–846.


114

Piller, F.T. (2004), “Mass customization: Reflections on the state of the concept”,International JournalofFlexibleManufacturingSystems,Vol.16No.4SPEC. ISS.,pp.313–334.

Pine,B.J.(1993),MassCustomization:TheNewFrontier inBusinessCompetition,HarvardBusinessPress.

Ragatz,G.L.,Handfield,R.B. and Scannell,T.V. (1997), “Success Factors for IntegratingSuppliers into New Product Development”, Journal of Product InnovationManagement,Vol.14No.3,pp.190–202.

Rask, I. (1998), “Rule-based product development-Report 1”, Industrial Research andDevelopmentCorporation,Mölndal.

Ruet,M.andGeneste,L.(2002),“SearchandAdaptation inaFuzzyObjectOrientedCaseBase”,Proc.6thEuropeanConferenceonCBR(ECCBR’02),pp.723–730.

Sabin, D. and Weigel, R. (1998), “Product configuration frameworks-a survey”, IEEEIntelligentSystemsandTheirApplications,Vol.13No.4,pp.42–49.

Saltelli,A.(2002),“Sensitivityanalysisfor importanceassessment”,RiskAnalysis,Vol.22No.3,pp.579–590.

Salvador, F. and Forza, C. (2004), “Configuring products to address the customization-responsiveness squeeze: A survey of management issues and opportunities”,InternationalJournalofProductionEconomics,Vol.91No.3,pp.273–291.

Salvador,F.andForza,C.(2007),“Principlesforefficientandeffectivesalesconfigurationdesign”,InternationalJournalofMassCustomisation,Vol.2,p.2007.

Schreier,M.andPrügl,R.W.(2011),“Thevalue incrementofmass-customizedproducts:Anempiricalassessment”,JournalofConsumerBehaviour,WileyOnlineLibrary,Vol.5No.4,pp.317–327.

Shafiee,S.(2015),“Anagiledocumentationsystemforhighlyengineered,complexproductconfigurationsystems”,Proceedingsofthe22ndEuromaConference,switzerland.

Shafiee,S.,Hvam,L.andBonev,M. (2014), “Scoping aproductconfigurationproject forengineer-to-order companies”, International Journal of Industrial Engineering andManagement,Vol.5No.4,pp.207–220.

Soininen,T.,Tiihonen,J.,Tomi,M.,Sulonen,R.andMännistö,T.(1998),“Towardsageneralontologyofconfiguration”,AIEDAM,Vol.12No.4,pp.357–372.

Stevens, G.C. (1989), “Integrating the Supply Chain”, International Journal of PhysicalDistribution&MaterialsManagement,Vol.19No.8,pp.3–8.

Stroeken, J. (2001), “Information technology, innovation and supply chain structure”,InternationalJournalofServicesTechnologyandManagement,Vol.2No.3–4,pp.269–288.

Studer,R.,Benjaminsc,R.andFensela,D.(1998),“Knowledgeengineering:Principlesandmethods”,Data&KnowledgeEngineering,Vol.25No.1–2,pp.161–197.

Stumptner, M., Friedrich, G.E. and Haselböck, A. (1998), “Generative constraint-based

115

configurationoflargetechnicalsystems”,AiEdam,Vol.12No.4,pp.307–320.

Sunnersjo,S.,Cederfeldt,M.,Elgh,F.andRask,I.(2006),“ATransparentDesignSystemforIterative Product Development”, Journal of Computing and Information Science inEngineering,Vol.6No.3,p.300.

Sviokla,J.(1990),“Anexaminationoftheimpactofexpertsystemsonthefirm:thecaseofXCON”,MISQuarterly,pp.127–140.

Tallon,P.andKraemer,K. (1996), “MultidimensionalAssessmentof theContributionofInformationTechnology to Firm Performance”, Center forResearch on InformationTechnologyandOrganizations.

Tan,K.C.,Kannan,V.R. andHandfield,R.B. (1998), “Supply chainmanagement: supplierperformance and firm performance”, International Journal of Purchasing andMaterialsManagement,No.summer,pp.2–9.

Tidstam, A. and Malmqvist, J. (2010), “Information Modelling for AutomotiveConfiguration”,ProceedingsofNordDesig.

Tidstam,A.andMalmqvist,J.(2011),“Authoringandverifyingvehicleconfigurationrules”,Proceedingsof8thConferenceonProductLifecycleManagement.

Tiihonen,J.,Felfernig,A.andMandl,M.(2014),“PersonalizedConfiguration”,Knowledge-BasedConfiguration:FromResearchtoBusinessCases,ElsevierInc.,pp.167–179.

Tiihonen, J. and Felfernig, a. (2010), “Towards recommending configurable offerings”,InternationalJournalofMassCustomisation,Vol.3No.4,p.389.

Tiihonen, J.,Heiskala,M.,Anderson,A. and Soininen,T. (2013), “WeCoTin – Apracticallogic-basedsalesconfigurator”,AICommunications,Vol.26No.1,pp.99–131.

Tiihonen, J., Lehtonen,T., Soininen,T., Pulkkinen, A. and Sulonen,R. (1998), “Modelingconfigurableproductfamilies”,12thInternationalConferenceonEngineeringDesign,Munich,.

Tiihonen, J., Soininen,T.,Männistö,T. and Sulonen,R. (1996a), “State-of-the-Practice inProductConfiguration—ASurveyof10Cases in theFinnish Industry”,KnowledgeIntensiveCAD,Vol.1,pp.95–114.

Tiihonen, J., Soininen,T.,Männistö,T. and Sulonen,R. (1996b), “State-of-the-Practice inProductConfiguration—ASurveyof10Cases in theFinnish Industry”,KnowledgeIntensiveCAD,SpringerUS,Vol.1,pp.95–114.

Tiihonen, J.,Soininen,T.,Männistö,T.andSulonen,R.(1998),“Configurableproducts—lessons learned from the Finnish industry”, Proceedings of the 2nd InternationalConferenceonEngineeringDesignandAutomation(ED&A ’98),Vol.Maui,Hawa,p.6pp.

Tomiyama,T.,Kiriyama,T.,Takeda,H.,Xue,D.andYoshikawa,H.(1989),“Metamodel:AkeytointelligentCADsystems”,ResearchinEngineeringDesign,Vol.1No.1,pp.19–34.

Trentin,A.,Perin,E.andForza,C.(2011),“Overcomingthecustomization-responsivenesssqueezebyusingproductconfigurators:Beyondanecdotalevidence”,Computers in


116

Industry,ElsevierB.V.,Vol.62No.3,pp.260–268.

Trentin, A., Perin, E. and Forza, C. (2012a), “Product configurator impact on productquality”, International JournalofProductionEconomics,Elsevier,Vol.135No.2,pp.850–859.

Trentin, A., Perin, E. and Forza, C. (2012b), “Product configurator impact on productquality”, International JournalofProductionEconomics,Elsevier,Vol.135No.2,pp.850–859.

Tseng, H.E., Chang, C.C. and Chang, S.H. (2005), “Applying case-based reasoning forproduct configuration in mass customization environments”, Expert Systems withApplications,Vol.29No.4,pp.913–925.

Turley, F. (2007), “An Introduction to PRINCE2”, JISC Infonet - Project ManagementMethodologies,ManagementPlaza.

Van de Ven, A.H. (2015), A Guide for Organizational and Social Research. :, StatewideAgricultural Land Use Baseline 2015, Vol. 1, OUP Oxford, availableat:https://doi.org/10.1017/CBO9781107415324.004.

Verhagen,W.J.C.,deVrught,B.,Schut,J.andCurran,R.(2015),“Amethodforidentificationof automation potential through modelling of engineering processes andquantificationofinformationwaste”,AdvancedEngineeringInformatics,ElsevierLtd,Vol.29No.3,pp.307–321.

Vickery, S., Jayaram, J. andDroge,C. (2003), “The effectsof an integrative supply chainstrategyoncustomerserviceandfinancialperformance:ananalysisofdirectversusindirectrelationships”,JournalofOperationsManagement,Vol.21No.5,pp.523–539.

Voss, C., Tsikriktsis, N. and Frohlich, M. (2002), “Case research in operationsmanagement”, International Journal of Operations & ManagementProductionManagement,Vol.22No.2,pp.195–219.

Wiegers,K. (1999), “First things first:prioritizingrequirements”,SoftwareDevelopment,Vol.7No.9,pp.48–53.

Wikner,J.andRudberg,M.(2005),“Integratingproductionandengineeringperspectiveson the customer order decoupling point”, International Journal of Operations &ProductionManagementVO-25,EmeraldGroupPublishingLimited,Vol.25No.7,p.623.

Yang,D.,Miao,R.,Wu,H.andZhou,Y.(2009),“Productconfigurationknowledgemodelingusingontologyweblanguage”,ExpertSystemswithApplications,ElsevierLtd,Vol.36No.3,pp.4399–4411.

Yin,R.K.(2013),CaseStudyResearch:DesignandMethods,Sagepublications.

Yu, B. and Skovgaard, H. (1998), “A configuration tool to increase productcompetitiveness”,IEEEIntelligentSystems,Vol.4,pp.34–41.

Zhang, L. (Linda) and Rodrigues, B. (2010), “Nested coloured timed Petri nets forproduction configuration of product families”, International Journal of ProductionResearch,Vol.48No.6,pp.1805–1833.

117

Zhang,L.L.,Vareilles,E. andAldanondo,M. (2013), “Genericbill of functions,materials,andoperationsforSAP2configuration”,InternationalJournalofProductionResearch,Vol.51No.2,pp.465–478.

Zhang, Y.,MacCallum,K.J. andDuffy,A.H.B. (1996), “ProductKnowledgeModelling andManagement”,DS53:Proceedingsof the2ndWDKWorkshoponProductStructuring1996.


118

7APPENDEDPAPERS

119

ARTICLEA


120

ARTICLEB

121

ARTICLEC


122

ARTICLED

123

ARTICLEE


124

ARTICLEF

125

ARTICLEG


126

ARTICLEH

127

ARTICLEI


128

ARTICLEJ

129

ARTICLEK

Documents

Conceptual Modelling for Product Configuration …orbit.dtu.dk/files/139515799/Final_Thesis_SARA_recent_changes.pdfCONCEPTUAL MODELLING FOR PRODUCT CONFIGURATION SYSTEMS Sara Shafiee