in automotive manufacturing process on FMEA and grey ...

69

AdvancesinProductionEngineering&Management ISSN1854‐6250

Volume13|Number1|March2018|pp69–80 Journalhome:apem‐journal.org

https://doi.org/10.14743/apem2018.1.274 Originalscientificpaper

Risk management in automotive manufacturing process based on FMEA and grey relational analysis: A case study

Baynal, K.a,*, Sarı, T.b,*, Akpınar, B.c aKocaeli University, Department of Industrial Engineering, Kocaeli, Turkey bKonya Food and Agriculture University, Department of International Trade and Business, Konya, Turkey cGeneral Electric, İstanbul, Turkey

A B S T R A C T A R T I C L E I N F O

Riskmanagement is an important issue inmanufacturing companies in to‐day’scompetitivemarket.Failuremodesandeffectsanalysis(FMEA)methodisa riskmanagement tool to stabilizeproductionandenhancemarketcom‐petitiveness by using risk priority numbers (RPN). Although the traditionalFMEA approach is an effectively and commonly used method, it has someshortcomingssuchasassumptionofequalimportanceofthefactors,severity,occurrence and detectability, and not following the ordered weighted rule.Thus, in order to improveRPN, an integratedmethod combining grey rela‐tional analysis (GRA)with FMEA is used in this study. The purpose of thispaperistocontributetoriskmanagementactivitiesbyproposingsolutionstoassembly line problems in an automotivemanufacturing company by usingcombinedGRAandFMEAmethod. In theproposedmethod, theprioritiesofproduction failures were determined by GRA approach and these failureswereminimized by using FMEA technique. The study results indicated theactionsthatleadtoenhancementintheproduct.Theimplementationofcor‐rective/preventiveactivitiesresultedin96%improvementindoorsealcutsproblemcausedbythedoorstepassembly.Doorsealcutsproblemcausedbyinstrument panel assembly and the noisy doorwindowproblem are solvedcompletely.

©2018PEI,UniversityofMaribor.Allrightsreserved.

Keywords:Automotivemanufacturing;Riskmanagement;Failuremodesandeffectanalysis(FMEA);Greyrelationalanalysis(GRA)

*Correspondingauthor:[email protected](Sarı,T.)

Articlehistory:Received12July2017Revised16January2018Accepted20February2018

1. Introduction

Before a newproduct is introduced to amarket, themanufacturing companies probably facemanyproblemsinthestagesofdesign,plan,productionanddelivery.Itisverycriticaltodetectandsolvetheseproblems,beforetheproductreachesacustomer.Somefailuresareeasytode‐tectwhilesomeofthemremainhidden.Thewholeprocessshouldbeevaluatedcarefullyandtheappropriatequalitycontroltechniquesshouldbeusedinordertofindoutthesehiddenfailures.OneofthemosteffectivemethodstodeterminethefailuresinanyprocessisFailureModeandEffectsAnalysis(FMEA).

FMEAcanbeexpressedasaspecificmethodologyinordertoevaluateaprocess,system,ser‐viceordesignforpossiblewaysinwhichfailurecanoccur[1].Risks,problems,concernsorer‐rorsaredifferenttypeoffailures.Failuremodecanbedescribedasaproductfailingtoperformitsdesired function,describedbytheexpectationsof thecustomers.Failureemerges fromthedeviationfromstandardsintheconditionsofmachine,method,materialandworkforce,affect‐ingthequalityofaproductoraprocess.TheFMEAanalysisfollowsawell‐definedsequenceof

Baynal, Sarı, Akpınar

70 Advances in Production Engineering & Management 13(1) 2018

stepsthatincludes(1)failuremode,(2)failureeffects,(3)causes,(4)detectability,(5)correc‐tiveorpreventiveactionsand(6)rationaleforacceptance[2].Today,withtheincreasingcom‐petition in themarket, any deficiency and deviation in product performance result inmarketshareloss.AlthoughthetraditionalFMEAemployingriskprioritynumbers(RPN)isanefficientandeffectivetooltostabilizeproductionandenhancemarketcompetitiveness,ithasbeencriti‐cizedforthefollowingshortcomings:its(1)highduplicationrate,(2)notfollowingtheorderedweightedrule,(3)assumptionofequalimportanceofseverity(S),occurrence(O),anddetecta‐bility(D)and(4)failuretoconsiderthedirectandindirectrelationshipsbetweenthemodesandthecausesoffailure[3].

In thisstudy,an integratedmethodcombiningFMEAandGreyRelationalAnalysis(GRA) isusedinordertoovercometheshortcomingsoftraditionalFMEAmethod.GRAisusedtodeter‐mine the priorities of production failures in an automotivemanufacturing company. The twofailures,doorsealcutandnoisywindowproblems,havebeenminimizedbyusingFMEAmeth‐odology.

2. Literature review

FMEAmethodologyiswidelyusedtomanageriskinindustriessuchasmanufacturing,automo‐tive,andaerospace.VinodhandSanthosh[4]reportedanapplicationofdesignFMEAtoanau‐tomotive leaf springmanufacturing organization in India. Implementation of fuzzy developedFMEAmethodtoaircraftlandingsystem,whichisoneoftheimportantpotentialfailuremodeinaerospaceindustry,hasshownthestrengthofthemethodinmanagingrisk[5].Chang[6]com‐binedgeneralizedmulti‐attributeFMEAandmulti‐attributeFMEAtoimproveLCDmanufactur‐ingprocessinacompanyinTaiwan.SegismundoandMiguel[7]proposedamethodologicalap‐proach to effective risk management in new product development in a Brazilian automakercompanybyusing FMEA technique. Bandukaetal. [8] integrated lean approachwith processFMEAinautomotiveindustry.Liuetal. introducedariskprioritymodelbycombininghesitant2‐tuple linguistic term sets and an extended QUALIFLEXmethod and FMEAmethodology forhandlingahealthcareriskanalysisproblem[9,10].Barkovicetal.usedFMEAmethod in im‐provementofnewspaperproductionsystemquality[11]. GRAhasbeenusedbymanagerstomakedecisionsunderuncertaintyinmanydifferentareassince1982.FengandWang[12]measuredthefinancialperformanceofairwaycompanieswiththehelpofGRA.HsuandWen[13]proposedadesigntodealwiththetrafficandflightfrequencyinairwaysusingGRA.InthestudyofLinandLin,oneofthetechniquesusedforoptimizationofwireerosionsystemwasgreyrelationanalysismethod[14].Wangetal.[15]proposedahybridmethodology using grey relational analysis and experimental design to solve several multi‐criteriadecisionmakingproblemssuchas,aflexiblemanufacturingsystem,arapidprototypingprocess and an automated inspection system. Palanikumar et.al. [16] optimized the results ofpolymermaterialprocesswithgreyrelationanalysismethod.RajeswariandAmirthagadeswa‐ran[17]usedgreyrelationalapproachtoimprovemachinabilitypropertiesofendmillingpro‐cess.Wang [18] developed amodel formeasuring the performance of logistic companies viagreyrelationalanalysismethod.Rameshetal. [19]proposedaneffectivemodel to investigateturningofmagnesiumalloybyusinggreyrelationalanalysismethod. A combinationof FMEAandGRA techniques areusedbyauthors inorder to eliminate theshortcomings of FMEA. Pillay andWang [2] used an integratedmethod combining FMEAandgrey theory to investigate thesystem failures in fuzzyenvironment foranocean‐going fishingvessel in their study. Bagheryetal. [20] implemented process FMEAmethod combiningwithDEA(dataenvelopmentanalysis)andGRAinanautomotivecompanyproducingautopartsforSamand,Peugeot405andPeugeot206.


Advances in Production Engineering & Management 13(1) 2018 71

3. Materials and methods

In thisstudyacombinedmethodologyof failuremodesandeffectanalysisandgreyrelationalanalysis is used. The priorities of production failureswere determined byGRA approach andfailureswereminimizedbyusingFMEAtechnique.

3.1 FMEA method

FMEAwasfirstdevelopedasanassessmenttool to improvetheevaluationofthereliabilityofmilitarysystemsandweaponsintheUSarmyinthelate1940s.ThismethodwasalsousedforApollo space missions in the 1960s by the National Aeronautics and Space Administration(NASA)[3].Inthelate1970sFMEAwasusedbyFordMotorCompanyinautomotiveproductionprocesses [6]. Because these applications resulted in satisfactory improvements in the FordCompany, themethodhasbeenwidelyused inautomotive industryasa riskassessment tool.TodayFMEA isappliedsuccessfully to industries suchasaircraft, automotive,medicine, semi‐conductorsandfoodindustry.IntheFMEAapproach,foreachofthefailuresidentified(whetherknownorpotential),anestimateismadeofitsoccurrence(O),severity(S)anddetection(D)[1].Occurrenceistheprobabilityofoccurrenceofthefailureanditscause.Detectionisanevalua‐tionprocesstofindpotentialfailuresintheproduct.Severityisanexpressionofimportanceandemergency of potential systemdefaultmode. FMEA technique evaluates the risk of failure byusingRPNs.TheRPNvalueisfoundbytakingtheproductofS,O,andDonascalefrom1to10.HigherRPNvalueindicatesahigherpriority.

3.2 Grey relational analysis

Greyrelationalanalysis isamulti‐criteriadecisionmakingmethodusedbydecisionmakerstotaketherightdecisionundercircumstanceswithlimitedanduncertaindata[21].GRAapproachexploressystembehaviorusingrelationalanalysisandmodelconstructions[2].Thegreysystemprovidessolutionstoproblemswheretheinformationisincomplete,limitedorcharacterizedbyrandomuncertainty.Thegrey theoryhasbecomeapopular techniqueprovidingmultidiscipli‐nary approaches in recent twenty years. The grey relational analysis was first developed byJulongDengin1982[22].Themodelincludesthreetypesofinformationpoints:white,greyorblack.Themaingoal is to transferblackpoints in thesystem to thegreypoints.Greyrelationanalysisconsistsofsixbasicsteps.Thesestepsareexplainedbelow[19,23,24]:

Step1:Constructanormmatrix . It is assumed that therearen data sequences includingmcriteria:

1 2 …1 2 …… … … …1 2 …

(1)

where istheentityinthei‐thdatasequencecorrespondingtothej‐thcriterion.

Step2:Sincemulti‐criteriadecisionmaking(MCDM)problemsmaycontainavariationofdiffer‐ent criteria, the solution needs normalization. Normalization process based on properties ofthreetypesofcriteria,largerthebetter,smallerthebetter,andnominalthebest:

; larger the better (2)

; smaller the better (3)

1| |

,; nominalthebest (4)

isthetargetvalueforthecriterionj,and ≤ ≤ .



Step3:NormalizethedatasetandgenerateareferencesequencebasedonEq.2toEq.4.Nor‐malizedmatrixisexpressedas :

1 2 …1 2 …… … … …1 2 …

(5)

Step4:Calculateabsolutevaluetable.Thedifferencebetweenanormalizedentityanditsrefer‐encevalueiscalculated.Thedifferenceisshownas∆ .

∆ | | (6)

∆

∆ 1 ∆ 2 … ∆∆ 1 ∆ 2 … ∆… … … …

∆ 1 ∆ 2 … ∆

(7)

Step5:Computegreyrelationalcoefficient ,applyingfollowinggreyrelationalequation:

∆ ∆∆ ∆

(8)

where ∆ ∆ , ∆ ∆ , and ∆ and ∈ 0,1 . isthe distinguishing index and inmost cases it takes the value of 0.5 offeringmoderate distin‐guishingeffect.

Step6:Compute thegreyrelationaldegree.Greyrelationaldegreewhich indicates themagni‐tudeofcorrelationorsimilarity.Theoverallgreyrelationaldegree Γ iscalculatedbytakingaveragevalueofgreyrelationalcoefficientsbyusingthefollowingequation:

Γ (9)

where referstotheweightofthej‐thcriterion.Thesumoftheweightsofallcriteriamustequalto1.

3.3 Integration of grey theory and FMEA method

ThetraditionalFMEAmethodcannotassignthepossibilityofoccurrenceoffailure,itsdetecta‐bilityandseveritycomplywith therealworld.The integrationofgrey theory toFMEAallowsengineersanddecisionmakerstoassignrelativeweightsdependingonresearchandproductionstrategies.Indecisionmakingproblems,thefactorserieswiththehighestgreyrelationdegreegives the best alternative. The greater the relation degreemeans the smaller effect of failuresourceinFMEAapplication.Forthisreason,theincreasingrelativedegreeshowsthedecreaseinriskpriorityofpotentialsourceswhichhavetobeimproved.

4. Case study

Thecasestudywasheld inacarmanufacturing company in theTurkishautomotive industry.Theaimofthestudywastosolvetheassemblylineproblems.Thecompanymainly facedtwotypesofproblems.Thefirstonewasthecardoorsealproblemandthesecondonewasthenoisycarwindowproblem.

4.1 Formulation of problems and causes

Thecardoorsealproblemcanbeexplainedasatearorcutinthesealofthecardoors.Thedoorsealservesasabarrierprotectingtheinnercaragainstdustandwaterfromtheoutside.Ifthe



sealisdamagedandifthisdamagecannotbedetectedthroughqualitycontrolprocesses,itmaycause severe customer complaints. The noisy carwindowproblem is the annoying noise andshakingproblemwhenthewindowglassmovesupanddown.

ThefactoryhasusedParetoanalysistofindoutthefailuresinmanufacturingandtheiroccur‐renceprobability.Thenumbersabouttheoccurrenceprobability,thecausesoffailuresandthefailuredetectionpointsaregiveninTable1.Thelinepointinthetablerepresentstheproblemsdetectedthroughthecontrolpointsof theassembly line itself.Finalpoint is thepointofade‐tailedcontrolofthecarjustbeforeitleavestheassemblyline.Pre‐qualitypointisacontrolpointfor repairedcarsbeforequalitycontrol.Afterqualitycontrolof theproducts, fiveof themareverycarefullycontrolledindetailataqualitycontrolpoint.Thecompany’sexpertteamhasde‐termined two important production problems and the most probable causes by using FMAEtechnique.Theresultingcausesarelistedbelow:

Causesforcardoorsealcutortearproblemaresummarizedbelow:

Step: TearorcutinthecardoorsealduringthedoorstepassemblyIP: TearorcutinthecardoorsealduringinstrumentpanelassemblyDoorlock: TearorcutinthecardoorsealduringdoorlockassemblySeat: TearorcutinthecardoorsealduringcarseatassemblyOperator: Damagingofthecardoorsealbyoperatorduringplacementoftheseal

Causesfornoisywindowproblemareasfollows:

Rivetposition: TheeffectofthepositionoftherivetofwindowmechanismFixingequipment: TheincompatibilitybetweenwindowmechanismandfixingequipmentHoleposition: Theeffectofthepositionoftherivethole

Table1Problemsandcauses

Failure Cause DetectionPoint

Cut/tearindoor

seal

Line Final Pre‐quality Quality Detailedquality Customer

Step 0 4 0 5 0 2IP 0 5 0 2 0 0Doorlock 0 2 0 2 0 0Seat 0 0 0 2 0 0Operator 0 0 0 2 0 0

Noisy

window Line Final Pre‐quality Quality Detailedquality Customer

Rivetposition 0 88 0 9 0 0Fixingequipment 0 108 0 20 0 0Holeposition 0 190 0 20 0 0

4.2 Probability of occurrence, detectability, severity

Occurrence(O):

Occurrenceistheprobabilityoffailureoccurrenceanditscause.Theprecautionsfordetectingthefailuresarenottakenintoconsiderationinthisstep.Onlythemethodsdeterminedforpre‐venting failureareconsidered. If theprocess isunderstatisticalprocesscontrol, theevolutiondependson thestatisticaldata.Otherwise, intangibledata fromjudgmentsareused forevolu‐tion.Inthefactory,theoccurrenceandthedetectionpointsoffailuresareexpressedbyParetoanalysisandthentransferredtoa"1‐10"scale.Inthetable, theOcolumndescribestheoccur‐renceofprobabilitybetweenthevalues1and10.



Table2CalculationofOvaluesbasedonParetoanalysis

Failure Cause DetectionPoint Cut/tearindoor

seal

Line Final Pre‐quality Quality Detailedquality Customer O

Step 0 54 0 55 0 12 10IP 0 25 0 32 0 0 7Doorlock 0 12 0 0 0 0 3Seat 0 0 0 12 0 0 3Operator 0 0 0 12 0 0 3

Noisy

window Line Final Pre‐quality Quality Detailedquality Customer O

Rivetposition 0 88 0 9 0 0 7Fixingequipment 0 190 0 20 0 0 10Holeposition 0 109 0 20 0 0 9

Detectability(D):

Detectionisanevaluationprocesstofindthepotentialfailuresinaproduct,beforeitleavestheassemblyline.Thefailureshouldbeacceptedasithasoccurredandthecriteriaforfailuredetec‐tionshouldbedetectedbefore theproducthasbeen introducedtoaconsumer. In the factory,according to results fromPareto analysis, the failures are scored for their detectionpoints tocalculateDvalues.ThescaleusedinthecalculationofDvalueisbelow:

Linepoint: 1‐2Finalpoint: 3‐4Pre‐quality: 5‐6Quality: 7‐8Detailedquality: 9Customer: 10

Sinceallthefailuresaredetectedmorethanonceindifferentpoints,Dvaluesarecalculatedbytheweighedmatrix(Table3)andbasedonQLSParetoanalysis(Table4).

Table3WeightedDvalues


Cut/tearindoor

seal

Line Final Pre‐quality Quality Detailedquality Customer D

Step 0 54X4 0 55X8 0 12X10 776IP 0 25X4 0 32X8 0 0 356Doorlock 0 12X4 0 0 0 0 48Seat 0 0 0 12X8 0 0 96Operator 0 0 0 12X8 0 0 96

Noisy

window Line Final Pre‐quality Quality Detailedquality Customer D

Rivetposition 0 88X4 0 9X8 0 0 424Fixingequipment 0 190X4 0 20X8 0 0 920Holeposition 0 109X4 0 20X8 0 0 596

Table4CalculationofDvaluesbasedonQLSParetoanalysis


Cut/tearindoor

seal

Line Final Pre‐quality Quality Detailedquality Customer D

Step 0 54X4 0 55X8 0 12X10 6IP 0 25X4 0 32X8 0 0 6Doorlock 0 12X4 0 0 0 0 4Seat 0 0 0 12X8 0 0 8Operator 0 0 0 12X8 0 0 86

Noisy

window Line Final Pre‐quality Quality Detailedquality Customer D

Rivetposition 0 88X4 0 9X8 0 0 4Fixingequipment 0 190X4 0 20X8 0 0 4Holeposition 0 109X4 0 20X8 0 0 5



Severity(S):

Severity is an expression of importance and urgency of a potential systemdefaultmode. Theonlyevaluationcriterionforseverity istheeffectofa failure.Theseveritydegreesaredefinedaccording to thedegreeof theeffectsonproduct, system, customerand legalobligations.Thefailurescalematrixof the factory’squalitysystem isuseddirectly in thisstudy(Table5).Theseverityvaluesarecalculatedwiththehelpofthistable.Table6givestheSvaluesdeterminedbythecompany’sexpertsusingqualityleadershipsystem(QLS)Paretoanalysis.

Table5Failureincreasematrixforseveritycalculations

FailureSeverity(QualityStandards)

PickLevel

IncreaseLevel

Lineteamleader

Teamleader

Areamanager

Qualityassur‐ancemanager

Factorymanager

CauseorQuality

Blitz(10‐9)Failureeffectsauto/drivercontrol,customersafetyandlegalconditions.

1 X X X 3 X X

Sampling 1 X X X X 3 X

CauseorQuality

A(8‐7)Failureisveryannoyingandcustom‐erfilesacomplainttovendor/service.

1 X X X 3 X

Sampling 1 X X X X 3 X

CauseorQuality

B(6‐5)Failureisannoying,causingcustomerunsatisfactionandcomplaintsofguarantee.

5 X X 8 X 10 X X

Sampling 2 X X X 4 X X

CauseorQuality

C(4‐3)Failureisdetectedbyeducat‐ed/criticalcustomersanditneedslongtimeimprovements.

10 X X 15 X

Sampling 4 X X X 6 X

Table6SeveritycalculationbasedonQLSParetovalues


Cut/tearindoor

seal

Line Final Pre‐quality Quality Detailedquality Customer S

Step 0 54 0 55 0 12 7IP 0 25 0 32 0 0 7Doorlock 0 12 0 0 0 0 7Seat 0 0 0 12 0 0 7Operator 0 0 0 12 0 0 7

Noisy

window Line Final Pre‐quality Quality Detailedquality Customer S

Rivetposition 0 88 0 9 0 0 6Fixingequipment 0 190 0 20 0 0 6Holeposition 0 109 0 20 0 0 6



4.3 Calculation of risk priority number (RPN)

RPNiscalculatedbymultiplicationofO,DandSvalues.RPNshowstherelativeimportanceoffailurecauses.TheresultingrankofRPNvalueshelpthedecisionmakerstodecidewhichcauseshouldbeimprovedfirst.ThehighesttheRPNvaluemeansthefirstrate.TherankingaccordingtoRPNisshowninTable7.

Table7Riskprioritynumbers

Failure Cause O D S RPN Rank

Cut/tearin

doorseal

Step 10 6 7 420 1

IP 7 6 7 294 2

Doorlock 3 4 7 84 6

Seat 3 8 7 168 5

Operator 3 8 7 168 5

Noisy

window Rivetposition 7 4 6 168 5

Fixingequipment 10 4 6 240 4

Holeposition 9 5 6 270 3

4.4 Calculation of grey relational coefficient

TheRPNinTable7aretransferredtogreyRPNvaluesandreorderedbyusinggreyrelationalanalysis.ThenthedifferencematrixisconstructedbyusingEq.8:

1 2 31 2 31 2 31 2 31 2 31 2 31 2 31 2 3

10 6 77 6 73 4 73 8 73 8 77 4 610 4 69 5 6

1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3

9 5 66 5 62 3 62 7 62 7 66 3 59 3 58 4 5

Accordingtodifferencematrix,∆min=2,∆max=9andisassumedas0.5value.Aftercalculationofdifferencematrix,greyrelationalcoefficientsarecalculatedbyusingEq.8.

∆ ∆∆ ∆

2 0.5 99 0.5 9

0.481

Thefollowingmatrixisconstructedbyusinggreyrelationalcoefficients:



1 2 31 2 31 2 31 2 31 2 31 2 31 2 31 2 3

0.481 0.684 0.6190.619 0.684 0.6191.000 0.867 0.6191.000 0.565 0.6191.000 0.565 0.6190.714 1.000 0.7890.556 1.000 0.7890.600 0.882 0.789

ThelaststepistocalculatetheGreyRPNtodeterminethepriorities.Table8showstheweightsofcostbasedpriorities.

Table8Costbasedweights WO WD WS

Cost 2.6€ 1.3€ 2.6€Weight 0.4 0.2 0.4

GreyrelationaldegreesarefoundbytheformulainEq.8.Thegreyrelationaldegreeofthefirstfailureleveliscalculatedas0.577byusingEq.9.

Γ 0.481 0.4 0.684 0.2 0.619 0.4 0.577

TheweightedgreyRPNvaluesarefoundasfollows:

WeightedGreyRPN=

0.5770.6320.8210.7610.7610.8230.7590.742

TheRPNandGreyRPNvalues are listed comparatively inTable9.As shown in the table, theweightofrivetpositiondiffersfromonemethodtotheother.AccordingtotherankinginTable9, themost importantproblem is thedoorseal cutscausedbystepassembly.Thesecond im‐portantproblemisthesealcutscausedbyinstrumentpanelassembly.Thethirdoneisthenoiseproblemof thewindowcausedbythepositionof thehole.The least importantproblemisde‐finedasnoisywindowproblemcausedbyrivetholeposition.Thepriorityofdecisionmakersistoinitiateimprovementonthesemosturgentproblems.

Table9ThecomparisonofFMEARPNandgreyRPN

Failure Cause O D S FMEARPN Rank GreyRPN Rank

Cut/tearin

doorseal

Step 10 6 7 420 1 0.577 1IP 7 6 7 294 2 0.632 2Doorlock 3 4 7 84 6 0.821 6Seat 3 8 7 168 5 0.761 5Operator 3 8 7 168 5 0.761 5

Noisy

win‐

dow Rivetposition 7 4 6 168 5 0.823 7

Fixingequipment 10 4 6 240 4 0.759 4Holeposition 9 5 6 270 3 0.742 3



4.5 Results and discussion

The solutions are developed according to the resulting grey RPN ranks. After improvementsbetterresultsinthequalitymetricsareobtained.

Thesolutionfordoorsealcutsproblemcausingfromdoorstepassembly:

Sinceadoorsealpreventswateranddustleakages,atearorcutinthedoorsealcausescus‐tomercomplaints.Therearefivebasiccausesforthedoorsealproblems.BasedonTable9,themostimportantreasonforthisproblemistearorcutindoorsealduringstepassemblyprocess.Whenthedoorstepassemblyprocesswasinspectedindetail,itwasdeterminedthatthesharpcornersofthestepcausedcutsinthesealduringassemblyofthestepbyanopera‐tor.Asapartofcorrectiveorpreventiveactivities,alltheareasofsealtowherethestepcor‐ners hitwere detected.Magnetic protectorsweremade. The operators have begun to usetheseprotectorsinrelevantareasduringassembly.Onemonthlater,qualityrecordsindicat‐edanimportantdecreaseinsealcutsbytherateof96%.

Thesolutionfordoorsealcutsproblemcausingfrominstrumentpanelassembly:

Cutortearindoorsealduringassemblyofinstrumentpanelgetsthesecondrankinpriority.Thesharp‐edgedframeoftheinstrumentpanelswasidentifiedasthecauseforthisproblem.Thecorrectiveandpreventiveactivitiesweredevelopedaseffectivesolutionstotheproblem.Asaresultofcorrectiveorpreventiveactivities,thepotentialtangibleareasofsealwerede‐termined.Magneticprotectorsweredesigned toprotect thesurfaceswhichare likely tobedamaged.Theoperatorshavebeguntousetheseprotectorsinrelevantareasduringassem‐bly.Onemonthlater,qualityrecordsindicatedthatcutsandtearsindoorsealcausedbyin‐strumentpanelassemblywerepreventedbytheratioof100%.

Thesolutionfornoisywindowglassproblemcausingfromholeposition:

Noisywindowglassisaproblemwhichcausesadisturbingnoiseandjoltinthevehicle,whilethewindowismovingupanddown.AccordingtoParetoanalysis,themostimportantreasonwith the third lowestdegree inpriority level is the rivet holeposition.Rivetingprocess inwindowinstallationwere inspected indetailand itwasdetectedthatthedistancebetweenmechanismandtherivetholewastoosmall(2mm).Thisshortdistancecausedthemecha‐nismpartstohittherivetwhichresultedinadisturbingnoiseandjoltinwindows.Asaresultofcorrective/preventiveactivities,thepositionofrivetholewasmovedtoa3mmlowerpo‐sition.Thereforethedistancebecame5mmwhichwassufficientforpreventingthehittingofwindowmechanismparts.Thepreventiveactivitieshaveresultedin100%improvementinthenoiseprobleminonemonth.

Thequality reports indicated that2operatorshave spent48workinghours in amonth todealwithqualityproblemsbeforeimprovements.AfterimplementationofgreyFMEAtechnique,thistimewasreducedto2hourswhichmeanssavingcostby2300Euroinamonthand27600Euroinayear.

5. Conclusion

FMEAiswidelyusedasanefficientdecision‐makingtooltocontrolthestabilityofthemanufac‐turingprocessandtoimproveproductandsystemperformancebydecreasingfailurerate.Alt‐houghthetraditionalFMEA,employingriskprioritynumbers,stabilizeproductionandincreasethemarketcompetitiveness,ithassomelimitationssuchasfailingtoevaluatetherelativerela‐tionshipofeachweightofthoseparameters.InthisstudythelimitationsofFMEAareovercomebyusinganintegratedmethodofgreytheoryandFMEA.First,thepossiblecausesoffailureandtheirdetectionpointsaredeterminedbyFMEA.Second,theprioritiesofthefactors(causes)aredeterminedbyusinggreyRPNvalues.Accordingtotheresultsofcaseapplicationinanautomo‐tivemanufacturing factory, a 96% improvement was achieved for a door seal cuts problemcausedbythedoorstepassembly.A furtherdoorsealcutsproblemcausedbythe instrument


panel assembly was solved completely. As a third improvement, the noisy door window prob-lem, caused by riveting hole position, is prevented by 100 %.

The main advantage of the integrated GRA and FMEA method in this study is the flexibility of assigning weight to each factor in FMEA, providing an effective and consistent methodology to identify weak parts in the component studied. This integrated approach is convenient to deal with risk assessment problems under circumstances where the information is incomplete or uncertain. The processing of linguistic information based on expert knowledge and experience enables a realistic, practical and flexible way to express judgments.

References

[1] Stamatis, D.H. (2003). Failure mode effect analysis: FMEA from theory to execution, 2nd edition, ASQ Quality Press, Wisconsin, USA.

[2] Pillay, A., Wang, J. (2003). Modified failure mode and effects analysis using approximate reasoning, Reliability Engineering & System Safety, Vol. 79, No. 1, 69-85, doi: 10.1016/S0951-8320(02)00179-5.

[3] Chang, K.H., Chang, Y.C., Tsai, I.T. (2013). Enhancing FMEA assessment by integrating grey relational analysis and the decision making trial and evaluation laboratory approach, Engineering Failure Analysis, Vol. 31, 211-224, doi: 10.1016/j.engfailanal.2013.02.020.

[4] Vinodh, S., Santhosh, D. (2011). Application of FMEA to an automotive leaf spring manufacturing organization, The TQM Journal, Vol. 24, No. 3, 260-274, doi: 10.1108/17542731211226772.

[5] Yazdi, M., Daneshvar, S., Setareh, H. (2017). An extension to fuzzy developed failure mode and effects analysis (FDMEA) application for aircraft landing system. Safety Science, Vol. 98, 113-123, doi: 10.1016/j.ssci.2017.06. 009.

[6] Chang, K.H. (2016). Generalized multi-attribute failure mode analysis, Neurocomputing, Vol. 175, Part A, 90-100, doi: 10.1016/j.neucom.2015.10.039.

[7] Segismundo, A., Miguel, P.A.C. (2008). Failure mode and effects analysis (FMEA) in the context of risk manage-ment in new product development: A case study in an automotive company, International Journal of Quality & Reliability Management, Vol. 25, No. 9, 899-912, doi: 10.1108/02656710810908061.

[8] Banduka, N., Veža, I., Bilić, B. (2016). An integrated lean approach to process failure mode and effect analysis (PFMEA): A case study from automotive industry, Advances in Production Engineering & Management, Vol. 11, No. 4, 355-365, doi:10.14743/apem2016.4.233.

[9] Liu, H.C., Li, P., You, J.X., Chen, Y.Z. (2015). A novel approach for FMEA: Combination of interval 2-tuple linguistic variables and gray relational analysis, Quality and Reliability Engineering International, Vol. 31, No. 5, 761-772, doi: 10.1002/gre.1633.

[10] Liu, H.C., You, J.X., Li, P., Su, Q. (2016). Failure mode and effect analysis under uncertainty: An integrated multiple criteria decision making approach, IEEE Trnasactions on Reliability, Vol. 65, No. 3, 1380-1392, doi: 10.1109/TR. 2016.2570567.

[11] Borković, J., Milčić, D., Donevski, D. (2017). Implementation of differentiated quality management system and FMEA method in the newspaper production, Tehnički Vjesnik – Technical Gazette, Vol. 24, No. 4, 1203-1211, doi: 10.17559/TV-20160222082713.

[12] Feng, C.M., Wang, R.T. (2000). Performance evaluation for airlines including the consideration of financial ratios, Journal of Air Transport Management, Vol. 6, No. 3, 133-142, doi: 10.1016/S0969-6997(00)00003-X.

[13] Hsu, C.I., Wen, Y.H. (2000). Application of grey theory and multi objective programming towards airline network design, European Journal of Operational Research, Vol. 27, No. 1, 44-68, doi: 10.1016/S0377-2217(99)00320-3.

[14] Lin, J.L., Lin, C.L. (2002). The use of the orthogonal array with grey relational analysis to optimize the electrical discharge machining process with multiple performance characteristics, International Journal of Machine Tools and Manufacture, Vol. 42, No. 2, 237-244, doi: 10.1016/S0890-6955(01)00107-9.

[15] Wang, P., Meng, P., Zhai, J.Y., Zhu, Z.Q. (2013). A hybrid method using experiment design and grey relational analysis for multiple criteria decision making problems, Knowledge-Based Systems, Vol. 53, 100-107, doi: 10.1016/j.knosys.2013.08.025.

[16] Palanikumar, K., Karunamoorthy, L., Karthikeyan, R. (2006). Multiple performance optimization of machining parameters on the machining of GFRP composites using carbide (K10) tool, Materials and Manufacturing Pro-cesses, Vol. 21, No. 8, 846-852, doi: 10.1080/03602550600728166.

[17] Rajeswari, B., Amirthagadeswaran, K.S. (2017). Experimental investigation of machinability characteristics and multi-response optimization of end milling in aluminium composites using RSM based grey relational analysis, Measurement, Vol. 105, 78-86, doi: 10.1016/j.measurement.2017.04.014.

[18] Wang, Y.J. (2009). Combining grey relation analysis with FMCGDM to evaluate financial performance of Taiwan container, Expert Systems with Applications, Vol. 36, No. 2, Part 1, 2424-2432, doi: 10.1016/j.eswa.2007.12.027.

[19] Ramesh, S., Viswanathan, R., Ambika, S. (2016). Measurement and optimization of surface roughness and tool wear via grey relational analysis, TOPSIS and RSA techniques, Measurement, Vol. 78, 63-72, doi: 10.1016/ j.measurement.2015.09.036.


https://doi.org/10.1016/S0951-8320(02)00179-5

https://doi.org/10.1016/j.engfailanal.2013.02.020

https://doi.org/10.1016/j.engfailanal.2013.02.020

https://doi.org/10.1108/17542731211226772

https://doi.org/10.1016/j.ssci.2017.06.009

https://doi.org/10.1016/j.ssci.2017.06.009

https://doi.org/10.1016/j.neucom.2015.10.039

https://doi.org/10.1108/02656710810908061

http://dx.doi.org/10.14743/apem2016.4.233

http://onlinelibrary.wiley.com/enhanced/exportCitation/doi/10.1002/qre.1633

https://doi.org/10.1109/TR.2016.2570567

https://doi.org/10.1109/TR.2016.2570567

https://doi.org/10.17559/TV-20160222082713

https://doi.org/10.17559/TV-20160222082713

https://doi.org/10.1016/S0969-6997(00)00003-X

https://doi.org/10.1016/S0377-2217(99)00320-3

https://doi.org/10.1016/S0890-6955(01)00107-9

https://doi.org/10.1016/j.knosys.2013.08.025

https://doi.org/10.1016/j.knosys.2013.08.025

https://doi.org/10.1080/03602550600728166

https://doi.org/10.1016/j.measurement.2017.04.014

https://doi.org/10.1016/j.eswa.2007.12.027




[20] Baghery, M., Yousefi, S., Rezaee, M.J. (2016). Risk measurement and prioritization of auto parts manufacturing processes based on process failure analysis, interval data envelopment analysis and grey relational analysis, Journal of Intelligent Manufacturing, Vol. 1, 1-23, doi: 10.1007/s10845-016-1214-1.

[21] Chan, J.W.K., Tong, T.K.L. (2007). Multi-criteria material selections and end-of-life product strategy: Grey rela-tional analysis approach, Materials & Design, Vol. 28, No. 5, 1539-1546, doi: 10.1016/j.matdes.2006.02.016.

[22] Deng, J.L. (1989). Introduction to grey system theory, The Journal of Grey System, Vol. 1, No. 1, 1-24. [23] Zhai, L.Y., Khoo, L.P., Zhong, Z.W. (2009). Design concept evaluation in product development using rough sets

and grey relation analysis, Expert System with Applications, Vol. 36, No. 3, Part 2, 7072-7079, doi: 10.1016/ j.eswa.2008.08.068.

[24] Wu, H.H. (2002). A comparative study of using grey relational analysis in multiple attribute decision making problems, Quality Engineering, Vol.15, No. 2, 209-2017, doi: 10.1081/QEN-120015853.


https://doi.org/10.1007/s10845-016-1214-1

https://doi.org/10.1016/j.matdes.2006.02.016



https://doi.org/10.1081/QEN-120015853

in automotive manufacturing process on FMEA and grey ...

Documents