6

Overview

SixSigma:-ADefinition-AppliedtoGE-GEQualityInitiative-WhyThisApproach?-OriginofSixSigma-The“BreakthroughStrategy”-ArrivingatSigmaSixSigmaStructureKeyConcepts&ToolsAPracticalExampleAnOverviewNotalotofDetails!!6

Overview

“SixSigma”

Ifwecan’texpresswhatweknowintheformofnumbers,wereallydon’tknowmuchaboutit.Ifwedon’tknowmuchaboutit,wecan’tcontrolit.Ifwecan’tcontrolit,weareatthemercyofchance.MikelJ.HarryPresident&CEOSixSigmaAcademy,Inc.ARigorousMethodforMeasuring&ControllingOurQuality“...willbringGEtoawholenewlevelofqualityinafractionofthe

timeitwouldhavetakentoclimbthelearningcurveonourown.”JohnF.Welch,Jr.1995GEAnnualReport6

Overview

WhatDoes“Sigma”Mean?

SigmaisaMeasureoftheConsistencyofaProcessIt(isAlsothe18thLetterintheGreekAlphabet!WhyDoesGENeedAQualityInitiative?GERaisingTheBarNewGoaltobe“BestintheWorld”vs.#1or#2CustomersareExpectingMore,weMustDeliver“Ship-and-fix”ApproachnoLongerToleratedintheMarketAimtoSpeedPastTraditionalCompetitorsin5YearsGoalConsistentwithReducedTotalCostsWeMustAcknowledgeOurVulnerabilitiesPoorQualityThatImpactsCustomersProblemswithNPITooHighInternalCosts6

Overview

WeNeedaMajorInitiativetoMoveFromWhereweAretoWhereweWanttobe6

Overview

WhyDoesGENeedAQualityInitiative?40%35%30%25%20%10%15%5%CostofFailure(%ofSales)DefectsperMillion3.4233621066,807308,537500,000Sigma654321

EstimatedCostofFailureinUSIndustryis15%ofSales;TakingGEFroma3toa6CompanyWillSave~$10.5BillionperYear!Why“SixSigma”?ProvenSuccessfulin“Quality-Demanding”Industriese.g.,Motorola,TexasInstruments(manyprocessstepsinseries)ProvenMethodtoReduceCostsHighlyQuantitativeMethod–ScienceandLogicInsteadofGutFeelIncludesManufacturing&Service(closetocustomer)andProvidesBridgetoDesignforQualityConceptsHasSupportandCommitmentofTopManagementItWorks!!!6

Overview

Sigma3456SpellingMoneyTime1.5

MisspelledWordsperPageinaBook1

MisspelledWordper30PagesinaBook1

MisspelledWordinasetofEncyclopedias1

MisspelledWordinalloftheBooksinaSmallLibrary$2.7MillionIndebtednessper$1BillioninAssets$570

Indebtednessper$1BillioninAssets$63,000

Indebtednessper$1BillioninAssets$2

Indebtednessper$1BillioninAssets31/2MonthsperCentury21/2DaysperCentury30MinutesperCentury6SecondsperCentury6

isSeveralOrdersofMagnitudeBetterThan3!!!Sigma:AMeasureofQuality6

Overview

WhereDoes“SixSigma”ComeFrom?

MikelJ.HarryoneoftheOriginalArchitectsPreviouslyHeadedQualityFunctionatABBandMotorola

NowPresident/CEOofSixSigmaAcademyinPhoenix,ArizonaHasConsultedforTexasInstruments,AlliedSignal(andothers)CurrentlyRetainedbyGEtoTeachtheImplementation,DeploymentandApplicationofSixSigmaConcepts&Tools

LearningfromThoseWhoHavehadSuccessWith6WillAccelerateitsImplementationatGE6

Overview

So...WhatisSixSigma?

AMeasurementSystem

AProblem-SolvingApproach

ADisciplinedChangeProcess“THESIXSIGMABREAKTHROUGHSTRATEGY”MeasureAnalyzeImproveControl6

Overview

How

DoWeArriveatSigma?Measuring&EliminatingDefectsisthe“Core”ofSixSigmaMeasurementSystemIdentifytheCTQsLookforDefectsinProductsorServices

“CriticaltoQuality”CharacteristicsortheCustomerRequirementsforaProductorService

CountDefectsorfailurestomeetCTQ

requirementsinallprocessstepsDefineDefectOpportunities

AnystepintheprocesswhereaDefectcouldoccurinaCTQ

ArriveatDPMO

UsetheSIGMATABLEConvertDPMOtoSigma

DefectsPerMillionOpportunities23456308,53766,8076,2102333.4PPM

DefectsperMillionofOpportunity

SigmaLevel6OverviewMeasurementSystem23456308,53766,8076,2102333.4PPMSIGMALEVELDEFECTSperMILLIONOPPORTUNITYIRSTaxAdviceBestCompaniesAirlineSafetyAverageCompanyGEAirlineBaggageDoctor’sPrescriptionRestaurantBillsAverageCompanyin3to4RangeSomeSigma““Benchmarks””6OverviewMeasurementSystemAGraphic/QuantitativePerspectiveonVariationAverageValueManyDataSetsHaveaNormalorBellShapeNumberofPeopleArrivingatCRDTime7:007:157:307:458:008:158:308:459:009:156OverviewProblemSolvingApproachCenterProcessReduceSpreadXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXOff-TargetUnpredictableOn-Target6HelpsusIdentifyandReduceVARIATIONdueto:-InsufficientProcessCapability-UnstableParts&Materials-InadequateDesignMarginTargetUSLLSLTargetUSLLSLTargetUSLLSLCenterProcessReduceSpreadOff-TargetUnpredictableOn-TargetDefects6OverviewProblemSolvingApproach“LowerSpecificationLimit””“UpperSpecificationLimit””LessVariationMeansFewerDefects&HigherProcessYields6OverviewProblemSolvingApproachKeyComponentsof“BREAKTHROUGHSTRATEGY”MeasureAnalyzeImproveControlIdentifyCTQ&CTP(CriticaltoProcess)VariablesDoProcessMappingDevelopandValidateMeasurementSystemsBenchmarkandBaselineProcessesCalculateYieldandSigmaTargetOpportunitiesandEstablishImprovementGoalsUseofParetoChart&FishboneDiagramsUseDesignofExperimentsIsolatethe“VitalFew””fromthe“TrivialMany”SourcesofVariationTestforImprovementinCenteringUseofBrainstormingandActionWorkoutsSetupControlMechanismsMonitorProcessVariationMaintain“InControl”ProcessesUseofControlChartsandProceduresAMixofConceptsandToolsWillAlsoIntegratewithNPIProcess6OverviewDisciplinedChangeProcessANewSetofQUALITYMEASURESCustomerSatisfactionCostofPoorQualitySupplierQualityInternalPerformanceDesignforManufacturabilityWillApplytoManufacturing&Non-ManufacturingProcessesandbeTracked&ReportedbyEachBusiness6OverviewStructureQualityCouncil

Members:Labs&Functions“Pipeline”&BBProjectPrioritiesTraining&CertificationMeasurements&RewardsCommunicationsChampions

Leadership:OverallInitiativeProjectFundingHR:Training&RewardsBlackBelts

Lead6ProjectTeams“Measure/Analyze”“Improve/Control”O(jiān)utwithBusinessesHereatCRDMasterBlackBelts

Teach6

MentorBlackBeltsMonitorBBProjectsWork“Pipeline”ProjectsAResourcePoolTeamMembers

Learn/Use6ToolsWorkonBBProjectsPartofTheJobOutwithBusinesses6ProjectswiththeGEBusinessesTabulationofGESixSigmaResultsBenefitTarget&UpdateCurrentbenefitslevel@10.865MMQPIDloading:Carryoverfrom1999:4.059CompletedProjects2000:3.313ActiveProjects2000:3.285Total:10.865MMKeyConcepts&Tools6Overview6OverviewChangingFocusFromOutputtoProcess

Y

Dependent

Output

Effect

Symptom

Monitor

X1...XN

Independent

Input-Process

Cause

Problem

ControlIdentifyingandFixingRootCausesWillHelpusObtaintheDesiredOutputf(X)Y=ProcessCapability6OverviewSustainedCapabilityoftheProcess(longterm)USLTTime1Time2Time3Time4InherentCapabilityoftheProcess(shortterm)LSLTargetOverTime,a““Typical”ProcessWillShiftandDriftbyApproximately1.56Overview“ShortTermCentered”versus““LongTermShifted””SixSigmaCenteredLSLUSLTProcessCapabilitySHORTTERM.001ppm.001ppm+6

LONGTERMLSLUSLT3.4ppmSixSigmaShifted1.5ProcessCapabilityHigherDefectYieldinLongTermProcessCapabilitythanShortTermProcessCapability-64.51.56OverviewTyingitAllTogethershiftCDAB0.51.01.52.02.5123456CONTROLPOORGOODTECHNOLOGYPOORGOODABCD

GoodControl/PoorTechnologyPoorControl/PoorTechnologyPoorControl/GoodTechnology

WORLDCLASS!!!shorttermProblemCouldbeControl,TechnologyorBoth6OverviewShortTermCapabilityShortTermCapabilityRatio(Cp)Cp=LSL-6USLExampleUSLLSL3.0==-3.063.0-(-3.0Cp=Cp=1LSLUSLProcessMeanTTargetA3ProcessThePotentialPerformanceofaProcess,ifitWereonTarget6OverviewLongTermCapability(Cpk)CpCpk=LongTermCapabilityRatioExampleCp=1(previouschart)Target=-0.5=0Cpk1-(-0.5-03=Cpk=0.83-Off-TargetPenalty

Target-3

ThePotentialPerformanceofaProcess,CorrectedforanOff-TargetMeanLSLUSLProcessMeanTTargetA3Process6OverviewZ-ScaleofMeasureZ=AUnitofMeasureEquivalenttotheNumberofStandardDeviationsthataValueisAwayfromtheTargetValue-3.0-0.53.0Z-ValuesUSLLSL=ProcessMeanZTTarget0A3ProcessTheDefinitionsofYieldFinalTestProcess(Process4)PassProcess3Process1Process2100(UnitsTested)65708291Yield1Yield2Yield3

Loss1

Loss3Rejects

Loss299125FirstTimeYield(Yft)=UnitsPassedUnitsTested=6570=0.93RolledThruputYield(Yrt)=(Yield1)(Yield2)(Yield3)=91826570(((())))=0.65100917082NormalizedYield(Ynm)==1/n(Yrt)(0.65)1/4=0.89(n:TotalNumberofProcesses)6OverviewYieldExclusiveofReworkProbabilityofZeroDefectsAverageYieldofAllProcesses6OverviewTheImpactofComplexityTheImpactofComplexityRolledRolled

Yield

YieldNumberofOperationsNumberofOperations1.001.000.900.900.800.800.700.700.600.600.500.500.400.400.300.3000.100.000.001101001,00010,000100.0001,000,0001101001,00010,000100.0001,000,000ProcessMeanCenteredonEachOperationProcessMeanCenteredonEachOperation1101001,00010,000100.0001,000,0001101001,00010,000100.0001,000,000RolledRolled

Yield

YieldNumberofOperationsNumberofOperations1.001.000.900.900.800.800.700.700.600.600.500.500.400.400.300.3000.100.000.00AstheNumberofOperationsIncreases,aHighRolledYieldRequiresaHighforEachOperation54366543ProcessMeanShifted1.5atEachOperation6OverviewBaselining&BenchmarkinganExistingProcessp(x)DefectsBenchmarkBaselineEntitlement

BenchmarkAWorld-ClassPerformance

EntitlementAchievablePerformanceGiventheInvestmentsAlreadyMade

BaselineTheCurrentLevelofPerformanceBaselining=CurrentProcess/Benchmarking=UltimateGoalSomeBasic6-RelatedTools6OverviewScatterDiagramOverSleptCarWouldNotStartWeatherFamilyProblemsOtherParetoDiagramFrequencyofOccurenceReasonsforBeingLateforWorkArrivalTimeatWorkTimeAlarmWentOffMaterialsPeopleTheHistogramControlCharts6OverviewSomeBasic6-RelatedToolsTheFishboneDiagramMeasurementsMethodsTechnologyStatementCause&EffectBeingLateforWorkPlotofDailyArrivalTime9:157:007:157:307:458:008:158:308:459:00AverageValueNumberofPeopleArrivingatCRDTime6OverviewLCLUCLRangeChartROutofControlConditionLCLXUCLXBarChartSomeBasic6-RelatedToolsLCL=LowerControlLimitUCL=UpperControlLimitX=MeanR=AverageRangeMonitorsChangesinAverageorVariationOverTimeDesignofExperiments6OverviewSCREENINGOPTIMIZATIONCHARACTERIZATIONForExperimentsInvolvingaLargeNumberofFactorsUsefulinIsolatingthe“VitalFew““fromthe“TrivialMany””ForExperimentsInvolvingaRelativelySmallNumberofFactorsUsefulWhenStudyingRelativelyUncomplicatedEffects&InteractionsForExperimentsInvolvingOnly2or3FactorsUsefulWhenStudyingHighlyComplicatedEffects&RelationshipsDOEisMoreEffectiveThanTestingOneFactorataTime6OverviewUsingthe““OneFactorataTime”Approach

ReduceCommutetoWorkto15Minutes(withoutworkinganabnormalworkschedule)

TheGoalTheVariables

TimeofDeparturefromHome&RouteTakentoWorkTheApproach

Try3PotentialRoutesatCurrentDepartureTime(7:45),SelecttheBest&VarytheDepartureTime‘tilwegetto15MinutesTimeofDeparture3217:157:307:458:008:15RouteCombinationSelectedTheResultUseRoute2andLeaveat7:15toReachGoal6OverviewUsing““DesignofExperiments”(DOE)TimeofDepartureDOE(i)BetterAccountsforInteractiveVariablesMissedby““OneFactorataTime”,and(ii)EfficientlySearchesfor“SweetSpot”inParameterSpaceTheVariablesTimeofDeparturefromHome&RouteTakentoWorkTheApproach

VarytimeofDepartureandRouteSimultaneously,inaSystematicFashionTheResultABetterCombinationAllowing15MoreMinutesofSleep!!!ActualCommutingTimeAverages(minutes)3217:157:307:458:008:15Route172023211915182019161215212018OriginalConclusionBestCombination“SweetSpot”

ReduceCommutetoWorkto15Minutes(withoutworkinganabnormalworkschedule)

TheGoalAPracticalExample(The““Cookbook”)6Overview6andBakingBreadYEAST

FLOURUsinga12StepProcess6OverviewThe“BETTERBREAD”CompanyStep1Selecting“CriticaltoQuality”(CTQsorY)WhatisImportanttotheCustomer?RiseTextureSmellFreshnessTasteY=Taste!!6OverviewMeasureStep2DefiningPerformanceStandardsforCTQsorY6OverviewHowCouldWeMeasureTaste(Y)?PanelofTastersRatingSystemof1to10Target:AverageRatingat8Desired:NoIndividualRatings(“defects”)Below7Y=12345678910TargetDefectsWorstBestButIsthistheRightSystem?Measure6OverviewStep3ValidatingtheMeasurementSystemforYHowCouldWeApproachThis?BlindfoldedPanelRatesSeveralLoafSamplesPut“Repeat”PiecesfromSameLoafinDifferentSamplesConsistentRatings*onPiecesfromSameLoaf=“Repeatability”ConsistentRatings*onSamplesAcrossthePanel=““Reproducibility”“Repeatability”&“Reproducibility”SuggestValidMeasurementApproachPanelMemberLoaf1Loaf2Loaf3A589B491C492D898E482F591G892*WithinOneTasteUnitMeasure6OverviewStep4EstablishProductCapabilityforY(Taste)Thisisa3Process!7Defects(ratingsbelow7)24Ratings(fromourpanel)=.292292,000Defectsper1,ooo,oooLoavesOR765432112345678910#ofRatingsRating64321143Defects<7Target=8AnalyzeHowDoWeApproachThis?BakeSeveralLoavesUnder““Normal”ConditionsHaveTasterPanelAgainDotheRatingAverageRatingis7.4ButVariationistooGreatfora6Process3x10+4x9+6x8+4x7+3x6+2x5+1x4+1x31+1+2+3+4+6+4+36Overview

Step5DefineImprovementObjectivesforY(Taste)HowdoweDefineImprovement?BenchmarktheCompetitionFocusonDefects(i.e.tasterating<7)DetermineWhatisan““AcceptableSigmaLevel”SetImprovementObjectivesAccordinglyMaybea5ProcessWillSuffice!1,000,000-100,000-.............................10,000-.............................1,000-.............................100-.............................10-.............................1-234567“BETTERBREAD””BakingProcessBestCompetitorRangeforImprovementDefectsPerMillionSigmaScale

Freihofer

WONDER

PepperidgeFarm

SunbeamAnalyze6Overview

Step6IdentifySourcesofVariationinY(Taste)HowdoweDeterminethePotentialSourcesofVariation(Xs)?HavetheChefsBrainstormSomeLikelyOnesMightbe:-AmountofSaltUsed-BrandofFlour-BakingTime-BakingTemperature-BrandofYeastYEAST

FLOURMultipleSources:Chefs,Suppliers,ControlsAnalyze6Overview

Step7ScreenPotentialCausesofVariation(Xs)HowdoweScreenforCausesofVariation(Xs)?DesignanExperimentUseDifferentSourcesofPotentialVariationHavePanelRatetheBreadUsedintheExperimentResultsLeadtothe“VitalFew”CausesYEAST

FLOURSourceConclusionNegligibleMajorCauseNegligibleMajorCauseNegligibleFocusonThe““VitalFew”Improve6OverviewStep8DiscoverVariableRelationshipsBetween“VitalFew”(Xs)andYHowdoweFindtheRelationshipBetweenthe“VitalFew””(Xs)andTaste(Y)?ConductaMoreDetailedExperimentFocus:OvenTemperaturefrom325to375and3BrandsofFlourRUN#TEMPBRAND1325A2325B3325C4350A5350B6350C7375A8375B9375C

FLOUR

FLOUR

FLOURBrandABrandBBrandCImproveResults:350&BrandAisBestCombinationofTemperature&Flour

Note:TimeisaFactorOnlyifTemperatureChangesSignificantly6OverviewStep9EstablishToleranceson“VitalFew”(Xs)HowdoweEnsureOvenTemperatureisControlled?DataSuggests350(5)isbestTemperaturetoReduceTasteVariation

BrandAFlourtobeUsedExceptinCaseofEmergency

“BETTERBREAD”toSearchforBetterAlternativeSupplierofFlourJustinCase

FLOURBrandAButIsOurMeasurementSystemCorrect?Improve6OverviewStep10ValidatetheMeasurementSystemforXsHowCouldWeApproachThis?NeedtoVerifytheAccuracyofOurTemperatureGaugesNeedfor“Benchmark”InstrumentationforComparisonRentSomeOther“HighEnd”GaugesComparetheResultsVerifythatourInstrumentsareAccurateControl6OverviewStep11DetermineAbilitytoControlVitalFewXsHowCouldWeApproachThis?CheckANumberofOvensMonitorTemperaturesOverTimeFocusontheProcessCapabilityLookforDegreeofVariationVariationOKBut...AverageisHigh(andthealgorithmshouldbechecked)30345#ofOvensTemperature346357347348349350351352353354355356252015105Control6OverviewStep12ImplementProcessControlSystemonXsWhatdowedoGoingForward?CheckOvensDailyforTemperatureLevelsAuditUsageFrequencyofAlternativeFlourSupplier(e.g.,BrandC)PeriodicallyReassemblethePaneltoTestTasteCharttheResultsAndPlottheDataOverTime

FLOUR“BrandC”354353352351350349348135791113151719212325Control9、靜夜四無無鄰，荒居居舊業(yè)貧。。。12月-2212月-22Wednesday,December7,202210、雨中黃葉葉樹，燈下下白頭人。。。22:38:3022:38:3022:3812/7/202210:38:30PM11、以我獨(dú)沈久久，愧君相見見頻。。12月-2222:38:3022:38Dec-2207-Dec-2212、故人江海別別，幾度隔山山川。。22:38:3022:38:3022:38Wednesday,December7,202213、乍見翻疑夢夢，相悲各問問年。。12月-2212月-2222:38:3022:38:30December7,202214、他他鄉(xiāng)鄉(xiāng)生生白白發(fā)發(fā)，，舊舊國國見見青青山山。。。。07