六西格瑪的基本術語課件

6綠帶介紹Introductionto6GreenBelt

六西格瑪的基本術語課件1Content1.6QualitySystem---Whyneed2.6---Overview3.6---Methodology

3.1.6---Define 3.2.6---Measure 3.3.6---Analysis 3.4.6---Improve 3.5.6---Control4.ConclusionContent21.6QualitySystem---Whyneed1.1.從產品飽受競爭威脅的觀點 以美國為例，1975-1985年，日本挾其『反向技術』(ReversedTechnology)優(yōu)勢，將其高品質的產品推向美洲大陸，使得美國一向以經濟與技術領先的盟主地位發(fā)生動搖。1.6QualitySystem---Wh3

在1970-1989年間美國的市場佔有率：＊電視機從50%降至2%＊收音機從50%降至2%＊汽車從78%降至28%＊影印機從90%降至20%＊照相機從90%降至5%＊鋼鐵從40%降至30%※其結果造成美國貿易赤字每月高達80-100億美元 在1970-1989年間美國的市場佔有率：4美國與日本製程能力之比較年代 美國 日本 製程能力品質水準製程能力品質水準 19700.67 2

1.00 3

1980初 4

1980中 4~5

1980末~1990初 Motorola6方法：不合格率3.4PPM相當於6

美國與日本製程能力之比較5美國企業(yè)競爭力衰退的原因美國管理文化中含有不信任的氣氛，這種表現在嚴格的審核、管制、檢查的管理制度上，無形造成過度的管理成本上漲?！哼@種管理監(jiān)督的作風，大體受到泰勒式科學管理影響的結果』。美國企業(yè)競爭力衰退的原因6日本經營之神松下幸之助在一次對美國企業(yè)界人士發(fā)表演說，提到：『你們的公司經營是以泰勒法則為基礎，更糟的是頭腦皆已泰勒化，因此堅信正確的管理，應是管理者在一邊，工人在另一邊，一邊的人只管思考，另一邊的只管工作。給你們一句忠告：「管理是執(zhí)行者將觀念轉移到員工身上的一種藝術」』。因此美國開始檢討其品質，各種品質系統(tǒng)亦相繼提出，其中包含6品質系統(tǒng)。日本經營之神松下幸之助在一次對美國企業(yè)界人士發(fā)表演說，提到：71.2.由需求的觀點在70年代，產品達到2便達到標準。在80年代，品質要求已提升至3

，但此標準美國會發(fā)生以下事件：※每年有20,000次配錯藥事件※每年有超過15,000個嬰兒出生時會被拋落地上※每年平均有9小時沒有水、電、暖氣供應※每星期有500宗做錯手術事件1.2.由需求的觀點8雖然3合格率已達到99.73%的水平，但相信各位對以上品質要求並不滿意。所以有很多公司已要求「6」的品質管理，其合格品率為99.99966%。在3水準，由1000個零件組成的產品中，每15個產品中只有1個產品是好的。在6水準則1000個產品卻有996.6好的。傳統(tǒng)以『百分率』水準作為設計品質水準，如今變更為以『百萬分率(ppm)』作為衡量品質的水準。雖然3合格率已達到99.73%的水平，但相信各位對以上品質91.3.從成本的觀點1.3.從成本的觀點101.4.從時代趨勢的觀點(1)時間經濟附加價值開端成長成熟衰退典型的產品生命週期1.4.從時代趨勢的觀點(1)時間經濟附加價值開端成長成熟111.4.從時代趨勢的觀點(2)經濟附加價值大型電腦主機迷你電腦與微電腦個人電腦掌上型電腦、電子書網路電視、GPS、行動電話電腦晶片的生命週期19471985199020001.4.從時代趨勢的觀點(2)經濟附加價值大型電腦主機迷你121.4.從時代趨勢的觀點(3)全球化與經濟附加價值農業(yè)時代工業(yè)時代資訊時代?時代時間與主要的技術時代6000BC1760195020001.4.從時代趨勢的觀點(3)全球化與經濟附加價值農業(yè)時代131.5.從品管大師的觀點J.M.Juran,1994年在美國品質管理學會年會會上說,“20世紀以生產力的世紀載入史冊,未來21世紀是品質的世紀”1.5.從品管大師的觀點141.6.從策點管理的觀點欲建立及維持組織競爭優(yōu)勢，效率、創(chuàng)新、品質及顧客回應扮演著主要角色。較佳的效率競爭優(yōu)勢低成本差異化較佳的創(chuàng)新較佳的品質較佳的顧客回應1.6.從策點管理的觀點較佳的效率競爭優(yōu)勢低成本較佳的創(chuàng)新151.7.從近代品質系統(tǒng)的觀點※ISO-9000Effectiveness:5※QS-9000Effectiveness:10※MalcolmBaldrigeGuidelineEffectiveness:25※EuropeanQualityAwardEffectiveness:301.7.從近代品質系統(tǒng)的觀點16※TQMEffectiveness：35※6

---TheLittleQEffectiveness:50※TheUltimate6

---TheBigQEffectiveness:90(上述品質系統(tǒng)均於80年代末期開展出)

※6&TheUltimate6將是一趨勢※TQM172.6Overview6,thewaytocreateprofit.SigmaImproveCustomerSatisfaction&ProfitsincreaseDefectsDecreasecostsDecrease2.6OverviewSigmaCustomer18

6History(1)Somethingmustbewrong1975MotorolaTVbusinessfailedduetopoorprofitandsoldtoaJapaneseCompanyAssigncorporatequalityoffices1980Corporatemovement“GreatQualityAwakening”program1981FocusonQualityandTotalCustomerSatisfaction(TCS)6History(1)Somethingmus19

6History(2)EstablishMotorolaTraining&EducationCenter1981LaunchQualitySystemReview(QSR)program19825years,10×Qualityimprovementgoalset6theoryandconceptinitialized1986BillSmith&MikelHarrypresentedtheideatoCEOGalvin6History(2)EstablishMot20

6History(3)QuantitativeOperationQualityInitiative1987MotorolaTexasInstrumentBusiness-wideStrategicManagement1995Nokia,GE,AlliedSignal,Hitachi,Panasonic,Sony,Whirlpool,Honeywell,Boeing,Dupont6History(3)Quantitative216InnovationModeling(3c-customer,competitor,company)(3e-excitation,employment,entertainment)(3p-product,process,person)InnovationBoxChangeofBusinessSituation(3C)Object(3P)Method(3E)Purpose.Profit.Skill-Up6InnovationModelingInnovat22TheManyAspectsof6

ToolSymbolMetricMethodValueGoalVisionBenchmarkPhilosophyTheManyAspectsof6ToolS236asaMetric---Thehighlevelofsigma,thelowertheprobabilityofproducingadefect.Spec.LimitTargetLimitSomeChanceofDefect3Spec.LimitTargetLimitMuchLessChanceofDefect66asaMetric---Thehighlev246asaMetric

1 691,500 2 308,537 3 66,807 4 6,210 5 233 6 3.4(Shifted1.5) From3processto6process:about20,000 timesimprovement6DPMO6asaMetric6DPMO25如何6導入一般來說，從3到4階段，是企業(yè)可以自行改善的範圍，但這樣的品質標準並沒有辦法讓企業(yè)變得很有競爭力；從4到5階段，就必須找尋可以學習的標竿企業(yè)(Benchmarking)，當成比較與學習的對象；而5到6階段時，品質就已經不是製造出來的，而是設計出來的(DFSS---DesignforSixSigma)。如何6導入266Activity6activityistofindoutcriticalfactorstoquality(CTQ)atcustomer’spointofviewandtoreducethedefectslessthan3.4DPMO(PPM).LSLUSL？

DefectM6inR&D6inMFG.CompanyCustomerCustomerCTQVOCVOB63.4DPMONewStd.366800DPMOPastStd.6ActivityLSLUSL？DefectM627HowDoWeImproveProcessCapability?IncreasethetoleranceDecreasethespreadorvariationofthe processShifttheaverageby:Centeringtheaverageifthespec.has twolimitsDecreaseorincreasetheaveragefor spec.withonelimitHowDoWeImproveProcessCapa28TheCostOpportunity1.523456051015202530About15%ofSales,CostOpportunityon3CompanyCostofFailure(%ofSales)TheCostOpportunity1.523429TheCostofPoorQuality(COPQ)LongCycleTimes,MoreSetups,ExpeditingCosts,LostSales,EngineeringChangeorder,Overtime,Latedelivery,LostOpportunity,LostCustomerLoyalty,ExcessInventoryRejects,Warranty,Inspection,Scrap,ReworkTheCostofPoorQuality(COPQ30WhoisImplementing6Motorola 1987TexasInstrument 1988ABB(AseaBrownBoveri) 1993AlliedSignal 1995GeneralElectric 1995Kodak 1995Siemens,Nokia,Sony 1997

WhoisImplementing6316Benefits(Case1)---Motorolaover12yrsIncreaseproductivityanaverageof12.3%peryearReducedthecostofpoorqualitybymorethan84%Eliminated99.7%ofinprocessdefectsSavemorethan$11billioninmanufacturingcostRealizedanaverageannualcompoundedgrowthrateof17%inearnings6Benefits(Case1)---Motoro326Benefits(Case2)GeneralElectric(MillionDollars)CostProfit1996240,2001997400,7001998450,12001999520,25202000600,30006Benefits(Case2)GeneralEl336Benefits(Case3)TexasInstruments

Before(1988) After(1993)COPQ 30% 7%DPMO 104000 9000Scrap $3M $0.3MYield 84.5% 98.9%CycleTime(Week) 11 4InventoryCost $3.9M $1.1M6Benefits(Case3)TexasIns34TheFocusof6IfweareperfectlycontrolX,shouldweconstantlytestandinspectY?Y=f(X)YX1X2X3X4X1、X2、…XnIndependentInput-ProcessCauseProblemControlYDependentOutputEffectSymptomMonitorTheFocusof6Y=f(X)YX1X356isApplyingOverallBusinessSystemR&D6Trans-actionMfgDesignSSManufacturingSSTransactionalSS6isApplyingOverallBusine366MethodologyMeasureAnalyzeDesignVerifyMeasureAnalyzeImproveControlDefineTranslateYNNewProcess/Product?AchievableGoal?NY6MethodologyMeasureAnalyzeD37Whatisdifferentwith6TraditionalQuality6CentralOrganizedNotFormalStructurefor ToolApplicationLackofSupportinusing ToolsLackofStructured TrainingInspectQualityin(Focus on“Y”)BlackBeltReportDirectlyIntoFunctionStructuredUseof StatisticalToolstoAidProblemSolvingData-BasedDecisionsStructuredTraininginAppliedStatisticsControlProcessInput(X’s)Whatisdifferentwith6Trad38

6Organization

6OrganizationExecutive

ProjectSelectionandSupport6TechnicalLead6CoachProject&BB6ProjectTeamLead6ProjectTeamMemberAllstaffs(6overview)

ChampionMasterBlackBelt(MBB)BlackBelt

(BB)GreenBelt(GB)WhiteBelt(WB)6Organization6Organizat393.6Methodology5StageofDMADVandNewTools ----DesignForSixSigma(DFSS) ----BlackBelt5StageofDMAICmethodology ----GreenBelt3.6Methodology403.1.DMAICMethodology 5StagesDMAICMethodologyandStatisticalTools Phase:DefineSteps;WorkBreakdown&Tool D1ValidateBusinessOpportunity;3CAnalysis,IdentifyCustomer,Cost/EffectAnalysisD2DefineCustomerRequirement;VOC,VOB,QFDD3ProjectPlanning;ProjectTeam,ProjectCharter,COPQ3.1.DMAICMethodology41DefineSteps(1)VisionBusinessStrategyBigY(CTQ)Smally---Project(Goal,Scope,PerformanceIndex)DefineSteps(1)VisionBusiness42DefineSteps(2-1)Vision:最具競爭力的企業(yè)(GE)Strategy---TheThreeCircle(GE)核心事業(yè)照明大型家電馬達高科技事業(yè)醫(yī)療系統(tǒng)工業(yè)電子航太服務事業(yè)信用公司資訊服務核能服務DefineSteps(2-1)核心事業(yè)高科技事業(yè)服務事43DefineSteps(2-2)---TheThreeCircle(GE)JackWelch(GE’SCEO)畫三個圓圈：核心、高科技與服務。GE公司未來都要在其中一個圓圈內。公司任何人不再任一個圓圈內者，未來將會被解雇。圓圈外的的事業(yè)將被整頓、關閉或出售。三個圓圈的策略讓GE公司找到焦點，不再是集團(似乎什麼都做)。DefineSteps(2-2)---TheThre44DefineSteps(3)SelectTargetProduct/ServiceAnalysisBusinessProcessAnalysisCoreProcessIdentifyCustomerListentoVOCSpecifyCustomerRequirementSelectCTQ(BigY)SpecifyCTQ(SmallY)EvaluatePotentialProjectsSelectProject&BuildEffectiveTeamDefineSteps(3)SelectTargetA45DefineSteps(4)CTQ---CriticaltoQuality

只要是顧客要求的，就是關鍵品質(CTQ)。亦稱為『重要成果』、『特殊限制』或流程的『Y變數』。CTQ係任何會直接影響顧客對產品/服務品質觀感之因素。顧客『在乎』的事，就是企業(yè)或組織在乎的事?！盒曼c子』要採納『外部觀點』(Outside-inPerspective)，即從顧客的眼光來看企業(yè)或組織的一切。DefineSteps(4)463.2.DMAICMethodology 5StagesDMAICMethodologyandStatisticalTools Phase:Measure

Steps;WorkBreakdown&Tool M1SpecifyProject;CTQTree,ProcessMap,PerformanceIndexM2AssessMeasurementSystem;MeasurementSystemAnalysis,GageR&RM3IdentifySigmaLevel;GraphAnalysis,CapabilityAnalysis,ConfidenceInterval3.2.DMAICMethodology47M1SpecifyProjectM1stepcoversthefollowings:CTQBreakdownDefineScopeProcessMappingParetoAnalysisDefineperformanceindexandspecificationsThesuccessofany6activitiesreliesmostlyontheCTQdefinitionandselectionM1SpecifyProjectThesuccess48DefinePerformanceIndexCustomerRequirementsInternalProcessImprovedProcessCustomerSatisfactionIndexofcurrentlevelIndexofimprovedlevelPerformprojectforimprovementClearQuantifiableSimpleDefinePerformanceIndexCustom49ExampleofPerformanceIndexYieldCycletimeDefectrateMachinefailurerateCustomerstand-byhoursNumberofinvoiceerrorsElapseTimefromloanapplicationtomoneytransfertothecustomeraccountHourstakenfromreceivingordertodeliverytothecustomerExampleofPerformanceIndex50M2MeasurementSystemAnalysisM2stepcoversthefollowings:StatisticfundamentalsSamplingplanDatacollectionplanVariationofmeasurementsystemGageR&RstudyImprovementofmeasurementsystemWithoutaccuratemeasure,can’tidentifyanysymptomofproblemM2MeasurementSystemAnalysis516MetricsDataType Statisticmeasurementdata Comparisonto specification Z-valueDiscreteDataContinuousDataDefectunitopportunity(DPMO)Average,Standardvariation,ShapeSPECZ-valueSigmacalculation6MetricsDataType Statistic52DefinitionofDPMOterminologyDPU:DefectperUnit(Defect:Anythingthatresultsincustomer dissatisfaction; Anythingthatresultsin nonconformance)DPO:DefectperOpportunityDPMO:DefectperMillionOpportunitiesDefinitionofDPMOterminology53ExampleofSigmacalculation:DiscreteDataCase(1):Overthelastseveralyears,youhavecollecteddataontripstotheairport.Ofthe100tripssampled,youhavemissedonly5flights.IfyouindicatethissituationasSigmalevel,whatwillitbe?Sol.DPU=DPO=5/100=0.05 1-0.05=0.95, Z=1.65

level=Z+1.5=1.65+1.5=3.15ExampleofSigmacalculation:54Case(2).Anaccountingtermconductedaninternalauditforthefinancialreportofyear2001andtheyobserved25nonconformingrecord.Therewere2stepsofprobableincorrectentryidentifiedintheprocessandthetotalobservationwas25,000records.WhatisDPO?WhatisDPMO?WhatisSigmaleveloffinancialreportingprocess?Sol. DPU=25/25000, DPO=25/(2*25000) =0.002,DPMO=0.002*10^6=2000 1-0.002=0.998, Z=2.878

level=Z+1.5=2.878+1.5=4.378Case(2).Anaccountingtermc55ExampleofSigmacalculation:ContinuousData(1)1MeanMedianMode708090100110120130+-NormalDistributionwithmean=100andstd.=10ExampleofSigmacalculation:56ContinuousData(2)(1)MathematicalModel；(2)Continuous；(3)Smooth；(4)Symmetrical;(5)TailasymptotictoX-axis；(6)Bellshaped；(7)Mean=Median=Mode(8)Totalareaundercurve=1(1)Thenormaldistributionhasthefollowingproperties.68.27%ofthedatafallwithin195.45%ofthedatafallwithin299.73%ofthedatafallwithin3(2)Inordertoassessthequalityoftheprocess,wemustcomparetheprocesscharacteristics(viathelocation,spreadandshape)tothespecificationlimitsandtargetedvalue.ContinuousData(2)(1)Mathemat57ContinuousData(3)StandardizationofNormalDistributionTheSigmaofaProcessisthenumberofstandarddeviationbetweenthemeanandtheSpecificationLimits.1Z=(X-)/

No.ofstandarddeviationUSLSigmaoftheprocessZ=4.5ContinuousData(3)1Z=(X-)58MeasurementSystemandMeasurementError(1)Measurementsystemisviewingasaprocess.Sourcesofvariation:5M1EValidatepossiblesourcesofvariationinthemeasurementprocessVariationinmeasurementSystemMeasurementSystemandMeasure59MeasurementSystemandMeasurementError(2)Averagem(Total)=m(Product)+m(Measurement)Variabilitys2(Total)=s2(Product)+s2(Measurement)Deflectionofmeasurementsystem(Tobedecidedbycalibration)Variationofmeasurementsystem(TobedecidedbyR&Rassessment)MeasurementSystemandMeasure60ProcessVarianceObservedProcessVariationActualProcessVariationMeasurementVariationLongtermprocessvariationShorttermprocessvariationVariationwithinasampleVariationduetooperatorsVariationduetogageReproducibilityLinearityStabilityRepeatabilityAccuracyProcessVarianceObservedProce61GageR&RAnalysisTypesofGageR&RAnalysisMethodologyXbar-RMethodANOVATypesofVariationestimationbytheGageR&REquipmentVariation:EVAppraiser(Operator):AVGageR&RAnalysis62GR&RDecisionandImprovementDirectionGageR&RDecisionCriteria%GR&R10%(Goodmeasurementsystem)10%%GR&R30%(Maybeused)%GR&R30%(Cannotused)GageimprovementdirectionForrepeatabilityerrorreproducibilityerror(Needtotakenanactiontooperator)Forreproducibilityerrorrepeatabilityerror(Needtotakenanactiontogage)GR&RDecisionandImprovement63M3IdentifySigmaLevelM3stepcoversthefollowings:DatastratificationGraphicalcauseeliminationUnderstandproductcapabilityandproductperformanceCalculatecurrentsigmalevelM3IdentifySigmaLevel64GraphicalAnalysisRunChartScatterDiagramBoxPlotHistogramChangestotheprocessmaybemoreeasilyrecognizedgraphicallythantabularly.

GraphicalAnalysis65UnderstandingVariationWhatisVariation?◎DifferentoutcomeofaprocessorresultofaproductorserviceMeasurementindexscatteredfromcentervalueVariationwillbeappearedineveryprocessandthetargetofimprovementistoreduceit’svariationsWhyavariationmightbeoccurred?◎Byacommoncause◎Byaspecialcause◎5M1EWhatimpactswillbehappenedifavariationbecomebig?Cannotpredict/forecastoutcomesOccurre-inspectionand/orreworkDelayschedule◎IncreasecustomerdissatisfactionUnderstandingVariation66Short-Term/Long-TermRelationshipsInherentCapabilityoftheProcess---ShortTermCapabilitySustainedPerformanceoftheProcess---LongTermPerformanceOvertime,a“typical”processwillshiftanddriftbyapproximately1.5Short-Term/Long-TermRelations67ProductCapability(Cp)Cp=|USL-LSL|/6st；

Zst=3CpCapabilityIndex(Cpk)Cpu=min(Cpu,Cpl)， WhereCpu=(USL-)/3st； Cpl=(-LSL)/3st； Zst=3CpkPerformanceIndex(Cpk)Cpk=min(Cpu,Cpl)， Where Cpu=(USL-)/3lt； Cpl=(-LSL)/3lt； Zst=3CpkProductCapability(Cp)683.3.DMAICMethodology 5StagesDMAICMethodologyandStatisticalTools Phase:Analysis

Steps;WorkBreakdown&Tool A1SetupImprovementGoal;Benchmarking,Entitlement,KANOA2IdentifyPotentialCauses;Pareto,Brainstorming,Cause&EffectDiagram,LogicTreeA3VerifyPotentialCauses;Regression,HypothesisTest,ANVOA,Multi-VariAnalysis3.3.DMAICMethodology69A1SetUpImprovementGoalInM3thecurrentproductcapabilitiesweredefined.StepA1providesthetoolstodefineperformanceobjectives.Thepurposeistosetobjectivetoestablishabalancebetweenimprovingcustomersatisfactionandavailableresources.A1SetUpImprovementGoal70A1stepcoversfollowingtopics:EstablishperformancegoalsAssessshort-termandlong-termsigmaintermsofmeasurementbenefits(decreaseinCOPQ,increaseinrevenue)DetermineimprovementdirectionbythegapanalysisbetweengoalandcurrentlevelA1stepcoversfollowingtopic71PathstoPerformanceGoalsPerformanceGoalsProcessImprovement(Entitlement)ProcessChange(Breakthrough)BaselineProcessMeasurementPathstoPerformanceGoalsPerf72Benchmarking---WhyBenchmark(1)?InnovationContinuousImprovementBreakthroughImprovementContinuousImprovementTimePerformanceBenchmarking---WhyBenchmark73Benchmarking---WhyBenchmark(2)?DevelopandImprovementStrategicGoalsEstablishactionableobjectivesProvidesenseofurgencyEncouragebreakthroughthinkingCreateabetterunderstandingofyourindustryBenchmarking---WhyBenchmark74A2IdentifyPotentialCausesFindingmainindependentvariables,andmakingthelist Y=f(X)Object Makelistofpotentialindependentvariablesforchanging&improvingYvaluemethodsBrainstorming◎Cause&EffectDiagramMulti-voting ◎LogicTreeA2IdentifyPotentialCauses75Cause&EffectDiagramVariationinmeasurementSystemCause&EffectDiagramVariatio76A3VerifyPotentialCausesOverviewInstepA3,settingapossiblepriorityastotheirimportancetothepotentialvariables(X’s)identifiedinstepA2PurposeGeneralalistofimportantfactors(vitalfew)fromthepotentialvariables. Y=f[(x1,x2,x3)(x4,x5,….,xn)]VitalFewTrivialManyA3VerifyPotentialCauses Y=77ParetoAnalysisParetoAnalysis78CorrelationAnalysisAstatisticalanalysistoinvestigate/measurementofassociationbetweentwovariables(X,Y)iscalledanalysis.CorrelationtellsyouthetrendofYwhenXvalueincrease/decrease.CorrelationAnalysisusingScatterDiagramAnalysis.CorrelationcoefficientindicatesclosenessofarelationshipbetweenXandY.CorrelationAnalysis79RegressionAnalysis(1)

Y=f(x1,x2,x3,…)StatusandcharacteristicsofaprocessModelingMathematicalequationXYRegressionAnalysis(1)Status80RegressionAnalysis(2)TypesofRegressionModelSimplelinearregressionMultiplelinearregressionNon-linearregressionY=f(X)Y:dependentvariableX:independentvariableRegressionAnalysis(2)Y=81TypesofHypothesisTestTypesofDataDiscreteDataContinuousDataMeanTestt–TestANOVAVarianceTestF–TestChi-SquareTypesofHypothesisTestTypes823.4.DMAICMethodology 5StagesDMAICMethodologyandStatisticalTools Phase:ImproveSteps;WorkBreakdown&Tool I1IdentifyVitalFew;ScreeningDOE,Streamlining,ForceFieldAnalysisI2FormulateOptimumModel;OptimizingDOE(RSM),SolutionSelectionMatrixI3SetUpOperatingWindow;OperatingWindow,ResponsibilityMatrix3.4.DMAICMethodology83I1IdentifyVitalFewInanalysisstep,potentialX’swereidentifiedwithapossibleprioritizationastotheirimportanceincontrollingY.ImprovestepprovidestoolstoexploreanduncoverthevitalX’sandhowmuchimpacteachXhasontheresponseY.OneofthemajortoolsusedforbothexplorationandconfirmationoftheeffectofX’sontheY’sisexperimentaldesign(DOE).Byexperimentallymanipulatingtheindependentvariables,DOEprovidesanefficientandeconomicalmeansforreachingvalidandrelevantconclusionsaboutaprocess.I1IdentifyVitalFew84TypesofExperimentalDesignScreening ◎Fullfactorial ◎2Kfactorial ◎FractionalfactorialOptimization ◎Multi-levelexperiments ◎Compositedesign ◎ResponsesurfacemethodTypesofExperimentalDesignSc85I2FormulateOptimumModelThevitalfewimportantX’swereidentifiedandcharacterizedinI1.ThetoolsinI2followamethodologyforestablishinglevelsoftheseX’sthatwillprovideimprovedperformancefortheY.I2FormulateOptimumModel863.5.DMAICMethodology 5StagesDMAICMethodologyandStatisticalTools Phase:ControlSteps;WorkBreakdown&Tool C1IdentifyControlCapability;GageR&R,CapabilityStudy,GapAnalysisC2EstablishControlPlan;ControlPlan,ControlChartC3LinktoManagementSystem;ControlSystem,SixSigmaAudit3.5.DMAICMethodology874.結語(1)6是一以統(tǒng)計為基礎之品質系統(tǒng),目的在使企業(yè)或組織的產品/服務品質達到近於完美的境界.盡量減少決策過程中的“人為因素”(HumanElement).成功取決於一致性(Consistency)---Vision,BusinessStrategy,SpecifyProject,andany6Activities.6品質系統(tǒng)猶“舊酒裝新瓶”,但有較強的結構性.其中,更強化IE&M的角色與重要性,非其他品質系統(tǒng)所能比擬.4.結語(1)884.結語(2)學習(Learning)、價值觀(Value)與創(chuàng)新(Innovation)為企業(yè)思想的核心,其凌駕階級與傳統(tǒng)之上.以學習建立自信,有勇氣嘗試,有勇氣追求與動手做以前認為不可能的事.一旦達成自我實現後,學習這個重要觀念就站上價值觀的中央舞臺上.『持續(xù)學習,並將其轉化為行動的欲望與能力』.4.結語(2)894.結語(3)外部觀點(Outside-inPerspective)

,『由外向內看是一個很重要的想法,強迫自己每件事都由外向內看』.以過程為導向,持續(xù)不斷改進,滿足顧客的需求與期望,『顧客把一切看在眼底.我們無法將任何事隱藏著不讓顧客看到』,『必須讓每一員工都關心顧客』.4.結語(3)90報告完畢敬請指教萬分謝謝六西格瑪的基本術語課件916綠帶介紹Introductionto6GreenBelt

六西格瑪的基本術語課件92Content1.6QualitySystem---Whyneed2.6---Overview3.6---Methodology

3.1.6---Define 3.2.6---Measure 3.3.6---Analysis 3.4.6---Improve 3.5.6---Control4.ConclusionContent931.6QualitySystem---Whyneed1.1.從產品飽受競爭威脅的觀點 以美國為例，1975-1985年，日本挾其『反向技術』(ReversedTechnology)優(yōu)勢，將其高品質的產品推向美洲大陸，使得美國一向以經濟與技術領先的盟主地位發(fā)生動搖。1.6QualitySystem---Wh94

在1970-1989年間美國的市場佔有率：＊電視機從50%降至2%＊收音機從50%降至2%＊汽車從78%降至28%＊影印機從90%降至20%＊照相機從90%降至5%＊鋼鐵從40%降至30%※其結果造成美國貿易赤字每月高達80-100億美元 在1970-1989年間美國的市場佔有率：95美國與日本製程能力之比較年代 美國 日本 製程能力品質水準製程能力品質水準 19700.67 2

1.00 3

1980初 4

1980中 4~5

1980末~1990初 Motorola6方法：不合格率3.4PPM相當於6

美國與日本製程能力之比較96美國企業(yè)競爭力衰退的原因美國管理文化中含有不信任的氣氛，這種表現在嚴格的審核、管制、檢查的管理制度上，無形造成過度的管理成本上漲?！哼@種管理監(jiān)督的作風，大體受到泰勒式科學管理影響的結果』。美國企業(yè)競爭力衰退的原因97日本經營之神松下幸之助在一次對美國企業(yè)界人士發(fā)表演說，提到：『你們的公司經營是以泰勒法則為基礎，更糟的是頭腦皆已泰勒化，因此堅信正確的管理，應是管理者在一邊，工人在另一邊，一邊的人只管思考，另一邊的只管工作。給你們一句忠告：「管理是執(zhí)行者將觀念轉移到員工身上的一種藝術」』。因此美國開始檢討其品質，各種品質系統(tǒng)亦相繼提出，其中包含6品質系統(tǒng)。日本經營之神松下幸之助在一次對美國企業(yè)界人士發(fā)表演說，提到：981.2.由需求的觀點在70年代，產品達到2便達到標準。在80年代，品質要求已提升至3

，但此標準美國會發(fā)生以下事件：※每年有20,000次配錯藥事件※每年有超過15,000個嬰兒出生時會被拋落地上※每年平均有9小時沒有水、電、暖氣供應※每星期有500宗做錯手術事件1.2.由需求的觀點99雖然3合格率已達到99.73%的水平，但相信各位對以上品質要求並不滿意。所以有很多公司已要求「6」的品質管理，其合格品率為99.99966%。在3水準，由1000個零件組成的產品中，每15個產品中只有1個產品是好的。在6水準則1000個產品卻有996.6好的。傳統(tǒng)以『百分率』水準作為設計品質水準，如今變更為以『百萬分率(ppm)』作為衡量品質的水準。雖然3合格率已達到99.73%的水平，但相信各位對以上品質1001.3.從成本的觀點1.3.從成本的觀點1011.4.從時代趨勢的觀點(1)時間經濟附加價值開端成長成熟衰退典型的產品生命週期1.4.從時代趨勢的觀點(1)時間經濟附加價值開端成長成熟1021.4.從時代趨勢的觀點(2)經濟附加價值大型電腦主機迷你電腦與微電腦個人電腦掌上型電腦、電子書網路電視、GPS、行動電話電腦晶片的生命週期19471985199020001.4.從時代趨勢的觀點(2)經濟附加價值大型電腦主機迷你1031.4.從時代趨勢的觀點(3)全球化與經濟附加價值農業(yè)時代工業(yè)時代資訊時代?時代時間與主要的技術時代6000BC1760195020001.4.從時代趨勢的觀點(3)全球化與經濟附加價值農業(yè)時代1041.5.從品管大師的觀點J.M.Juran,1994年在美國品質管理學會年會會上說,“20世紀以生產力的世紀載入史冊,未來21世紀是品質的世紀”1.5.從品管大師的觀點1051.6.從策點管理的觀點欲建立及維持組織競爭優(yōu)勢，效率、創(chuàng)新、品質及顧客回應扮演著主要角色。較佳的效率競爭優(yōu)勢低成本差異化較佳的創(chuàng)新較佳的品質較佳的顧客回應1.6.從策點管理的觀點較佳的效率競爭優(yōu)勢低成本較佳的創(chuàng)新1061.7.從近代品質系統(tǒng)的觀點※ISO-9000Effectiveness:5※QS-9000Effectiveness:10※MalcolmBaldrigeGuidelineEffectiveness:25※EuropeanQualityAwardEffectiveness:301.7.從近代品質系統(tǒng)的觀點107※TQMEffectiveness：35※6

---TheLittleQEffectiveness:50※TheUltimate6

---TheBigQEffectiveness:90(上述品質系統(tǒng)均於80年代末期開展出)

※6&TheUltimate6將是一趨勢※TQM1082.6Overview6,thewaytocreateprofit.SigmaImproveCustomerSatisfaction&ProfitsincreaseDefectsDecreasecostsDecrease2.6OverviewSigmaCustomer109

6History(1)Somethingmustbewrong1975MotorolaTVbusinessfailedduetopoorprofitandsoldtoaJapaneseCompanyAssigncorporatequalityoffices1980Corporatemovement“GreatQualityAwakening”program1981FocusonQualityandTotalCustomerSatisfaction(TCS)6History(1)Somethingmus110

6History(2)EstablishMotorolaTraining&EducationCenter1981LaunchQualitySystemReview(QSR)program19825years,10×Qualityimprovementgoalset6theoryandconceptinitialized1986BillSmith&MikelHarrypresentedtheideatoCEOGalvin6History(2)EstablishMot111

6History(3)QuantitativeOperationQualityInitiative1987MotorolaTexasInstrumentBusiness-wideStrategicManagement1995Nokia,GE,AlliedSignal,Hitachi,Panasonic,Sony,Whirlpool,Honeywell,Boeing,Dupont6History(3)Quantitative1126InnovationModeling(3c-customer,competitor,company)(3e-excitation,employment,entertainment)(3p-product,process,person)InnovationBoxChangeofBusinessSituation(3C)Object(3P)Method(3E)Purpose.Profit.Skill-Up6InnovationModelingInnovat113TheManyAspectsof6

ToolSymbolMetricMethodValueGoalVisionBenchmarkPhilosophyTheManyAspectsof6ToolS1146asaMetric---Thehighlevelofsigma,thelowertheprobabilityofproducingadefect.Spec.LimitTargetLimitSomeChanceofDefect3Spec.LimitTarget