统计过程控制简介

连接控制图方法与流程改进方法LinkControlChartmethodstotheProcessImprovementMethodologyTM讨论不同类型的变差Discussdifferenttypesofvariation介绍各种类型的控制图IntroducevarioustypesofControlCharts讨论如何解释控制图DiscusstheinterpretationofControlCharts目的ObjectivesProjectTrackerRevision5ProjectStartDate:21/01/2004ProjectTypeDataCollectionPlanforYDevelopProjectScheduleLaunchCharterImprovementDefineIdentify'YanddraftcharterCharterApprovedProcessMapC&EMatrixorFTA30DayMBBReviewPreliminaryFMEAorFTAMSADataCollectionPlanforKeyXsMBBReviewMeasure21/01/200404/02/200411/02/200425/02/200409/03/200409/03/200409/03/2004InitialCapabilityStudyMulti-VariMBBReviewContractApprovalAnalyze22/03/200415/04/200415/04/200415/04/200415/04/2004SingleorMultiFactorTests(DOE)MBBReviewImprove31/05/200431/05/200431/05/2004ControlPlanFinalCapabilityStudyFinalFMEARevisedRPNMBBReviewFinalProjectPresentation&ReportProjectSign-off&ClosureControl21/06/200429/06/200429/06/200405/07/200409/07/200409/07/200419/07/2004(Useasrequired)VOC/VOBRequirementsAnalysisProcessRedesignSolutionDesignRedesignEnterStartDateHereImprovementsDefinedProjectHopperandSponsorRoleReview6SigmaDatabasesforlikeprojectsImprovementsImplementedHandOffTraining/ProcessOwnerSign-offRedesignmapiscalculatedseparatelywithnodependenciesonotherblockdatesActualcompletiondatePlannedcompletiondateLegend2/1/020022/3/02Completioncheck我们是否应该采取行动？Shouldwetakeaction?每天我们都被数据淹没,而且不得不作出决定Everydaywearefloodedbydataandweareforcedtomakedecisions工厂产量下降４％Plant’sOutputDecreasesBy4%美国贸易赤字增加４００亿USTradeDeficitRisesBy$40Billion某公司获利比上季度降低２．４亿CompanyX’sEarningsAreOff$240MillionFromPreviousQuarter我们需要解释数据的方法WeNeedWaystoInterpretData今天采集什么样的数据？WhatTypeOfDataIsCollectedToday?制造业Manufacturing:_____________________________________非制造业Non-Manufacturing___________________________________如何分析数据？HowIsItAnalyzed?制造业Manufacturing:_____________________________________非制造业Non-Manufacturing___________________________________得知数据好坏后该当如何？WhatHappensIfItIsBad/Good?制造业Manufacturing:_____________________________________非制造业Non-Manufacturing___________________________________客户需求下限Lower“Customer”Requirement这一方法THISMETHOD告诉你关于客户的需求Tellsyouwhereyouareinregardstocustomer’sneeds不告诉你怎么满足用户需求及下一步怎么办ItwillNOTtellyouhowyougotthereorwhattodonext客户需求上限Upper“Customer”Requirement我们管理数据的方式－过去(历史来讲)的方式TheWayWeManageData–Historically不用管它,不会坏的LeaveItAlone.ItAin’tBroke痛苦,受累Pain&Suffering痛苦,受累Pain&Suffering这一方法导致何种管理行为？Thismethodcauseswhattypeofmanagementbehavior?客户需求下限Lower“Customer”Requirement客户需求上限Upper“Customer”Requirement我们管理数据的方式－历史来讲的方式TheWayWeManageData–Historically不用管它,不会坏的LeaveItAlone.ItAin’tBroke痛苦,受累Pain&Suffering痛苦,受累Pain&Suffering23ScrapLevel(%)废品率11996CelebrationTime•工厂废品率为年度最低的２％Thefactoryscraplevelisatayearlowof2%•经理给工厂颁奖Managerpresentsanawardtotheplant•在餐厅进行庆祝：每人都可分享免费皮萨饼和饮料Ceremonyinthecafeteria:pizzaandrefreshmentsforall!•“每人都应为他们的成就骄傲．”“Everyoneshouldbeproudofwhatthey’veaccomplished”.DerivedfromUnderstandingVariation:TheKeyToManagingChaos,DonaldJ.Wheeler,SPCPress.1993.１９９６年４月APRIL1996JFMA2311996经理希望能将发出去的奖收回来Managerwantstotakebackaward•废品率连续三个月持续增长Threeconsecutivemonthsofscrapincreases.•经理希望能将发出去的奖收回来Managerwisheshecouldtakebacktheaward•经理考虑要采取行动了ManageristhinkingabouttakingactionScrapLevel(%)废品率１９９６年６月JUNE1996DerivedfromUnderstandingVariation:TheKeyToManagingChaos,DonaldJ.Wheeler,SPCPress.1993.JFMAMJ2311996Nomore“NiceGuy”不再充好人了•废品率上升到２．６％Scraprisestoavalueof2.6%•经理决定采取行动Managerdecidestotakeaction•召开一个“特别会议”来寻求一个永久性的解决方案A“specialmeeting”iscalledtosolvethisproblemonceandforall.•经理在长篇大论次品率多么重要后离开了．雇员们不知道该干什么．另外，他们有其他更重要的评估标准．于是，他们什么也没做．Afterasoundlectureontheimportanceofscrap,themanagerleaves.Employeesaren’tsurewhattodo.Besides,theyhaveothermetricswhichhavemoreimportance.Sotheydonothing.ScrapLevel(%)废品率１９９６年１１月NOVEMBER1996DerivedfromUnderstandingVariation:TheKeyToManagingChaos,DonaldJ.Wheeler,SPCPress.1993.JFMAMJJASON•经理看到从去年开始废品率持续下降Managerhasseenreducedscraplevelssincetheendoflastyear•教训：“严格的管理会出成效！”TheLearning:“Atoughmanagementstylegetsresults!”Managerconcludes:“ToughLoveMakesThingsHappen”23119961997ScrapLevel(%)废品率1997年6月JUNE1997DerivedfromUnderstandingVariation:TheKeyToManagingChaos,DonaldJ.Wheeler,SPCPress.1993.JFMAMJJASONDJFMAMJDerivedfromUnderstandingVariation:TheKeyToManagingChaos,DonaldJ.Wheeler,SPCPress.1993.将数据置于统计流程控制图中PuttingTheDataInASPCChart23119961997ScrapLevel(%)废品率JFMAMJJASONDJFMAMJUCLLCL统计流程控制图显示不同的解释,可为什么呢?SPCTellsADifferentStory.ButWhy?23119961997ScrapLevel(%)废品率JFMAMJJASONDJFMAMJUCLLCL“人们已知的最佳方式之一是如不能使用控制图分析数据会：增加成本,浪费的努力和降低士气;-DonaldJ.Wheeler博士“Failuretousecontrolchartstoanalyzedataisoneofthebestwaysknowntomankindto:increasecostswasteeffortandlowermorale.”-Dr.DonaldJ.Wheeler统计流程控制图显示不同的解释,可为什么呢?SPCTellsADifferentStory,ButWhy?S=统计技术:检查偏差StatisticaltechniquesusedtoexamineprocessvariationC=控制过程通过积极管理ControllingtheprocessthroughactivemanagementP=过程,任何过程Process,ANYProcess现在我们管理数据的方法-SPCTheWayWeManageData-TodaySPC•显示过程偏差随时间变化的图形控制图方法ControlChartsMethod它从哪里来的?WhereDidItComeFrom?•19世纪20年代-西部电器的WalterShewhart博士:1920’s-WesternElectric/Dr.WalterShewh