MarketingLeadershipPresentedby:EdenH.C.ChenSixSigmaAcademyCTIGroup,USAR&DSelf-SupportsProcessIndependentlyVariationReductionWastesReductionSixSigmaStrategiesforExcellenceSixSigma是一種經營管理理念,也是一套有系統的管理手法,目的是在定義的目標之下,追求最小的變異,獲得最大的獲利水準。CTI集團陳憲章博士六標準差為一全面且具彈性的系統,可用於獲取、維持和擴大企業成功。六標準差的趨動要素,在於洞悉顧客之需,嚴格使用事實、資料、統計分析,以及全力關注業務流程的管理、改善和創新。Thesixsigmaway-SixSigmaisamanagementphilosophywhichsearchfortheperfectwithoutanysignificantvariationsbasedonthepre-setperformancerequirements.-Todefinetherequirements,findingouttheundesirableoutcomes(y)andassociatingitwiththepossibleinputsistheconceptstoavoid,ifnot100%sure,eliminatethecausesofvariancesinintentoutputs.Bydoingsowecanminimizethewaste–lean.-Usebesttoolstoachievethegoalsofavoidingcause,detectfailure,&compensatetheeffectsSixSigmaBasicConceptTakeawildguessonwhatisthenormalaveragemistakesbyanygivenprocessedoperation?Semi-Professionalsare7%(stddev.Is6%)Professionalsare3%(std.dev.Is2%)AbsoluteProfessionalswithoutcheck/balanceare11%(stddev.Is11%)ForWhiteCollarworksForBlueCollarworksSkilledOperatorsare1%(stddev.Is.8%)withoutErrorProofingOptionsUnskilledOperatorsare7%(stddev.Is6.8%)withoutErrorProofingOptionsForInstrumentControlOperationsBasedonaccuracyandsensitivityoftheinstrumentitcanbeanythingfrom0.05%to3.0%Likely,formostindustryinstrumentitisaround0.5%SupplyInputProcessOutputCustomer供應者品質輸入流程功能產出要求客戶滿意i.e.KPIVi.e.VOC,CTCi.e.可歸因的外在變異(Assignablecause)i.e.一般內在因素變異&KCCetc.i.e.CTQSpec.KPC,&etc.SIPOC的觀念InputProcessOutput首先看一看輸入與輸出的關係OutputInputYX如果結果只被一個輸入因素影響,則1:1,suchasY=X.套模零件尺寸模具尺寸YX範例:在無外部噪音與內部變因的完美狀況50%50%00.10.20.30.40.50.60.7HeadTail擲一塊硬幣的機率Distribution(分配)andProbability(機率)Frequency123465432150.0270.0560.0630.1110.1390.1670.1390.1110.0630.0560.027%23456789101112兩粒骰子的機率和出現次數頻率圖0.180.160.140.120.100.080.060.040.020.機率出現次數12345654321mean-1s-3s-2s2s3s1s68%95%99.7%中心線與標準差OutputInputYX然而在很多條件情況下,結果是一個分佈Undesirable非期待流程/製程InputDesirable期待OutputControlControlNoises噪音Man人Machine機Material料Method法Environment環境考慮噪因(noise)的情況OutputInputYXNoise流程/製程Man人Machine機Material料Method法Environment環境Input好的產品有缺點的產品OutputProcessControl&SPCQualityinspection&SQC品質檢驗&統計品管洪水,地震,罷工等在實際的環境下ControlControlControlControlMMMMEMMMMEMMMMEMMMMEY#1#2#3#lastOutput=InputY1=X1Y1=F(X1)Y1=F(Xi)Y1=ΣF(Xi)Y1=πF(Xi)Yj=πF(Xi)•有多少輸出變數?•有多少輸入變數?•那幾個是重要輸出變數?•那幾個是重要輸入變數?•將重要輸出變數與重要輸入變數連結現實環境下的流程模式OutputYY=X1+X2+X3+…Xn結果就更複雜了…當不是只有一個輸入變數時InputX1InputX2InputX3InputXn工程下規集中點工程上規完美狀態.工程規格與實際產出現象工程下規工程上規Target幾乎完美接近完美大多數狀況Spec.Limit+/-1+/-2+/-3+/-4+/-5+/-6Percent(%)68.2795.4599.7399.993799.99994399.9999998PossibleDefectiveppm3173004500270083.57.002完美狀況理論值Spec.Limit+/-1+/-2+/-3+/-4+/-5+/-6Percent(%)30.2369.1393.3299.379099.9767099.999660PossibleDefectiveppm6977003087006681062102333.4大多數狀況實際值現實的世界Spec.Limit+/-1+/-2+/-3+/-4+/-5+/-6Percent(%)30.2369.1393.3299.379099.9767099.999660PossibleDefectiveppm6977003087006680062102333.4分佈範圍與集中程度失敗成本4SigmaCondition--6,000PPM(DefectivePartsPerMillion)6SigmaCondition--3-4PPM(DefectivePartsPerMillion)1%25%SalesSalesSource:Monro&Assoc.Inc.企業經營成功的指標投資報酬未來成長的體質客戶滿意度保固成本回收(Warrantypay-off(DPTP)生產力第一時間點的良率(1stTimeyieldandQuality)產業排比(IndustrialRating)品質指標允收水準(AQL/AOQL)SPCwithestimateddefectiverate&trend製程(流程)能力指標ProcessCapabilityIndex1%不良的品質水準代表什麼意義?每小時有20﹐000郵件丟失每天有15分鐘飲用不淨水每周有5﹐000手術錯誤主要機場每天至少有4次事故每年有200﹐000藥物處方錯誤每月有七天停電28零缺點-“Almost(幾乎)”完美的流程設計當你買一瓶藥,你是否期待每一顆都是好的?當你搭飛機,你是否期待每一次起飛與降落都成功?是否零缺點的觀念可以應用到你的流程上甚至產品上?29SixSigma的目標:面對全球市場與競爭力,消除變異追求最大的獲利目標不斷提高客戶滿意度持續測量提高制程能力降低不良品增加獲利持續改善導致經營成功30SixSigma成功策略過程內的數據資料引導決定Y=SF(Xi)流程內的各個輸出結果=各種機會最終產品產出將F(Xi)作最大的發揮,則就可控制Y.31機會?即刻–短期因應(Reactive)解決現在的問題確實指出高風險的問題所在(suchascost,reworks,scraps,&ect.)預見–長期預應(Proactive)防止潛在的弱點造成問題指出潛在的機會32MeasureAnalyzeImproveControlCharacterizationOptimizationwithinthesystemSixSigma應用戰術-MAIC33Define/定義掌握特性(Characterization)在體制內追求完美(Optimization)Control/管制Improve/改進Analyze/分析Measure/衡量SixSigma應用戰術-DMAIC34Define/定義Control/管制DESIGN/重新設計Analyze/分析Measure/衡量掌握特性(Characterization)在體制外追求完美(Optimization)SixSigma應用戰術-DMADV35Define定義Validate/驗證Incorporate/組成團隊執行之Evaluate/評估Optimize/自我調適掌握特性(Characterization)自我調適(Optimize)驗證執行SixSigma應用戰術-DEOVI36MAIC現狀點的改善6sMeasurement测量Control控制Improvement改进Analysis分析37DMAIC是在現有體制之下變革式的改善6sMeasurementControlImprovementAnalysisDefine38DMADV是打破體制的突破式變革6sMeasurementValidationDesignAnalysisDefine39DEOVI是在最早的時間點上將事情有效率的完成6sEvaluateIncorporateValidateOptimizeDefine40Define,Measure,Analyze,Improve,Control1SelectOutputCTQorCTC2Defineperformancestandardsforresponsevariable3Identifymeasurementsystemforinputandresponsevariables4Establishprocesscapabilityofresponsevariable1Defineimprovementobjectives2Identifyallsourcesofvariationiny(inputs=xi)3Screensourcesforpotentialcausesofvariationiny&identifyvitalfewx–KPIV1Determinemathematicalrelationshipbetweenvitalfewxi2Establishoperatingtolerancesforvitalfewxi1Validatemeasurementsystemforxi2Determineabilitytocontrolvitalfewxi3EstablishprocesscontrolplanforvitalfewxiStepFocusyyx&yyx1,x2,...xnx1,x2,...xnVitalFewxiVitalFewxiVitalFewxiPhaseMeasureAnalyzeImproveCont