6SIGMA綠帶管理知識(shí)簡介

1、http:/6SIGMA綠帶管理知識(shí)簡介http:/ Content1. 6 Quality System -Why need2. 6 - Overview3. 6 - Methodology3.1. 6 - Define3.2. 6 - Measure3.3. 6 - Analysis3.4. 6 - Improve3.5. 6 - Control4. Conclusionhttp:/ 1. 6 Quality System -Why need 1.1. 從產(chǎn)品飽受競(jìng)爭威脅的觀點(diǎn)從產(chǎn)品飽受競(jìng)爭威脅的觀點(diǎn)以美國為例， 1975-1985年，日本挾其反向技術(shù)反向技術(shù)(Reversed Techn

2、ology)優(yōu)勢(shì)，將其高品質(zhì)的產(chǎn)品推向美洲大陸，使得美國一向以經(jīng)濟(jì)與技術(shù)領(lǐng)先的盟主地位發(fā)生動(dòng)搖。http:/在1970-1989年間美國的市場(chǎng)佔(zhàn)有率： 電視機(jī)從50%降至2% 收音機(jī)從50%降至2% 汽車從78%降至28% 影印機(jī)從90%降至20% 照相機(jī)從90%降至5% 鋼鐵從40%降至30% 其結(jié)果造成美國貿(mào)易赤字每月高達(dá)80-100億美元http:/美國與日本製程能力之比較 年代 美國 日本 製程能力 品質(zhì)水準(zhǔn) 製程能力 品質(zhì)水準(zhǔn) 1970 0.67 2 1.00 3 1980初 4 1980中 4 5 1980末 1990初 Motorola 6方法： 不合格率3.4PPM 相當(dāng)於 6

3、http:/ 美國企業(yè)競(jìng)爭力衰退的原因美國企業(yè)競(jìng)爭力衰退的原因 美國管理文化中含有不信任的氣氛不信任的氣氛，這種表現(xiàn)在嚴(yán)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。這種管理監(jiān)督的作風(fēng)，大體受到泰勒式科學(xué)管理泰勒式科學(xué)管理影響的結(jié)果。http:/ 日本經(jīng)營之神松下幸之助在一次對(duì)美國企業(yè)界人士發(fā)表演說，提到：你們的公司經(jīng)營是以泰勒法則為基礎(chǔ)，更糟的是頭腦皆已泰勒化，因此堅(jiān)信正確的管理，應(yīng)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)行者將觀管理是執(zhí)行者將觀念轉(zhuǎn)移到員工身上的一種藝術(shù)念轉(zhuǎn)移到員工身上的一種藝術(shù)。 因此美國開始檢討其品質(zhì)，各

4、種品質(zhì)系統(tǒng)亦相繼提出，其中包含6 品質(zhì)系統(tǒng)品質(zhì)系統(tǒng)。http:/ 1.2. 由需求的觀點(diǎn)由需求的觀點(diǎn) 在70年代，產(chǎn)品達(dá)到2 便達(dá)到標(biāo)準(zhǔn)。 在80年代，品質(zhì)要求已提升至3 ，但此標(biāo)準(zhǔn)美國會(huì)發(fā)生以下事件： 每年有每年有20,000次配錯(cuò)藥事件次配錯(cuò)藥事件 每年有超過每年有超過15,000個(gè)嬰兒出生時(shí)會(huì)個(gè)嬰兒出生時(shí)會(huì) 被拋落地上被拋落地上 每年平均有每年平均有9小時(shí)沒有水、電、暖小時(shí)沒有水、電、暖氣供應(yīng)氣供應(yīng) 每星期有每星期有500宗做錯(cuò)手術(shù)事件宗做錯(cuò)手術(shù)事件http:/ 雖然3 合格率已達(dá)到99.73%的水平，但相信各位對(duì)以上品質(zhì)要求並不滿意。所以有很多公司已要求6 的品質(zhì)管理，其合格品率為99

5、.99966%。 在3 水準(zhǔn)，由由1000個(gè)零件組成的產(chǎn)品中個(gè)零件組成的產(chǎn)品中，每每15個(gè)產(chǎn)品中只有個(gè)產(chǎn)品中只有1個(gè)產(chǎn)品是好的。在個(gè)產(chǎn)品是好的。在6 水準(zhǔn)則水準(zhǔn)則1000個(gè)產(chǎn)品卻有個(gè)產(chǎn)品卻有996.6好的。好的。 傳統(tǒng)以百分率百分率水準(zhǔn)作為設(shè)計(jì)品質(zhì)水準(zhǔn)，如今變更為以百萬分率百萬分率(ppm)作為衡量品質(zhì)的水準(zhǔn)。http:/ 1.3. 從成本的觀點(diǎn)從成本的觀點(diǎn)http:/ 1.4. 從時(shí)代趨勢(shì)的觀點(diǎn)從時(shí)代趨勢(shì)的觀點(diǎn)(1)時(shí)間經(jīng)濟(jì)附加價(jià)值開端成長成熟衰退典型的產(chǎn)品生命週期http:/ 1.4. 從時(shí)代趨勢(shì)的觀點(diǎn)從時(shí)代趨勢(shì)的觀點(diǎn)(2)經(jīng)濟(jì)附加價(jià)值大型電腦主機(jī)迷你電腦與微電腦個(gè)人電腦掌上型電腦、電子書

6、網(wǎng)路電視、GPS、行動(dòng)電話電腦晶片的生命週期1947 1985 1990 2000 http:/ 1.4. 從時(shí)代趨勢(shì)的觀點(diǎn)從時(shí)代趨勢(shì)的觀點(diǎn)(3)全球化與經(jīng)濟(jì)附加價(jià)值農(nóng)業(yè)時(shí)代工業(yè)時(shí)代資訊時(shí)代?時(shí)代時(shí)間與主要的技術(shù)時(shí)代6000BC 1760 1950 2000http:/ 1.5. 從品管大師的觀點(diǎn)從品管大師的觀點(diǎn) J. M. Juran, 1994年在美國品質(zhì)管理學(xué)會(huì)年會(huì)會(huì)上說,“20世紀(jì)以生產(chǎn)力的世紀(jì)載入世紀(jì)以生產(chǎn)力的世紀(jì)載入史冊(cè)史冊(cè), 未來未來21世紀(jì)是品質(zhì)的世紀(jì)世紀(jì)是品質(zhì)的世紀(jì)”http:/ 1.6. 從策點(diǎn)管理的觀點(diǎn)從策點(diǎn)管理的觀點(diǎn) 欲建立及維持組織競(jìng)爭優(yōu)勢(shì)，欲建立及維持組織競(jìng)爭優(yōu)勢(shì)，

7、效率效率、創(chuàng)創(chuàng)新新、品質(zhì)品質(zhì)及顧客回應(yīng)及顧客回應(yīng)扮演著主要角色。扮演著主要角色。較佳的效率較佳的效率競(jìng)爭優(yōu)勢(shì)競(jìng)爭優(yōu)勢(shì)低成本低成本差異化差異化較佳的創(chuàng)新較佳的創(chuàng)新較佳的品質(zhì)較佳的品質(zhì)較佳的顧客回應(yīng)較佳的顧客回應(yīng)http:/ 1.7. 從近代品質(zhì)系統(tǒng)的觀點(diǎn)從近代品質(zhì)系統(tǒng)的觀點(diǎn) ISO-9000 Effectiveness: 5 QS-9000 Effectiveness: 10 Malcolm Baldrige Guideline Effectiveness: 25 European Quality Award Effectiveness: 30http:/ TQM Effectiveness：3

8、5 6 -The Little Q Effectiveness: 50 The Ultimate 6 -The Big Q Effectiveness: 90 (上述品質(zhì)系統(tǒng)均於上述品質(zhì)系統(tǒng)均於80年代末期開展出年代末期開展出) 6 & The Ultimate 6 將是一趨勢(shì)將是一趨勢(shì)http:/ 2. 6 Overview 6 , the way to create profit.SigmaImproveCustomer Satisfaction&Profits increaseDefectsDecreasecostsDecreasehttp:/ 6 History (1)Somethin

9、g must be wrong1975Motorola TV business failed due to poor profit and sold to a Japanese CompanyAssign corporate quality offices1980Corporate movement “Great Quality Awakening” program1981Focus on Quality and Total Customer Satisfaction (TCS)http:/ 6 History (2)Establish Motorola Training & Educatio

10、n Center1981Launch Quality System Review (QSR) program19825 years, 10 Quality improvement goal set6 theory and concept initialized1986Bill Smith & Mikel Harry presented the idea to CEO Galvinhttp:/ 6 History (3)Quantitative Operation Quality Initiative1987Motorola Texas InstrumentBusiness-wide Strat

11、egic Management1995Nokia, GE, Allied Signal, Hitachi, Panasonic, Sony, Whirlpool, Honeywell, Boeing, Duponthttp:/ 6 Innovation Modeling(3c-customer, competitor, company)(3e-excitation, employment, entertainment)(3p-product, process, person)Innovation BoxChange of Business Situation (3C)Object (3P)Me

12、thod (3E)Purpose. Profit. Skill-Uphttp:/ The Many Aspects of 6 ToolSymbolMetricMethodValueGoalVisionBenchmarkPhilosophyhttp:/ 6 as a Metric-The high level of sigma, the lower the probability of producing a defect.Spec. LimitTarget LimitSome Chanceof Defect 3 Spec. LimitTarget LimitMuch Less Chanceof

13、 Defect 6 http:/ 6 as a Metric1691,5002308,537366,80746,210523363.4 (Shifted 1.5 )From 3 process to 6 process: about 20,000 times improvement6 DPMOhttp:/ 如何如何 6 導(dǎo)入導(dǎo)入 一般來說，從從3 到到4 階段階段，是企業(yè)可以自行改善的範(fàn)圍以自行改善的範(fàn)圍，但這樣的品質(zhì)標(biāo)準(zhǔn)並沒有辦法讓企業(yè)變得很有競(jìng)爭力； 從從4 到到5 階段階段，就必須找尋可以學(xué)習(xí)學(xué)習(xí)的標(biāo)竿企業(yè)的標(biāo)竿企業(yè)(Benchmarking)，當(dāng)成比較與學(xué)習(xí)的對(duì)象；而而5 到到6 階段

14、時(shí)，品階段時(shí)，品質(zhì)就已經(jīng)不是製造出來的，而是設(shè)計(jì)出質(zhì)就已經(jīng)不是製造出來的，而是設(shè)計(jì)出來的來的(DFSS-Design for Six Sigma)。http:/ 6 Activity 6 activity is to find out critical factors to quality (CTQ) at customers point of view and to reduce the defects less than 3.4 DPMO(PPM).LSLUSL ？ DefectM6 in R&D6 in MFG.CompanyCustomerCustomerCTQVOCVOB6 3.4

15、DPMONew Std.3 66800 DPMOPast Std.http:/ How Do We Improve Process Capability? Increase the tolerance Decrease the spread or variation of the process Shift the average by: Centering the average if the spec. has two limits Decrease or increase the average for spec. with one limithttp:/ The Cost Opport

16、unity1.5 2 3 4 5 6 051015202530About 15% of Sales, Cost Opportunity on 3 CompanyCost of Failure(% of Sales)http:/ The Cost of Poor Quality (COPQ)Long Cycle Times, More Setups, Expediting Costs, Lost Sales, Engineering Change order, Overtime, Late delivery, Lost Opportunity,Lost Customer Loyalty, Exc

17、ess InventoryRejects, Warranty, Inspection, Scrap, Reworkhttp:/ Who is Implementing 6 Motorola1987 Texas Instrument1988 ABB (Asea Brown Boveri)1993 Allied Signal1995 General Electric1995 Kodak1995 Siemens, Nokia, Sony1997 http:/ 6 Benefits(Case 1) -Motorola over 12 yrs Increase productivity an avera

18、ge of 12.3% per year Reduced the cost of poor quality by more than 84% Eliminated 99.7% of in process defects Save more than $11billion in manufacturing cost Realized an average annual compounded growth rate of 17% in earningshttp:/ 6 Benefits(Case 2)General Electric(Million Dollars) Cost Profit1996

19、240, 2001997400, 7001998450, 12001999520, 25202000600, 3000http:/ 6 Benefits( Case 3)Texas InstrumentsBefore (1988)After (1993)COPQ30%7%DPMO1040009000Scrap$3M$0.3MYield84.5%98.9%Cycle Time (Week)114Inventory Cost$3.9M$1.1Mhttp:/ The Focus of 6 If we are perfectly control X, should we constantly test

20、 and inspect Y?Y = f (X)YX1X2X3X4X1、X2、XnIndependentInput-ProcessCauseProblemControlYDependentOutputEffectSymptomMonitorhttp:/ 6 is Applying Overall Business SystemR&D6 Trans-actionMfgDesign SSManufacturing SSTransactional SShttp:/ 6 MethodologyMeasureAnalyzeDesignVerifyMeasureAnalyzeImproveControlD

21、efineTranslateYNNew Process/ Product ?Achievable Goal ?NYhttp:/ What is different with 6 Traditional Quality6 Central OrganizedNot Formal Structure for Tool ApplicationLack of Support in using ToolsLack of Structured TrainingInspect Quality in (Focus on “Y”)Black Belt Report Directly Into FunctionSt

22、ructured Use of Statistical Tools to Aid Problem SolvingData-Based DecisionsStructured Training in Applied StatisticsControl Process Input (Xs)http:/ 6 Organization6 OrganizationExecutive Project Selection and Support6 Technical Lead6 Coach Project &BB6 Project Team Lead6 Project Team MemberAll staf

23、fs (6 overview) ChampionMaster Black Belt(MBB)Black Belt (BB)Green Belt(GB)White Belt(WB)http:/ 3. 6 Methodology 5 Stage of DMADV and New Tools- Design For Six Sigma (DFSS)- Black Belt 5 Stage of DMAIC methodology - Green Belthttp:/3.1. DMAIC Methodology5 Stages DMAIC Methodology and Statistical Too

24、ls Phase: Define Steps; Work Breakdown & Tool D1 Validate Business Opportunity; 3C Analysis, Identify Customer, Cost/Effect Analysis D2 Define Customer Requirement; VOC, VOB, QFD D3 Project Planning; Project Team, Project Charter, COPQhttp:/ Define Steps (1)VisionBusiness StrategyBig Y (CTQ)Small y

25、-Project(Goal, Scope, Performance Index)http:/ Define Steps (2-1)Vision:最具競(jìng)爭力的企業(yè)最具競(jìng)爭力的企業(yè)(GE)Strategy-The Three Circle (GE)核心事業(yè)核心事業(yè)照明照明大型家電大型家電馬達(dá)馬達(dá)高科技事業(yè)高科技事業(yè)醫(yī)療系統(tǒng)醫(yī)療系統(tǒng)工業(yè)電子工業(yè)電子航太航太服務(wù)事業(yè)服務(wù)事業(yè)信用公司信用公司資訊服務(wù)資訊服務(wù)核能服務(wù)核能服務(wù)http:/ Define Steps (2-2) -The Three Circle (GE) Jack Welch (GES CEO)畫三個(gè)圓圈：核畫三個(gè)圓圈：核心、高科技與服務(wù)。

26、心、高科技與服務(wù)。GE公司未來都要在公司未來都要在其中一個(gè)圓圈內(nèi)。公司任何人不再任一其中一個(gè)圓圈內(nèi)。公司任何人不再任一個(gè)圓圈內(nèi)者，未來將會(huì)被解雇。圓圈外個(gè)圓圈內(nèi)者，未來將會(huì)被解雇。圓圈外的的事業(yè)將被整頓、關(guān)閉或出售。三個(gè)的的事業(yè)將被整頓、關(guān)閉或出售。三個(gè)圓圈的策略讓圓圈的策略讓GE公司找到焦點(diǎn)，不再是公司找到焦點(diǎn)，不再是集團(tuán)集團(tuán)(似乎什麼都做似乎什麼都做)。http:/ Define Steps (3)Select TargetProduct/ServiceAnalysis BusinessProcessAnalysis Core ProcessIdentifyCustomerListen t

27、o VOCSpecify CustomerRequirementSelect CTQ(Big Y)Specify CTQ(Small Y)Evaluate PotentialProjectsSelect Project & Build Effective Teamhttp:/ Define Steps (4) CTQ - Critical to Quality只要是顧客要求的，就是關(guān)鍵品質(zhì)只要是顧客要求的，就是關(guān)鍵品質(zhì)(CTQ)。亦稱為重要成果、特殊限制或流程的Y變數(shù)。CTQ係任係任何會(huì)直接影響顧客對(duì)產(chǎn)品何會(huì)直接影響顧客對(duì)產(chǎn)品/服務(wù)品質(zhì)觀感服務(wù)品質(zhì)觀感之因素之因素。顧客顧客在乎在乎的事，就是企

28、業(yè)的事，就是企業(yè)或組織在乎的事或組織在乎的事。新點(diǎn)子新點(diǎn)子要採納要採納外部觀點(diǎn)外部觀點(diǎn)(Outside-in Perspective)，即從顧客的眼光來看企業(yè)或組織的一切。即從顧客的眼光來看企業(yè)或組織的一切。http:/3.2. DMAIC Methodology5 Stages DMAIC Methodology and Statistical Tools Phase: Measure Steps; Work Breakdown & Tool M1 Specify Project; CTQ Tree, Process Map, Performance Index M2 Assess Meas

29、urement System; Measurement System Analysis, Gage R&R M3 Identify Sigma Level; Graph Analysis, Capability Analysis, Confidence Intervalhttp:/ M1 Specify Project M1 step covers the followings:CTQ BreakdownDefine ScopeProcess Mapping Pareto AnalysisDefine performance index and specificationsThe succes

30、s of any 6 activities relies mostly on the CTQ definition and selectionhttp:/ Define Performance IndexCustomer RequirementsInternal ProcessImprovedProcessCustomer SatisfactionIndex of current levelIndex of improved levelPerform project for improvement ClearQuantifiableSimplehttp:/ Example of Perform

31、ance IndexYieldCycle time Defect rateMachine failure rateCustomer stand-by hoursNumber of invoice errorsElapse Time from loan application to money transfer to the customer accountHours taken from receiving order to delivery to the customerhttp:/M2 Measurement System Analysis M2 step covers the follo

32、wings:Statistic fundamentalsSampling planData collection planVariation of measurement systemGage R&R studyImprovement of measurement systemWithout accurate measure, cant identify any symptom of problemhttp:/ 6 MetricsData TypeStatistic measurement dataComparison to specificationZ-valueDiscrete DataC

33、ontinuousDataDefect unit opportunity (DPMO)Average, Standard variation, ShapeSPECZ-valueSigma calculationhttp:/ Definition of DPMO terminology DPU: Defect per Unit (Defect: Anything that results in customer dissatisfaction; Anything that results in nonconformance) DPO: Defect per Opportunity DPMO: D

34、efect per Million Opportunitieshttp:/ Example of Sigma calculation: Discrete Data Case (1): Over the last several years, you have collected data on trips to the airport. Of the 100 trips sampled, you have missed only 5 flights. If you indicate this situation as Sigma level, what will it be? Sol. DPU

35、 = DPO =5/100 = 0.05 1-0.05 = 0.95, Z = 1.65 level = Z + 1.5 =1.65 + 1.5 = 3.15http:/Case (2). An accounting term conducted an internal audit for the financial report of year 2001 and they observed 25 nonconforming record. There were 2 steps of probable incorrect entry identified in the process and

36、the total observation was 25,000 records. What is DPO? What is DPMO? What is Sigma level of financial reporting process? Sol. DPU = 25/25000, DPO = 25/(2*25000) =0.002, DPMO = 0.002*106 =20001- 0.002 = 0.998,Z = 2.878 level = Z + 1.5 = 2.878 + 1.5 = 4.378http:/ Example of Sigma calculation: Continuo

37、us Data(1)1 MeanMedianMode 70 80 90 100 110 120 130+ - Normal Distribution with mean = 100 and std. = 10http:/ Continuous Data(2)(1) Mathematical Model；(2) Continuous；(3) Smooth；(4) Symmetrical; (5)Tail asymptotic to X-axis；(6) Bell shaped；(7) Mean = Median = Mode (8) Total area under curve = 1(1) T

38、he normal distribution has the following properties.68.27% of the data fall within 1 95.45% of the data fall within 2 99.73% of the data fall within 3 (2) In order to assess the quality of the process, we must compare the process characteristics (via the location, spread and shape) to the specificat

39、ion limits and targeted value.http:/ Continuous Data(3)Standardization of Normal Distribution The Sigma of a Process is the number of standard deviation between the mean and the Specification Limits.1 Z = (X- )/ No. of standard deviationUSLSigma of the process Z = 4.5http:/ Measurement System and Me

40、asurement Error(1) Measurement system is viewing as a process. Sources of variation: 5M1E Validate possible sources of variation in the measurement processP ersonnelM ac hinesM aterialsM ethodsM easurem entsE nvironm entC a u s e -a n d -E ffe c t D ia g ra mVariation in measurementSystemhttp:/ Meas

41、urement System and Measurement Error(2)Averagem (Total) = m (Product) + m(Measurement)Variabilitys2 (Total) = s2 (Product) + s2(Measurement)Deflection of measurement system(To be decided by calibration)Variation of measurement system(To be decided by R&R assessment)http:/ Process VarianceObserved Pr

42、ocess VariationActual Process VariationMeasurement VariationLong term process variationShort term process variationVariation within a sampleVariation due to operatorsVariation due to gageReproducibilityLinearityStabilityRepeatabilityAccuracyhttp:/ Gage R&R Analysis Types of Gage R&R Analysis Methodo

43、logyX bar- R MethodANOVATypes of Variation estimation by the Gage R&REquipment Variation: EVAppraiser (Operator): AVhttp:/ GR&R Decision and Improvement Direction Gage R&R Decision Criteria %GR&R 10% (Good measurement system) 10% %GR&R 30% (May be used) %GR&R 30% (Can not used) Gage improvement dire

44、ctionFor repeatability error reproducibility error (Need to taken an action to operator)For reproducibility error repeatability error (Need to taken an action to gage)http:/ M3 Identify Sigma Level M3 step covers the followings:Data stratificationGraphical cause eliminationUnderstand product capabil

45、ity and product performanceCalculate current sigma levelhttp:/ Graphical AnalysisRun Chart Scatter Diagram Box PlotHistogram Changes to the process may be more easily recognized graphically than tabularly. http:/ Understanding Variation What is Variation? Different outcome of a process or result of

46、a product or service Measurement index scattered from center value Variation will be appeared in every process and the target of improvement is to reduce its variations Why a variation might be occurred? By a common cause By a special cause 5M1EWhat impacts will be happened if a variation become big

47、? Can not predict/ forecast outcomes Occur re-inspection and/ or rework Delay schedule Increase customer dissatisfactionhttp:/ Short-Term/Long-Term RelationshipsInherent Capability of the Process-Short Term CapabilitySustained Performance of the Process-Long Term PerformanceOver time, a “typical” pr

48、ocess will shift and drift by approximately 1.5 http:/ Product Capability(Cp) Cp = |USL-LSL| / 6st ； Zst = 3 Cp Capability Index(Cpk) Cpu = min (Cpu, Cpl)，Where Cpu = (USL-)/3st；Cpl = (-LSL)/3st；Zst = 3 Cpk Performance Index(Cpk) Cpk = min (Cpu, Cpl)，Where Cpu = (USL-)/3lt；Cpl = (-LSL)/3lt；Zst = 3 C

49、pkhttp:/3.3. DMAIC Methodology5 Stages DMAIC Methodology and Statistical Tools Phase: Analysis Steps; Work Breakdown & Tool A1 Set up Improvement Goal; Benchmarking, Entitlement, KANO A2 Identify Potential Causes; Pareto, Brainstorming, Cause & Effect Diagram, Logic Tree A3 Verify Potential Causes;

50、Regression, Hypothesis Test, ANVOA, Multi-Vari Analysishttp:/ A1 Set Up Improvement GoalIn M3 the current product capabilities were defined.Step A1 provides the tools to define performance objectives. The purpose is to set objective to establish a balance between improving customer satisfaction and

51、available resources.http:/ A1 step covers following topics:Establish performance goalsAssess short-term and long-term sigma in terms of measurement benefits (decrease in COPQ, increase in revenue)Determine improvement direction by the gap analysis between goal and current levelhttp:/ Paths to Perfor

52、mance GoalsPerformanceGoalsProcess Improvement(Entitlement)Process Change(Breakthrough)BaselineProcess Measurementhttp:/ Benchmarking - Why Benchmark (1)?InnovationContinuous ImprovementBreakthrough ImprovementContinuous ImprovementTimePerformancehttp:/ Benchmarking - Why Benchmark (2)?Develop and I

53、mprovement Strategic GoalsEstablish actionable objectivesProvide sense of urgencyEncourage breakthrough thinkingCreate a better understanding of your industryhttp:/ A2 Identify Potential Causes Finding main independent variables, and making the listY = f (X) ObjectMake list of potential independent

54、variables for changing & improving Y value methods Brainstorming Cause & Effect Diagram Multi-voting Logic Treehttp:/ Cause & Effect DiagramPersonnelMachinesMaterialsMethodsMeasurementsEnvironmentCause-and-Effect DiagramVariation in measurementSystemhttp:/ A3 Verify Potential Causes OverviewIn step

55、A3, setting a possible priority as to their importance to the potential variables (Xs) identified in step A2 PurposeGeneral a list of important factors (vital few) from the potential variables.Y = f (x1,x2,x3) (x4,x5,.,xn)Vital FewTrivial Manyhttp:/ Pareto AnalysisRottenBruisedUndersizedOthers235100

56、 87 1953.322.719.7 4.3 53.3 76.0 95.7100.0050100150200250300350400450020406080100DefectCountPercentCum %PercentCountPareto Chart for CATEGORIEShttp:/ Correlation AnalysisA statistical analysis to investigate / measurement of association between two variables (X, Y) is called analysis.Correlation tel

57、ls you the trend of Y when X value increase/decrease.Correlation Analysis using Scatter Diagram Analysis.Correlation coefficient indicates closeness of a relationship between X and Y.http:/ Regression Analysis (1)Y = f (x1, x2, x3,)Status and characteristics of a processModelingMathematical equation

58、XYhttp:/ Regression Analysis (2) Types of Regression ModelSimple linear regressionMultiple linear regressionNon-linear regressionY = f(X)Y: dependent variableX: independent variablehttp:/ Types of Hypothesis TestTypes of DataDiscrete DataContinuous DataMean Testt TestANOVAVariance TestF TestChi- Squ

59、arehttp:/3.4. DMAIC Methodology5 Stages DMAIC Methodology and Statistical Tools Phase: Improve Steps; Work Breakdown & Tool I1 Identify Vital Few; Screening DOE, Streamlining, Force Field Analysis I2 Formulate Optimum Model; Optimizing DOE(RSM), Solution Selection Matrix I3 Set Up Operating Window; Operating Window, Responsibility Matrixhttp:/ I1 Identify Vital FewIn analysis step, potential Xs were identified with a possible prioritization as to their importance in