6SIGMA绿带介绍(PPT 91页).ppt

下载

下载

48积分

60积分

下载

60积分

VIP价：48积分

6  綠 帶 介 紹 Introduction to 6  Green Belt 1 • Content 1. 6  Quality System ---Why need 2. 6  --- Overview 3. 6  --- Methodology . 6  --- Define . 6  --- Measure . 6  --- Analysis . 6  --- Improve . 6  --- Control 4. Conclusion 2 • 1. 6  Quality System ---Why need • . 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』(Reversed Technology)優勢， 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 發生動搖。 3 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 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 • 1970 2 • 3 • 1980初 4 • 1980中 4 ~ 5 • 1980末 • ~ 1990初 Motorola • 6方法： • 不合格率 • 相當於 6 5 • 美國企業競爭力衰退的原因 • 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。『這種管理監督的作風，大體受到 泰勒式科學管理影響的結果』。 6 • 日本經營之神松下幸之助在一次對美國 企業界人士發表演說，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 統亦相繼提出，其中包含6 品質系統。 7 • . 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 8 • 雖然3合格率已達到%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%。 • 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個產品卻有好的。 • 傳統以『 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。 9 • . 從成本的觀點 10 • . 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 11 • . 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 1947 1985 1990 2000 12 • . 從時代趨勢的觀點(3) 全 球 化 與 經 濟 附 加 價 值 農業時代 工業時代 資訊時代 ?時代 時間與主要的技術時代 6000BC 1760 1950 2000 13 • . 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” 14 • . 從策點管理的觀點 • 欲建立及維持組織競爭優勢，效率、創 新、品質及顧客回應扮演著主要角色。 較佳的效率 競爭優勢 低成本 差異化 較佳的創新 較佳的品質 較佳的顧客回應 15 • . 從近代品質系統的觀點 •※ ISO-9000 • Effectiveness: 5 •※ QS-9000 • Effectiveness: 10 •※ Malcolm Baldrige Guideline • Effectiveness: 25 •※ European Quality Award • Effectiveness: 30 16 •※ TQM • Effectiveness：35 •※ 6  ---The Little Q • Effectiveness: 50 •※ The Ultimate 6  ---The Big Q • Effectiveness: 90 (上述品質系統均於80年代末期開展出) ※ 6  & The Ultimate 6 將是一趨勢 17 • 2. 6  Overview • 6 , the way to create profit. Sigma Improve Customer Satisfaction & Profits increase Defects Decrease costs Decrease 18 • 6  History (1) Something must be wrong 1975 Motorola TV business failed due to poor profit and sold to a Japanese Company Assign corporate quality offices 1980 Corporate movement “Great Quality Awakening” program1981 Focus on Quality and Total Customer Satisfaction (TCS) 19 • 6  History (2) Establish Motorola Training & Education Center 1981 Launch Quality System Review (QSR) program 1982 5 years, 10×Quality improvement goal set 6 theory and concept initialized 1986 Bill Smith & Mikel Harry presented the idea to CEO Galvin 20 • 6  History (3) Quantitative Operation Quality Initiative 1987 Motorola Texas Instrument Business-wide Strategic Management 1995 Nokia, GE, Allied Signal, Hitachi, Panasonic, Sony, Whirlpool, Honeywell, Boeing, Dupont 21 • 6  Innovation Modeling • (3c-customer, competitor, company) • (3e-excitation, employment, entertainment) • (3p-product, process, person) Innovation Box Change of Business Situation (3C) Object (3P) Method (3E) Purpose . Profit . Skill-Up 22 • The Many Aspects of 6  Tool Symbol Metric Method Value Goal Vision Benchmark Philosophy 23 • 6  as a Metric---The high level of sigma, the lower the probability of producing a defect. Spec. LimitTarget Limit Some Chance of Defect 3  Spec. LimitTarget Limit Much Less Chance of Defect 6  24 • 6  as a Metric 1 691,500 2 308,537 3 66,807 4 6,210 5 233 6 (Shifted ) From 3 process to 6 process: about 20,000 times improvement 6 DPMO 25 • 如何 6  導入 • 一般來說，從3  到4  階段，是企業可 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(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---Design for Six Sigma)。 26 • 6  Activity • 6 activity is to find out critical factors to quality (CTQ) at customer’s point of view and to reduce the defects less than DPMO(PPM). LSL USL ？ Defect M 6 in R&D 6 in MFG. CompanyCustomer Customer CTQ VOC VOB 6 DPMO New Std. 3 66800 DPMO Past Std. 27 • 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 limit 28 • The Cost Opportunity  2 3 4 5 6 0 5 10 15 20 25 30 About 15% of Sales, Cost Opportunity on 3 Company Cost of Failure (% of Sales) 29 • 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, Excess Inventory Rejects, Warranty, Inspection, Scrap, Rework 30 • Who is Implementing 6  • Motorola 1987 • Texas Instrument 1988 • ABB (Asea Brown Boveri) 1993 • Allied Signal 1995 • General Electric 1995 • Kodak 1995 • Siemens, Nokia, Sony 1997 31 • 6  Benefits(Case 1) ---Motorola over 12 yrs • Increase productivity an average of % per year • Reduced the cost of poor quality by more than 84% • Eliminated % of in process defects • Save more than $11billion in manufacturing cost • Realized an average annual compounded growth rate of 17% in earnings 32 • 6  Benefits(Case 2) General Electric (Million Dollars) Cost Profit 1996 240, 200 1997 400, 700 1998 450, 1200 1999 520, 2520 2000 600, 3000 33 • 6  Benefits( Case 3) Texas Instruments Before (1988) After (1993) COPQ 30% 7% DPMO 104000 9000 Scrap $3M $ Yield % % Cycle Time (Week) 11 4 Inventory Cost $ $ 34 • The Focus of 6  If we are perfectly control X, should we constantly test and inspect Y? Y = f (X) YX1 X2 X3 X4 X1、X2、…Xn ◎Independent ◎Input-Process ◎Cause ◎Problem ◎Control Y ◎Dependent ◎Output ◎Effect ◎Symptom ◎Monitor 35 • 6  is Applying Overall Business System R&D 6 Trans- action Mfg Design SS Manufacturing SS Transactional SS 36 • 6  Methodology Measure Analyze Design Verify Measure Analyze Improve Control Define Translate Y NNew Process/ Product ? Achievable Goal ? N Y 37 • What is different with 6  Traditional Quality 6 ◎Central Organized ◎Not Formal Structure for Tool Application ◎Lack of Support in using Tools ◎Lack of Structured Training ◎Inspect Quality in (Focus on “Y”) ◎Black Belt Report Directly Into Function ◎Structured Use of Statistical Tools to Aid Problem Solving ◎Data-Based Decisions ◎Structured Training in Applied Statistics ◎Control Process Input (X’s) 38 • 6  Organization • 6  Organization ◎Executive Project Selection and Support ◎6Technical Lead 6 Coach Project &BB ◎6 Project Team Lead ◎6 Project Team Member ◎All staffs (6 overview) Champion Master Black Belt (MBB) Black Belt (BB) Green Belt (GB) White Belt (WB) 39 • 3. 6  Methodology • 5 Stage of DMADV and New Tools ---- Design For Six Sigma (DFSS) ---- Black Belt • 5 Stage of DMAIC methodology ---- Green Belt 40 . DMAIC Methodology 5 Stages DMAIC Methodology and Statistical Tools • 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, COPQ 41 • Define Steps (1) Vision Business Strategy Big Y (CTQ) Small y ---Project (Goal, Scope, Performance Index) 42 • Define Steps (2-1) • Vision:最具競爭力的企業(GE) • Strategy---The Three Circle (GE) 核心事業 ◎照明 ◎大型家電 ◎馬達 高科技事業 ◎醫療系統 ◎工業電子 ◎航太 服務事業 ◎信用公司 ◎資訊服務 ◎核能服務 43 • Define Steps (2-2) ---The Three Circle (GE) • Jack Welch (GE’S CEO)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)。 44 • Define Steps (3) Select Target Product/Service Analysis Business Process Analysis Core Process Identify Customer Listen to VOC Specify Customer Requirement Select CTQ (Big Y) Specify CTQ (Small Y) Evaluate Potential Projects Select Project & Build Effective Team 45 • Define Steps (4) • CTQ --- Critical to Quality • 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 或組織在乎的事。『新點子』要採納『 外部觀點』(Outside-in Perspective)， 即從顧客的眼光來看企業或組織的一切。 46 . DMAIC Methodology 5 Stages DMAIC Methodology and Statistical Tools • Phase: Measure • Steps; Work Breakdown & Tool • M1 Specify Project; • CTQ Tree, Process Map, Performance Index • M2 Assess Measurement System; • Measurement System Analysis, Gage R&R • M3 Identify Sigma Level; • Graph Analysis, Capability Analysis, Confidence Interval 47 • M1 Specify Project • M1 step covers the followings: ◎CTQ Breakdown ◎Define Scope ◎Process Mapping ◎Pareto Analysis ◎Define performance index and specifications The success of any 6 activities relies mostly on the CTQ definition and selection 48 • Define Performance Index Customer Requirements Internal Process Improved Process Customer Satisfaction Index of current level Index of improved level Perform project for improvement Clear Quantifiable Simple 49 • Example of Performance Index ◎Yield ◎Cycle time ◎Defect rate ◎Machine failure rate ◎Customer stand-by hours ◎Number of invoice errors ◎Elapse Time from loan application to money transfer to the customer account ◎Hours taken from receiving order to delivery to the customer 50 M2 Measurement System Analysis • M2 step covers the followings: •Statistic fundamentals •Sampling plan •Data collection plan •Variation of measurement system •Gage R&R study •Improvement of measurement system Without accurate measure, can’t identify any symptom of problem 51 • 6  Metrics Data Type Statistic measurement data Comparison to specification Z-value Discrete Data Continuous Data Defect unit opportunity (DPMO) Average, Standard variation, Shape SPEC Z-value Sigma calculation 52 • 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: Defect per Million Opportunities 53 • 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 = DPO =5/100 = • = , Z = •  level = Z + = + = 54 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 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) =, DPMO = *10^6 =2000 • 1- = , Z = •  level = Z + = + = 55 • Example of Sigma calculation: • Continuous Data(1) 1  Mean Median Mode 70 80 90 100 110 120 130 +  -  Normal Distribution with mean = 100 and std. = 10 56 • 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) The normal distribution has the following properties. ◎% of the data fall within 1 ◎% of the data fall within 2 ◎% 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 specification limits and targeted value. 57 • 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 deviation USL Sigma of the process Z = 58 • Measurement System and Measurement Error(1) • Measurement system is viewing as a process. • Sources of variation: 5M1E • Validate possible sources of variation in the measurement process Variation in measurement System 59 • Measurement System and Measurement Error(2) Average m (Total) = m (Product) + m(Measurement) Variability s2 (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) 60 • Process Variance Observed Process Variation Actual Process Variation Measurement Variation Long term process variation Short term process variation Variation within a sample Variation due to operators Variation due to gage Reproducibility LinearityStabilityRepeatabilityAccuracy 61 • Gage R&R Analysis • Types of Gage R&R Analysis Methodology ◎X bar- R Method ◎ANOVA Types of Variation estimation by the Gage R&R ◎Equipment Variation: EV ◎Appraiser (Operator): AV 62 • 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 direction For repeatability error  reproducibility error • (Need to taken an action to operator) For reproducibility error  repeatability error • (Need to taken an action to gage) 63 • M3 Identify Sigma Level • M3 step covers the followings: •Data stratification •Graphical cause elimination •Understand product capability and product performance •Calculate current sigma level 64 • Graphical Analysis ◎Run Chart ◎Scatter Diagram ◎Box Plot ◎Histogram • Changes to the process may be more easily recognized graphically than tabularly. 65 • Understanding Variation • What is Variation? ◎ Different outcome of a process or result of 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 it’s variations • Why a variation might be occurred? ◎ By a common cause ◎ By a special cause ◎ 5M1E • What impacts will be happened if a variation become big? • Can not predict/ forecast outcomes •Occur re-inspection and/ or rework •Delay schedule ◎ Increase customer dissatisfaction 66 • Short-Term/Long-Term Relationships Inherent Capability of the Process---Short Term Capability Sustained Performance of the Process---Long Term Performance Over time, a “typical” process will shift and drift by approximately  67 • Product Capability(Cp) • Cp = |USL-LSL| / 6st ； Zst = 3 Cp • Capability Index(Cpk) • Cpu = min (Cpu, Cpl)， Where Cpu = (USL-)/3st； Cpl = (-LSL)/3st； Zst = 3 Cpk • Performance Index(Cpk) • Cpk = min (Cpu, Cpl)， Where Cpu = (USL-)/3lt； Cpl = (-LSL)/3lt； Zst = 3 Cpk 68 . DMAIC Methodology 5 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; • Regression, Hypothesis Test, ANVOA, Multi-Vari Analysis 69 • A1 Set Up Improvement Goal •In 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 available resources. 70 • A1 step covers following topics: –Establish performance goals –Assess 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 level 71 • Paths to Performance Goals Performance Goals Process Improvement (Entitlement) Process Change (Breakthrough) Baseline Process Measurement 72 • Benchmarking --- Why Benchmark (1)? Innovation Continuous Improvement Breakthrough Improvement Continuous Improvement Time P erform ance 73 • Benchmarking --- Why Benchmark (2)? •Develop and Improvement Strategic Goals •Establish actionable objectives •Provide sense of urgency •Encourage breakthrough thinking •Create a better understanding of your industry 74 • A2 Identify Potential Causes • Finding main independent variables, and making the list Y = f (X) • Object Make list of potential independent variables for changing & improving Y value methods • Brainstorming ◎ Cause & Effect Diagram • Multi-voting ◎ Logic Tree 75 • Cause & Effect Diagram Variation in measurement System 76 • A3 Verify Potential Causes • Overview •In step A3, setting a possible priority as to their importance to the potential variables (X’s) identified in step A2 • Purpose •General a list of important factors (vital few) from the potential variables. Y = f [(x1,x2,x3) (x4,x5,….,xn)] Vital Few Trivial Many 77 • Pareto Analysis 78 • Correlation Analysis •A statistical analysis to investigate / measurement of association between two variables (X, Y) is called analysis. •Correlation tells 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. 79 • Regression Analysis (1) Y = f (x1, x2, x3,…) Status and characteristics of a process Modeling Mathematical equation X Y 80 • Regression Analysis (2) • Types of Regression Model •Simple linear regression •Multiple linear regression •Non-linear regression Y = f(X) Y: dependent variable X: independent variable 81 • Types of Hypothesis Test Types of Data Discrete DataContinuous Data Mean Test ◎t –Test ◎ANOVA Variance Test ◎F –Test ◎Chi- Square 82 . DMAIC Methodology 5 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 Matrix 83 • I1 Identify Vital Few • In analysis step, potential X’s were identified with a possible prioritization as to their importance in controlling Y. Improve step provides tools to explore and uncover the vital X’s and how much impact each X has on the response Y. •One of the major tools used for both exploration and confirmation of the effect of X’s on the Y’s is experimental design (DOE). By experimentally manipulating the independent variables, DOE provides an efficient and economical means for reaching valid and relevant conclusions about a process. 84 • Types of Experimental Design Screening ◎ Full factorial ◎ 2K factorial ◎ Fractional factorial Optimization ◎ Multi-level experiments ◎ Composite design ◎ Response surface method 85 • I2 Formulate Optimum Model •The vital few important X’s were identified and characterized in I1. The tools in I2 follow a methodology for establishing levels of these X’s that will provide improved performance for the Y. 86 . DMAIC Methodology 5 Stages DMAIC Methodology and Statistical Tools • Phase: Control • Steps; Work Breakdown & Tool • C1 Identify Control Capability; • Gage R&R, Capability Study, Gap Analysis • C2 Establish Control Plan; • Control Plan, Control Chart • C3 Link to Management System; • Control System, Six Sigma Audit 87 • 4. 結語 (1) • 6 是一以統計為基礎之品質 系統 , 目的在 使企業或組織的產品 /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”(Human Element).成功取決於 一致性 (Consistency)---Vision, Business Strategy, Specify Project, and any 6  A c t i v i t i e s. • 6 品質 系統 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 系統所 能 比擬 . 88 • 4. 結語 (2) • 學習(Learning) 、價值觀(Value)與 創新(Innovation)為企業思想的核心, 其凌駕階級與傳統之上. 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 中央舞台上.『持續學習,並將其轉化為 行動的欲望與能力』. 89 • 4. 結語 (3) • 外部觀點(Outside-in Perspective) ,『 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』. 90 • 報 告 完 畢 •敬 請 指 教 • 萬 分 謝 謝 91