问题解决工具-七步法.ppt

《问题解决工具-七步法.ppt》由会员分享，可在线阅读，更多相关《问题解决工具-七步法.ppt（62页珍藏版）》请在装配图网上搜索。

1、问题解决工具七步法,Problem Solving-7 STEP,Quality Issue,S1 Identify the Problem,S2 Determine Possible Causes & Rank their importance,Incoming issue (PUR) In process issue (FIRG,MQV,PQA,EQA) Customer complains (BIS,ELF,PIR),Brainstorming 5W1H Picture/ Sketch Check sheet tracking spreadsheet PFD Histograms Par

2、eto Diagram VOC Control chart Run chart Relationship Maps MSA PPM Is/Is not Team Leader Cross functional team,S3 Short Term Action?,Temporary Countermeasure,Open Brainstorming Fishbone Check sheet Pictograms FTA Flow Diagrams FMEA 5WHY Rank Is/Is not Multi-voting Pareto Diagram,S5 Conduct Tests Anal

3、yze Data Identify Root Cause(s) Select Solution,S6 Plan and Implement Permanent Solution,Tools Brainstorming Design/ Process Error proofing Flow chart GanttChart PFD Action: Engineering change Control plan D/PFMEA Job instruction Lesson learnt Layered process audit Check list Boundary sample Machine

4、 adjustment Production Plan Service Parts Plan Service Information PITS Documentation SPT,Pareto Diagram Run Chart Trend Chart Control Chart Histogram Design/ Process Error proofing D/PFMEA Process Control Plan Key team achievement Fix Effectiveness GPRP,Y,N,Protecting the customer and satisfying th

5、e customer,S4 Gather Data & Design Tests,At supplier On way material At plant WIP Inventory Test samples Scrap Rework On way finish goods to customer At customer Temporary action/ TRP Campaign,Histogram Run Chart Trend Chart Check Sheet Control Chart Pareto Charts Capability Index DOE Concentration

6、Diagram Regression Analysis Anova FTA,S7 Measure and Evaluate for Continuous Improvement & Recognize the Team,Brainstorming Capability Index Control Chart Decision Tree / analysis Histogram Hypothesis Testing Linear Regression Analysis Run Chart Trend Chart,问题解决-7步法,质量事故,S1 问题描述,S2 确定可能的原因& 排列它们重要性,

7、来料问题 (PUR) 过程问题 (FIRG,MQV,PQA,EQA) 客户抱怨 (BIS,ELF,PIR),头脑风暴 5W1H 照片/草图 检查表 跟踪表 过程流程图 直方图 排列图 客户的意见VOC 控制图表 运行图 关联图 测量系统分析 PPM 是与不是 小组负责人 跨部门的功能小组,S3 采取短期行动,临时措施,Open 头脑风暴 鱼骨图 检查表 图示效果图 FTA故障树分析 作业流程图 FMEA 5WHY 排序 是与不是 Multi-voting 排列图,S5 实施试验/分析数据，验证根本原因，选择解决方案,S6 计划和执行永久改善方案,工具 头脑风暴 设计/过程防错 流程图 干特图

8、工作流分析 措施 工程变更 防错验证 控制计划 设计/过程FMEA 操作指导书 经验教训 分层审核 检查表 边界样件 设备调整 生产计划 服务件计划 服务信息 PITS 文件化 服务零件简报,排列图 运行图 趋势图 控制图 直方图 设计/过程防错 设计/过程FMEA 过程控制计划 关键小组成就 措施有效性 全球问题再发生预防,Y,N,保护客户和使客户满意,S4 收集数据& 为原因制定试验计划,在供应商处 在途原材料 在工厂内 正在工作中的WIP 库存 测试样件 废料 返工 在途发往客户的产品 客户处 临时行动/TRP 召回,直方图 运行图 检查表 控制图 排列图 能力分析 DOE 集中图 回归

9、分析 方差分析 故障树分析,S7 测量和评估持续改进，认可团队,头脑风暴 过程能力分析 控制图 决定树/分析 设计过程防错 DOE 直方图 假设检验 线形回归分析 运行图 趋势图,STEP 7. MEASURE & EVALUATE (PREVENT RECURRENCE) RECOGNIZE THE TEAM,STEP 6. PLAN & IMPLEMENT PERMANENT SOLUTION,STEP 5. CONDUCT TEST/ ANALYZE DATA/IDENTIFY ROOT CAUSE(S) SELECT SOLUTIONS,STEP 1. IDENTIFY THE PRO

10、BLEM,STEP 2. DETERMINE POSSIBLE CAUSES & RANK THEIR IMPORTANCE,STEP 3. TAKE SHORT TERM ACTION,STEP 4. GATHER DATA/ DESIGN TESTS OF CAUSES,PLAN,DO,CHECK,ACT,Problem Solving-7 STEP 问题解决-7步法,走入现场 数据说话 团队协作,Problem Solving-5WHY 问题解决-5WHY,Cookies taste really bad 甜点口味不好,Problem问题,5个为什么分析,Problem Solving

11、-5WHY 问题解决-5WHY,5 WHY ANALYSIS 5个为什么分析 If you found more than one Possible or Direct Cause, please utilize additional sheets for each Direct Cause 如果你发现了不只一个可能或直接的原因，请对每一个直接原因利用另外的表,1). Why did the manufacturing process not prevent this failure mode?为什么制造过程没有预防这种失效？,2). Why did the quality process n

12、ot protect customer from this failure mode?为什么质量过程没有保护客户远离这次失效？,预防：制造过程-预防&标准操作,保护：质量过程-探测&围堵政策,发生源,流出源,3).Why did the planning process not predict this failure mode?为什么计划过程没有预见此次失效？,预见：计划过程-FEMA和控制计划 中的信息内容,系统源,Problem Solving -5WHY 问题解决-5WHY,User Guide 使用说明 以上5页为使用工具和改进思路指导，供大家参考！ 以下为项目PPT模板和培训材料

13、页为培训材料 其他页为PPT模板， 在PPT模板中，每一步都罗列了很多工具以及这些工具的使用说明。在每一页中有特别注明：如必须使用，建议使用(选择性使用) 等 在模板中蓝色字体， 为对相关工具的解释，可在今后使用中用实际问题描述代替。 对如何使用该页模板进行解释,XXX,Identify The Problem问题描述,“A PROBLEM WELL STATED IS A PROBLEM HALF SOLVED.” “准确的问题描述是解决问题的一半” - Charles F. Kettering,PURPOSE目的: TO DEFINE THE PROBLEM IN TERMS RELATE

14、D TO THE PRODUCT, PROCESS OR OTHER REQUIREMENTS 识别与产品，过程和其他要求相关的问题 TO LEAD TO QUICK STEP 3 AND STEP 6 ACTION 确保有快速的第三步和第六步措施 DEFINE THE CUSTOMER REPORTED PROBLEM IN AS MUCH DETAIL AS POSSIBLE 尽可能详细地描述客户报告的问题。 Understand your customers desire this may be different to your perception of the problem 了解

15、你用户的期望，可能不同于自己对问题的理解 FORMULATE A CROSS-FUNCTIONAL TEAM AS APPROPRIATE ( INCLUDING CUSTOMERS AND SUBCONTRACTORS)确定适当的多功能小组（包括客户和分供方） DRAW IN APPROPRIATE EXPERTISE TO HELP DEFINE THE PROBLEM 征集专家意见帮助定义问题 DEVELOP A CLEAR, CONCISE, VISIBLE AND QUANTIFIABLE STATEMENT DEFINING THE PROBLEM, BACKED BY ONE O

16、R MORE OF THE FOLLOWING TOOLS : 采用一个或多个如下工具， 进行清晰扼要的， 可视的，量化的描述定义问题。 Check sheet or tracking spreadsheet, Picture/ Sketch, control charts, run charts, Pareto diagrams, PFD-flow charts ， Histograms，5W1H, PPM, MSA ， Is or is not etc. 检查单或跟踪表， 照片，控制图， 运行图，排列图，流程图， 直方图， 5W1H， PPM， MSA， 是与不是等,Identify Th

17、e Problem问题描述,ASSESS OVERALL RISKS BY CONSIDERING ”MULTIPLE CUSTOMER LOCATIONS?” AND SEVERITY OF RISK 通过考虑是否多客户发生，风险严重度评估总体风险 IDENTIFY AND QUANTIFY 识别和量化: Is the information supplied adequate? 提供的信息足够吗？ What additional information is required? 还需要什么补充信息？ Were parts returned? 零件退回没有？ What inspection

18、data is required? 需要什么检验数据？ What historical information is needed - change history etc. 需要什么历史数据 变更历史 等 What is the application? 什么应用？ What is the circumstances surrounding the failure? 与失效相关的坏境? TRANSLATE THE REPORTED PROBLEM INTO A TECHNICAL PROBLEM DESCRIPTION OR STATEMENT将报告的问题转化成“技术上的问题描述”: Wha

19、t is the exact problem? 准确的问题是什么？ What was the specification or requirement that wasnt met? 没满足的规范和要求是什么？ How often? 多久？ THE FINAL STATEMENT SHOULD ALLOW AN INDIVIDUAL OR TEAM TO KNOW WHAT THEY HAVE TO FIX, AND TO BE ABLE TO KNOW WHEN THEY HAVE FIXED IT 最终的陈述应该让个人或小组知道他们要解决什么，并且知道何时解决。,“A PROBLEM WE

20、LL STATED IS A PROBLEM HALF SOLVED.” “准确的问题描述是解决问题的一半” - Charles F. Kettering,CUSTOMER REPORTED PROBLEM 客户报告的问题 Brief description of customer symptom. Can include part / defect information if known but should focus on the problem that the customer experiences as a result of the issue. 简要描述客户的症状。如果知道

21、，可以描述零件和失效的信息，但必须集中在客户的体验上，作为问题造成的结果。,EMERGENCY RESPONSE ACTION紧急反应活动 Description描述: Brief description of ERA (if any). If no ERA then “Not applicable” with justification. 简要描述ERA(如果有)。如果没有，则认为“不适用” Verification details验证细节: Details of test to ensure ERA will protect customer & not cause other conce

22、rns 进行详尽的试验确保ERA可以保护客户并不引起其它失效,CUSTOMER DESIRE IDENTIFICATION 客户期望识别,Step 1 Identify The Problem 第一步 问题描述,BIS售时问题：必须用 Warranty 售后问题：建议用 工厂内部问题：不用 供应商在工厂内发生问题：供应商必须用,7 STEP TEAM DETAILS 7步法团队 Team Leader 小组负责人：XXX Team Members小组成员: Note: If necessary, including customer and /or subcontractor into 7 s

23、tep team.如果必要在小组中增加客户和/或分供方。 Team need include hourly/production employees? (where applicable)小组成员是需包括临时工/生产工人？（如适用） Meeting schedule (or see next slide) 会议时间表（或见下页）: Daily / weekly, location, day, time 日常/每周，地点，日期，时间,Required 必须用,Step 1 Identify The Problem 第一步 问题描述,Failure Mode 失效模式: Fail Codes/ L

24、OP 失效代码/LOP: Predicted RPH (C/1000) 预测RPH(C/1000): CPE (EPUS)每台发动机的金额: Total Cost 金额总计: Additional Trends 其它趋势: (application应用, build month生产月份, mileage行驶里程, etc等等。) Additional Key Step 1 information其它主要第一步信息:,“能定量描述，不定性描述”,Step 1 Identify The Problem 第一步 问题描述,Warranty 售后问题：必须用， 多件时，应使用清单,MSA,MAB (M

25、OP/MIS) Chart,Chart should be for component or system level depending on the scope of the project. 基于项目的范围确定图表是零件级或系统级。 Clearly identify dates previous improvement, production changes, etc. 清楚地识别前期改进日期，生产变更日期等。 Identify area of concern project is addressing and % of failure rate project will address

26、. 所关注项目的识别区域被确定 并且项目失效百分比将被确定。,Step 1 Identify The Problem 第一步 问题描述,Warranty 售后问题：必须用,“能定量描述，不定性描述”,PARETO OF FAILURE MODES失效模式排列图 Pareto chart showing results of analysis of returned parts (teardown). Can substitute warranty verbatim pareto as an initial analysis. 排列图展示了退回件的分析结果（拆卸）。初始分析，可以逐步地替换售后零

27、件。,PROBLEM DESCRIPTION 问题描述 5W1H, DRAWINGS, PICTURES, DATA, SAMPLES. 5W1H, 图纸, 图片, 数据, 样品 1. Does the problem statement tell what, when where and the magnitude of the issue? 问题描述是否告知了时间，地点，发生了什么事，问题的影响有多大？ 2 .Does the problem statement avoid predicting cause? 问题描述是否已避免直接写出原因？ 3. Problem includes cus

28、tomer and Cummins Part Number? (where applicable) 问题中是否包含了顾客和公司的零件号？（如适用 4. Are there pictures/sample parts etc? 有图片/样件吗？ 5. Are Paretos approached from multiple angles? 是否从多个角度应用了排列图的方法？,RECURRENCE 是否重复发生 Yes No,Step 1 Identify The Problem 第一步 问题描述,MSA,根据不同的问题，可采取不同的质量工具， 但一定要确认是否重复发生,根据不同的问题，可采取不同

29、的质量工具， 但一定要确认是否重复发生,HISTOGRAM 直方图,CONTROL CHART 控制图,Step 1 Identify The Problem 第一步 问题描述,PROCESS FLOW 流程图,MSA,First 30 days planned/ target dates auto-populate based on launch date 基于项目开始日期，前30天的计划/目标日期会自动生成。 Track first 30 in days, and weekly or as needed after 根据需要按日/周/更长时间 跟踪前30天状态。 Include secti

30、on of tracker relevant since last update (1 month) 应包括上次更新后（1个月）与跟踪相关的内容。 Identify gaps and recovery plan.识别不足并完善计划。,Project Tracker项目跟踪,Step 1 Identify The Problem 第一步 问题描述,建议用,Determine Possible Causes & Rank their importance确定可能的原因& 排列它们重要性,PURPOSE 目的: TO DOCUMENT ALL THE POSSIBLE CAUSES FOR THE

31、DEFINED PROBLEM, SELECT THE MOST LIKELY CAUSE (S) FOR A MORE FOCUSED APPROACH 针对定义的问题，列出所有可能的原因，为关注的方法选择更可能的原因 Note: Step 2 and Step 3 should be concurrent activities. 第二步和第三步应该同时发生 DETERMINE MAJOR CATEGORIES OF CAUSE确定原因的主要分类: Cause can come from everywhere (including subcontractors, customer). 原因可

32、能来自任何地方（分供方， 客户） Identify possible process and system cause. 确定可能的过程和系统原因 USE A SYSTEMATIC LOGICAL REASONING APPROACH TO IDENTIFY THE ROOT CAUSE(S) OF A PROBLEM 使用系统的，逻辑的原因接近方法， 识别问题的根本原因。 Utilize team BRAINSTORMING techniques to capture the cause and effect in the form of an ISHIKAWA DIAGRAM (also

33、called Fishbone OR cause and effect diagram).利用头脑风暴法技术， 以鱼骨图（因果 图）形式捕捉原因和结果 ENSURE THAT POSSIBLE CAUSES HAVE BEEN REDUCED TO THE LOWEST LEVEL BY USING THE 5 WHYS, BRAIN- STORMING ETC.通过5 个为什么，头脑风暴法， 确保可能的原 因分解到最底层。 RANK MOST LIKELY CAUSE 排列更可能的原因: It is acceptable for rank to be based on opinions an

34、d best guesses at this stage.该阶段，根据意见和最佳推测排序 Ranks should be determined by consensus。(Notes: Data collection and verification will be conducted in subsequent steps.)按对数人意见 进行排列,Brainstorming 头脑风暴法：“THE BEST WAY TO GET GREAT IDEAS IS TO GET LOTS OF IDEAS AND THROW THE BAD ONES AWAY”. 获得好想法的最佳方法是收集大量想

35、法，然后去掉不好的想法。,Step 2: Determine Possible Causes and Rank Their Importance第二步：确定可能的原因& 排列它们重要性,建议用,WHEN BRAINSTORMING 在头脑风暴法时 START WITH SILENT THINK TIME 开始静静思考一下 AVOID CRITICISM OR EVALUATIONS WHILE BRAINSTORMING 避免评判和评估 AVOID HITCHHIKE - BUILD ON IDEAS 避免搭便车 MORE IDEAS THE BETTER 越多越好 DO NOT EXPLAI

36、N IDEAS 不要解释想法 AVOID ATTEMPTING TO FIND SOULTIONS TO CAUSES 防止试图找到原因的解决办法 THERE ARE NO BAD IDEAS 没有坏的想法,DATA DRIVEN TOOL ANALYSIS数据驱动分析 Data driven tool used to sort through potential causes while focusing on finding the root cause. Eliminates the need to brainstorm many other causes that have been

37、ruled as “not” causing the problem. 数据驱动分析用来分类潜在原因而聚焦在根本原因上。排除头脑风暴中那些被裁定为不会引起问题的其它原因。,Step 2: Determine Possible Causes and Rank Their Importance第二步：确定可能的原因& 排列它们重要性,Notes： How to Rank Their Importance, 如何排列重要性 0 =不相关no correlation 1 =对失效模式（效果）影响很小 Remotely affects the failure mode (EFFECT) 3 =对失效模式

38、（效果）有一定影响 Moderate effect on the failure mode (EFFECT) 9 =对失效模式（效果）有重要的影响 Strong effect on the failure mode (EFFECT),使用数据驱动分析和因果分析之一,FISHBONE DIAGRAM ANALYSIS因果分析 Show completed Fishbone (cause & effect diagram) to demonstrate how root cause theories were derived. 展示完成的鱼骨图(原因&效果图表)来证明怎样得到根本原因推测。 Not

39、e： 1) Indicate possible direct causes on diagram. 2) Circle most likely direct causes. 注意: 1) 在图表上简要的说明可能的直接原因。 2) 标出非常可能的直接原因。,Step 2: Determine Possible Causes and Rank Their Importance第二步：确定可能的原因& 排列它们重要性,Notes： How to Rank Their Importance, 如何排列重要性 0 =不相关no correlation 1 =对失效模式（效果）影响很小 Remotely

40、affects the failure mode (EFFECT) 3 =对失效模式（效果）有一定影响 Moderate effect on the failure mode (EFFECT) 9 =对失效模式（效果）有重要的影响 Strong effect on the failure mode (EFFECT),使用数据驱动分析和因果分析之一,FTA失效树分析: A method to analyze the failure causes from top event to basic characteristics, in order to identify possible root

41、causes and find the failure mechanism.一种从顶端事件到基本特性的分析失效原因的办法，用以识别可能的根本原因及寻找失效机制。 Discourage the user from jumping to a conclusion阻止使用者直接跳到结论； Facilitate problem solving 促进问题的解决 Organize failure analysis investigations.组织故障分析调研,Step 2: Determine Possible Causes and Rank Their Importance第二步：确定可能的原因& 排

42、列它们重要性,Required 必须用,Go slow in order to go fast 欲速则不达,Engine wont start 引擎不能发动,Wrong Injection timing错误的注油时间 Low compression低压缩 Lack of fuel缺少燃料 Lack of air缺少空气,Top Event首要事件 The problem问题,Basic Events基本事件 Basic Characteristics基本特征,Intermediate Events中间事件 Interface Mechanisms划分机理,FTA Concept FTA基本概念

43、,Determine Possible Causes & Rank their importance确定可能的原因& 排列它们重要性,Basic Events & Characteristics Example基本事件和特征实例,Basic Events基本事件 Low torque低扭矩 Low hardness低硬度 Excessive engine speed引擎速度过快,Basic Characteristics 基本特征 Torque - ft/lbs or N/m扭矩 Hardness- in Rockwell C硬度 Engine speed RPM引擎速度,NOTE: The

44、basic events must be stated in terms of quantifiable and measurable basic characteristics. 注意：基本事件必须陈述可量化和可测量的基本特征。,Determine Possible Causes & Rank their importance确定可能的原因& 排列它们重要性,5 Whys,Determine Possible Causes & Rank their importance确定可能的原因& 排列它们重要性,PURPOSE目的: TO DEVELOP AND IMPLEMENT TEMPORARY

45、 FIX(ES) 开展并实施临时修正. TO TAKE IMMEDIATE ACTION TO PROTECT THE CUSTOMER FROM FURTHER DISRUPTION. 采取立即措施，保护客户 TO ENSURES THAT THE PROBLEM IS CONTAINED, AND THAT THE CUSTOMER IS NOT AFFECTED AGAIN.确保问题得到围堵， 不让客户再受影响 TO MAINTAIN GOOD CUSTOMER RELATIONS.与客户保持好的关系 TO PROTECT THE SAFETY OF EMPLOYEES OR CUSTO

46、MERS.保护员工和客户安全 ENSURE THAT THE DATE, DATE CODE, SERIAL NUMBER ETC., CORRESPONDING TO IMPLEMENTATION OF EACH SHORT TERM ACTION IS DOCUMENTED FOR TRACEABILITY. 确保与每一个短期措施相对应的日期，批号， 序列号等文件化，可用于追溯。,Containment 围堵,Take Short Term Action 采取短期行动,TYPICAL SHORT TERM ACTIONS典型的短期行动: INCREASE INSPECTION FREQUE

47、NCY (OFTEN TO 100%)增加检验频次 INCREASE FUNCTIONAL TEST FREQUENCY 增加功能检查 CHANGE OR ADD NEW FUNCTIONAL TESTS TO DETECT THE SUSPECTED FAILURE 改变或增加新的功能测试，探测可疑的失效 IMPROVE PROCESS DOCUMENTATION, WORK INSTRUCTIONS 改进过程文件， 作业指导书。 SORT/RE-INSPECT ALL STOCK - INCLUDING STOCK HELD AT THE CUSTOMER/IN-TRANSIT ETC.

48、挑选/再检查所有库存- 包括存在客户或在途的产品 COMMUNICATION TO EMPLOYEES, OTHER PLANTS, DEPARTMENTS OR FUNCTIONS AS APPROPRIATE 告知员工， 其他工厂，部门等 VERIFY COMPETENCY 验证有效性,SOME OF THESE ACTIVITIES MAY ALREADY HAVE BEEN INITIATED PRIOR TO THE START OF FORMAL PROBLEM SOLVING 在开始正式的问题解决前，这些活动中的一部分也许已经进行。,Take Short Term Action

49、采取短期行动,Take short action (s) -Interim Containment临时措施,Avoid Suspect Finish Goods to be Shipped to Customer 防止可疑产品交付顾客,Verification details验证: Is there documented verification that containment actions are effective?是否有文件证明临时遏止措施是有效的？,Step 3: Take Short Term Action 第三步：采取短期行动,First Choice 首选,List deta

50、ils of step 3 fix implemented, 30days to completed step 3 fix. 列出详细的第三步确定执行的内容，完成时间控制在项目起动30天内。 Record date, date codes, ESNs, etc corresponding to each short term action 记录日期，日期代码，ESNs等其它符合短期措施的行动。 If no Step 3 fix implemented, explain why and/or recovery plan. 如果第三步没有确定的执行内容，则需要说明为什么或是覆盖计划。,Step 3

51、: Take Short Term Action 第三步：采取短期行动,Take short action (s) -Interim Containment临时措施,Avoid Suspect Finish Goods to be Shipped to Customer 防止可疑产品交付顾客,Second Choice 其次选,List details of step 3 fix implemented, 30days to completed step 3 fix. 列出详细的第三步确定执行的内容，完成时间控制在项目起动30天内。 Record date, date codes, ESNs,

52、 etc corresponding to each short term action 记录日期，日期代码，ESNs等其它符合短期措施的行动。 If no Step 3 fix implemented, explain why and/or recovery plan. 如果第三步没有确定的执行内容，则需要说明为什么或是覆盖计划。,Step 3: Take Short Term Action 第三步：采取短期行动,Take short action (s) -Interim Containment临时措施,Avoid Suspect Finish Goods to be Shipped to

53、 Customer 防止可疑产品交付顾客,Third Choice 第三选项,PURPOSE目的: TO IDENTIFY, WITH DATA, THE PROBABLE PROBLEM CAUSE(S) AND TO DESIGN THE TESTS REQUIRED TO VERIFY AND REPLICATE THE CAUSES. 数据说话，识别非常可能的问题原因， 并且设计需要的实验去验证和再现原因。 TO DEVELOP A DATA GATHERING PLAN TO CONFIRM THE POSSIBLE CAUSES IDENTIFIED IN STEP 2, 制定数据

54、收集计划去确认第二步识别的可能原因。 E.G. REVIEW HISTORICAL RECORDS, MEASURE PARTS ETC.如审阅历史数据， 测量零件等 TO RUN THE ACTUAL PROCESS - OBSERVE, MEASURE, CHART/SPC ETC.运行实际流程-观察，测量，图示/SPC等 IF ABOVE TWO ACTIONS DO NOT REPLICATE THE REPORTED PROBLEM, THEN DESIGN ONE OR MORE TEST TO CONFIRM POTENTIAL CAUSE AND REPLICATE THE R

55、EPORTED PROBLEM. 如果上两项不能在现问题， 需要设计一个或多个试验， 以便确认潜在的原因并再现问题。 ALWAYS INVOLVE THE PEOPLE CLOSEST TO THE PROCESS 总是让人们靠近过程。 Take care to explain what and why when attempting to collect 在设法收集数据时要关注”什么”，”为什么”.,Gather Data & Design Tests to Determine Cause(s)收集数据& 为原因制定试验计划,DEFINE TEST OBJECTIVES 设计试验的目标,CH

56、OOSE WHAT TO MEASURE 选择如何衡量,DESIGN THE EXPERIMENT, IF NECESSARY 如必要， 设计实验,PREPARE 准备,BE PRECISE 准确 WHAT QUESTIONS ARE YOU TRYING TO ANSWER ?你要设法回答什么问题？ HOW GENERAL DO YOU WANT THE RESULT TO BE ?你期望的结果如何？,WHAT QUALITY CHARACTERISTIC ARE YOU GOING TO MEASURE ? 你要测量什么质量特性？,WHAT FACTORS ARE LIKELY TO AF

57、FECT THE RESPONSES YOU HAVE IDENTIFIED ?什么因数影响你已识别响应 WHAT ARE EFFICIENT WAYS TO TEST VARIOUS COMBINATIONS OF THESE FACTORS ?有何有效方法测试因数间的不同组合？,IS TRAINING NEEDED ? 需要培训吗？ WHAT CAN GO WRONG ? 什么能变差,TIPS FOR DESIGNING TESTS设计试验,DESIGN TEST TO DETERMINE CONTRIBUTING AND NON-CONTRIBUTING CAUSES 设计实验确定涉及和

58、不涉及的原因,Gather Data & Design Tests to Determine Cause(s)收集数据& 为原因制定试验计划,QUALITY TOOLS质量工具 HISTOGRAMS RUN CHARTS CHECK SHEETS 柱状图 趋势图 检查单 CONTROL CHARTS PARETO CHARTS CAPABILITY INDEX 控制图表 排列图 能力指标 CONCENTRATION DIAGRAMDOE PFD 集中图 实验设计 过程流程图 REGRESSION ANALYSIS, ANOVA FTA 回归分析 方差分析 故障树分析,对FTA每一条结果，列出收

59、集数据或进行试验的计划；并根据计划逐条实施，获得结果。（以下工具可参考使用）,Gather Data & Design Tests to Determine Cause(s)收集数据& 为原因制定试验计划,What,When,Who,Step 4: Gather Data & Design Test 第四步：收集数据& 为原因制定试验计划,Notes: WHAT: 尽可能使用相关质量工具。 1. 对部分数据， 可以通过快速数据收集方法得到原始数据证据， 2. 对一些数据， 需要制定更详细的数据收集/试验计划（DOE等）,Step 4: Gather Data & Design Test 第四步

60、：收集数据& 为原因制定试验计划,When,Who,Notes: WHAT: 尽可能使用相关质量工具。 1. 对部分数据， 可以通过快速数据收集方法得到原始数据证据， 2. 对一些数据， 需要制定更详细的数据收集/试验计划（DOE等）,PURPOSE 目的: TO CONDUCT TESTS IDENTIFIED IN STEP 4 执行在第4步制定的测试. TO CONFIRM THE TRUE ROOT CAUSE OF THE PROBLEM, 确定问题的真正原因 ANALYZE THE RESULTS OF THE TESTS TO ENSURE THE SOLUTION WILL N

61、OT CAUSE OTHER PROBLEMS分析测试结果确保解决方案不会产生其它问题。 Did the test (S) /historical data replicate the failure mode? If not, go back to Step 4.测试或历史数据能再现失效模式吗？ 如不能再现，回到第四步 Of the test (S)/ historical data that replicated the failure mode, what is /are the most likely root causes)？能再现失效模式的测试或历史数据， 是更可能的根本原因吗？

62、Review and update the cause and effect diagram or FTA in Step 2.回顾第二步的因果分析或FTA 。 TO SELECT THE MOST EFFECTIVE SOLUTION(S) 选择最有效的解决方案。 Beware of the single solution fix 谨防单一的纠正解决方案 Look for fail-safes/design changes 寻找防错/设计变更 Consider training & documentation needs 考虑需要的培训及文件 Can the root cause (s) b

63、e controlled? How? 能控制根本原因吗？如何？ What systems and procedures must be changed? 什么系统和程序需要更改？ Will the control ensure the solution is sustained? 控制能保证解决方案 持续不变吗？ Can the controls (solution) be automated? 控制（解决方案）是自动的吗？ Is there a way to prevent the controls (solution) from being bypassed?有何方法防止解决方案被绕过。

64、IDENTIFY MEASURES FOR A SUCCESSFUL IMPLEMENTATION 设别成功设施的测量指标。,Conduct Test, Analyze Data, Select Solution 实施试验/分析数据，识别根本原因，选择解决方案,QUALITY TOOLS质量工具 BRAINSTORMING DECISION MATRIX CHECK SHEETS 头脑风暴法 抉择矩阵 检查单 HISTOGRAMS RUN CHARTS CONTROL CHARTS 柱状图 趋势图 控制图表 CAPABILITY INDEX CONCENTRATION DIAGRAM DOE

65、能力指标 集中图 实验设计 REGRESSION ANALYSIS, ANOVA FTA 回归分析 方差分析 故障树分析 HYPOTHESIS TESTING DESIGN/PROCESS FAIL-SAFE 假设检验 设计/过程防错,Conduct Test, Analyze Data, Select Solution 实施试验/分析数据，识别根本原因，选择解决方案,Step 5: Conduct Tests/Analyze Data/Identify Root Cause(s) Select Solution第五步：实施试验/分析数据，识别根本原因，选择解决方案,Step 5: Conduct Tests/Analyze Data/Identify Root Cause(s) Select Solution第五步：实施试验/分析数据，识别根本原因，选择解决方案,5 WHY ANALYSIS 5个为什么分析 If you found more than one Possible or Direct Cause, please uti