《SPC统计技术流程控制--stevexinli.pptx》由会员分享，可在线阅读，更多相关《SPC统计技术流程控制--stevexinli.pptx（68页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、惠亚集团SPC课程 A Workshop from Our Worldwide SPC CurriculumSPC for Operators - Version 2.1 Mar. 10, 20012 培训目的培训目的Training Goal SPC for Operators - Version 2.1 Mar. 10, 20013 z什么是SPC及SPC的益处 What is SPC and benefits. z数据的收集及测量 Data collection and measurement. z使用反应计划解决问题 Troubleshooting using reaction pla
2、ns z流程的稳定&能力(CP和CPK) Process Stability and Capability.(CP and CPK)培训内容培训内容 Training Agenda SPC for Operators - Version 2.1 Mar. 10, 20014 什么是什么是SPC?SPC?What is SPC? 利用统计技术如控制图去分析流程或其产物利用统计技术如控制图去分析流程或其产物, ,以采取适当以采取适当的行动去达致及维持一个受统计技术控制的状态以及改善的行动去达致及维持一个受统计技术控制的状态以及改善流程能力流程能力. . The use of statistical
3、 techniques such as control charts to analyze a process or its outputs so as to take appropriate actions to achieve and maintain a state of statistical control and to improve the process capability. z SPC for Operators - Version 2.1 Mar. 10, 20015 Out of control 失控失控Trends 趋势趋势Shifts 偏移偏移统计技术统计技术 St
4、atistical 流程流程 Process控制控制 Control LCL UCL 什么是什么是SPC?What is SPC?SPC for Operators - Version 2.1 Mar. 10, 20016 Measurement 测量 Machine 机 Environment 环境 Man 人 Material 物料 Methods 方法 什么是什么是SPC?What is SPC?SPC for Operators - Version 2.1 Mar. 10, 20017 16.7%16.7%50.0%16.7%0.0%20.0%40.0%60.0%80.0%100.0%
5、DefectivesScratchPlugholeOpenShortSPC for Operators - Version 2.1 Mar. 10, 20018 Determine Key Product Characteristics 介定关键的产品特性介定关键的产品特性Determine Key Variables 介定关键的变量介定关键的变量Plan the Matrix of Validation Tests 计划确认测试的矩阵计划确认测试的矩阵 什么是什么是SPC?What is SPC?SPC for Operators - Version 2.1 Mar. 10, 20019 F
6、unction affected 被影响的功能被影响的功能 Potential Failure mode 潜在缺陷模式潜在缺陷模式 - Potential Effect 潜在影响 - Potential Cause 潜在原因 - Design Control 设计控制 Action Plan 行动计划行动计划 Result 结果结果 什么是什么是SPC?What is SPC?SPC for Operators - Version 2.1 Mar. 10, 200110 统计技术统计技术 Statistical 收集及整理数字化收集及整理数字化的事实数据的事实数据 Collection & a
7、rrangement of numerical facts. Data! 控制控制 Control 预防预防, ,预测预测, ,监管监管 Prevention. Prediction. Regulation. 流程流程 Process 执行一件工作的一执行一件工作的一连串步骤连串步骤 Series of steps to perform a task z是否一组图表是否一组图表? ? Is it a set of charts? z是否一个警告系统是否一个警告系统? ? Is it a warning system? z是否一个预防性方法是否一个预防性方法? ? Is it a prevent
8、ive method?SPC for Operators - Version 2.1 Mar. 10, 200111 使用使用SPCSPC目的目的Using Using SPC Objectives Our SPC objectives is to provide you with a tool for continual improvement that helps to: y将流程变异减致最少 minimize process variation. y消除流程的错败 eliminate failures in our processes. y将产品能力尽量提升 maximize produ
9、ct capability. y加强客户满意程度 increase customer satisfaction. y SPC for Operators - Version 2.1 Mar. 10, 200112 & 少报废少报废 Less scarp & 少返工少返工 Less rework & 少停机少停机 Less down time & 更佳利润更佳利润 Better profit & 更佳市场信誉更佳市场信誉 Good reputation in marketplace ! 惠亚集团惠亚集团Viasystems Group SPC for Operators - Version 2.
10、1 Mar. 10, 2001SPC for Operators - Version 2.1 Mar. 10, 200113 & 更佳产品质量更佳产品质量 Better quality product & 更多可靠产品更多可靠产品 More reliable delivery & 可以信赖的供应商可以信赖的供应商 A supplier they can count on客户客户 CustomersSPC for Operators - Version 2.1 Mar. 10, 2001SPC for Operators - Version 2.1 Mar. 10, 200114 & 少挫败感及
11、无聊时间少挫败感及无聊时间 Less frustration & down time & 更能控制工作更能控制工作 More control & 更能对流程及产品问题作正确决定更能对流程及产品问题作正确决定 Ability to make the right decision about process of product problems & 能生产更优值的产品能生产更优值的产品 Better product coming from the are in front of me你你 YouSPC for Operators - Version 2.1 Mar. 10, 200115 生产中
12、生产中 During Production z发生问题时(特别成因)SPC数据会发出讯号 SPC data send a signal that theres a problem (special cause). z我们采取纠正行动防止问题 We take corrective action to prevent defects.将来将来 For the Future zSPC数据提供一个流程变异的历史记录 SPC data provide a history of the variation in the process. z我们改善流程以减少一般原因导致的变异 We improve the
13、 process to reduce common cause variation.SPC for Operators - Version 2.1 Mar. 10, 200116 z收集数据-数目或量度某物 Collect data - count or measure something. z z于控制图上画上数值 Plot the data on the control chart. z 或. OR . . . z计算统计数值于控制图上画上数值 Calculate a statistic and plot the statistic on the chart.SPC for Operato
14、rs - Version 2.1 Mar. 10, 200117 z什么数据需要收集? What data do we need? z为什么需要收集数据? Why do we need to collect data? z如何得到数据? How will we get the data? z何时收集数据;多久收集一次? When do we collect it; how often? z流程中哪处收集数据? Where in the process do we get the data? z谁收集数据? Who collects the data?收集数据收集数据 Collect data
15、SPC for Operators - Version 2.1 Mar. 10, 200118 收集数据收集数据 Collect data 客户要求客户要求: : 功能功能; ;外观外观 Customer Requirements Functional Aesthetic 产品特性产品特性 Product Features 流程变数流程变数 Process Variables SPC for Operators - Version 2.1 Mar. 10, 200119 为什么需要收集数据为什么需要收集数据?Why should we collect the data?收集数据收集数据 Col
16、lect data 监控变异监控变异 To monitor variation. 尽早指出流程问题尽早指出流程问题 To identify process problems as early as possible.为什么为什么? ? Why? SPC for Operators - Version 2.1 Mar. 10, 200120 如何如何? ? How? 收集数据收集数据 Collect data z方法方法-技巧技巧, ,工具工具, ,经培训的员工经培训的员工 Method - technique, tool, trained people z记录记录-数据收集表格数据收集表格,
17、,表达数据的格式表达数据的格式( (即控制图即控制图) ) Record - data collection sheet, format to present data (i.e., control chart) z z计划计划-负责人负责人, ,收集数据地点收集数据地点, ,频率频率, ,时间时间 Plan - who, where, how often, whenSPC for Operators - Version 2.1 Mar. 10, 200121 那一个比较好那一个比较好? ? Which is better?流程中作测量流程中作测量 Measure on-line 最后检查最后
18、检查 Inspect at end NoImageNoImageSPC for Operators - Version 2.1 Mar. 10, 200122 流程中作测量流程中作测量 Measure On-line, During the Process SPC SPC 预防性 Prevention focus 避免浪费 Avoids waste 提供质量一致的产品 Delivers consistent products 减少变量 Reduces variation 主动性 Proactive 于流程终点作检查于流程终点作检查 Inspect Product at the End of t
19、he Process 检查 Inspection 侦察性 Detection focus 容忍浪费 Tolerates waste 返工 Reworks products 容忍变异 Tolerates variation 被动性被动性 Reactive SPC for Operators - Version 2.1 Mar. 10, 200123 计量的计量的 Variables z可用连续性的单位 Measure on a continuous scale z可以是流程或量度的特性 For a process or product 例子: Examples: z线宽 line width z
20、板料厚度 lamination thickness 计数的计数的 Attributes z数量,等级,类别 Count, grade, categorize z不能量度的 Cant measure it z一般是产品特性 Usually for a product 例子: Examples: z坏品百分比 % defective parts z甩货百分比 % orders late SPC for Operators - Version 2.1 Mar. 10, 200124 计量控制图计量控制图 Variables Charts 用于连续单位的量度用于连续单位的量度 For data we
21、get by measuring on a continuous scale. 例子 Examples 温度 temperature 线宽 circuit width 厚度 pin height分别监察两组统计数值的三对控制图分别监察两组统计数值的三对控制图Three pairs of charts that monitor two statistics.平均及范围平均及范围(X-R)(X-R)Average & R (average and range) 平均及标准差平均及标准差(X-S)(X-S)Average & S (average and standard deviation) 个别
22、值移动范围个别值移动范围(X-MR)(X-MR)X & MR (individuals and moving range)SPC for Operators - Version 2.1 Mar. 10, 200125 计数控制图计数控制图 Attribute Charts 用于数量等级或分类的数据用于数量等级或分类的数据 For data we get by counting, grading or categorizing things. 例子 Examples 甩货百分比 # of late deliveries 坏板百分比 # of defective circuit boards 计数
23、控制图计数控制图Attribute Charts NP P U CSPC for Operators - Version 2.1 Mar. 10, 200126 z客户客户, ,供应商或内部工程要求供应商或内部工程要求 Customer, supplier or internal engineering requirements z包括包括: : Include: y 目标或中间值目标或中间值( (最理想的最理想的) ) target or nominal value (the ideal!) y 在目标值两旁可接受的变异在目标值两旁可接受的变异 allowable variation on
24、either side of the target value z规格上限规格上限= =目标值目标值+ +可接爱的变异可接爱的变异 USL = target + allowable variation z规格下限规格下限= =目标值目标值- -可接受的变异可接受的变异 LSL = target - allowable variation SPC for Operators - Version 2.1 Mar. 10, 200127 规格是130克/升+/-30克升 Specification is 130 grams/liter +/- 30 grams/liter z规格上限=目标值+可接受
25、的变量 USL = target + allowable variation 130 + 30 = 160 z规格下限=目标值-可接受的变异 LSL = target - allowable variation 130 - 30 = 100 z可接受变异的总值=规格上限=规格下限 Total allowable variation = USL - LSL 160 - 100 = 60 SPC for Operators - Version 2.1 Mar. 10, 200128 z若流程受到控制: If your process is in control: y 99.7%的数据将会在控制上限
26、及下限之间. 99.7 % of the data points will be between the UCL and LCL. y 数据没有特定的图样=随机的. Data points will show no pattern = random. y 大部分数据会趋向平均值. Most data points will be close to the average. y 数据应有一半在平均值以上及一半在以下. Data points be 50/50 above and below average. SPC for Operators - Version 2.1 Mar. 10, 200
27、129 z减少两个可能的错误 Minimizes two possible mistakes: y 忽略问题 Ignoring a problem. y 采取不必要的纠正行动 Taking unnecessary corrective action. z曲线下的出现可能性已知 Probabilities under the curve are known. y 有利于预测和计划 Good for prediction and planning. y 有利于侦察不正常的数据 Good for spotting weird patterns of points. SPC for Operators
28、 - Version 2.1 Mar. 10, 200130 使用标准差决定控制上限及下限,如钟状曲线. To determine UCL & LCL, use standard deviation (s) - just like the bell curve! UCL = Average + 3s Average LCL = Average - 3s 利用变异的数值(标准差)去建立变异的一般范围 (控制上限及下限之间的范围)是否合理? Doesnt it make sense to use a measure of variation (standard deviation) to esta
29、blish a normal range of variation (area between UCL and LCL) for a process? SPC for Operators - Version 2.1 Mar. 10, 200131 Average -3s +3s -2s +2s -1s +1s LCL UCL -3s +3s -2s +2s -1s +1s LCL UCL Average -3s +3s -2s +2s -1s +1s LCL UCL Average 将钟状曲线转将钟状曲线转9090度度 Flip the bell curve on its side. SPC
30、for Operators - Version 2.1 Mar. 10, 200132 UCL=Average +3s LCL=Average - 3s -3s +3s -2s +2s -1s +1s Average LCL UCL 这是什么意思? What does it all mean? SPC for Operators - Version 2.1 Mar. 10, 200133 1.收集数据 Collect data 3. 理解数据 Interpret the data 2.在图上记录数据 Record data on chart 4.特别成因 Special Cause? 6.采取
31、纠正行动 Take corrective action 5.继续执行流程 Continue to run the process 是Yes否No 7. 准备文件 Document SPC for Operators - Version 2.1 Mar. 10, 200134 8 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 mils 是否有特别是否有特别问题发生问题发生? ? Did anything special happen? NoImage1.收集数据 Collect data 3.理
32、解数据 Interpret the data 2.在图上记录数据 Record data on chart 4.特别成因 Special Cause?SPC for Operators - Version 2.1 Mar. 10, 200135 一般原因导致的变异一般原因导致的变异 Common cause variation 特别原因导致的变异特别原因导致的变异 Special cause variation 8 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 mils UCL LCL Y轴-
33、量度 y-axis - scale for measurements X轴-时间 X-axis - time sequence SPC for Operators - Version 2.1 Mar. 10, 200136 z平均值(典形数值) zAverage (the typical value) z平均年龄=(27+ 27+33+46+57) 5 = 38 岁 zAverage age = (27+27+33+46+57) 5 = 38 years 27 27 33 46 57 Average age = 38SPC for Operators - Version 2.1 Mar. 10
34、, 200137 z范围(R)-最大数字减最小数字 Range (R) - highest number minus lowest number zR= (57 - 27) = 30 岁 R = (57 - 27) = 30 years幅度=30岁 Range = 30 years 27 27 33 46 57 SPC for Operators - Version 2.1 Mar. 10, 200138 z标准差(S)=个别数值与平均值相差的平方 Standard deviation (S) - How far away, on average, are the individual dat
35、a values from the average age?zS=13.2 岁 S = 13.2 yearsAverage = 38 11 years11 years19 years8 years5 years2727334657SPC for Operators - Version 2.1 Mar. 10, 200139 P Chart - VAAS Fastech Final Inspection Average LCL UCL 0 15000 30000 45000 60000 75000 90000 105000 120000 135000 150000 1 2 3 4 5 6 PPM
36、 SPC for Operators - Version 2.1 Mar. 10, 200140 稳定的流程稳定的流程 Stable Process z一致的 Consistent z可预测的 Predictable z特别原因 No special causes z受控的 In control 稳定性可用控制图量度 Stability is measured using control charts. 流程可以是受控流程可以是受控, ,但不符合规格的但不符合规格的. A process could be in control, but not meet specifications.? ?
37、SPC for Operators - Version 2.1 Mar. 10, 200141 完成一件工作的一连串步骤完成一件工作的一连串步骤 A series of steps we follow to accomplish a task. 步骤步骤Steps 成品成品 Outputs 输入输入 Inputs ? ? SPC for Operators - Version 2.1 Mar. 10, 200142 NoImageNoImageNoImageNoImageNoImage物料物料Materials 人人People 机器机器/ /设备设备Machines 方法方法Methods
38、环境环境 Environment NoImageSPC for Operators - Version 2.1 Mar. 10, 200143 为能做到优越的表现为能做到优越的表现, ,工作流程必须工作流程必须: :To achieve performance excellence, work processes must be: 稳定稳定 Stable z一致的一致的, ,可预计的可预计的 consistent, predictable 有能力有能力 Capable z持续符合规格持续符合规格 continually meets specifications 持续改善持续改善 Continu
39、ally improved惠亚集团品质政策惠亚集团品质政策 Viasystems Global Quality Policy 惠亚集团承诺通过优越的表现以惠亚集团承诺通过优越的表现以满足客户要求满足客户要求 Viasystems is committed to Customer Satisfaction through Performance Excellence.SPC for Operators - Version 2.1 Mar. 10, 200144 有能力的流程有能力的流程 Capable Process z流程有能力持续的满足规格要求 Process has the ability
40、 to continually meet specifications. 流程必须处于受控状态(统计技术角度),才可被认为有能力 Process must be in statistical control before it can be considered capable.SPC for Operators - Version 2.1 Mar. 10, 200145 z稳定的流程 Stable process y 受统计技术控制 in statistical control y 无特别原因导致的变异 no special causes of variation z z稳定流程的控制图 C
41、ontrol chart from a stable process y 无超越控制上限或下限的点 no points outside UCL or LCL y 无偏移或走势 no shifts or trends流程可以说是于一定时间内受到控制,而只有一般原因导致的变异 Such a process is said to be in control over a series of time, with only common causes present. LCL UCL CL SPC for Operators - Version 2.1 Mar. 10, 200146 个别成品之间不可
42、避免的分别 The inevitable differences among individual outputs of a process. 一般原因导致的变异Common Cause Variationy正常,常见,可预计的-自然的 normal, usual, expected things - natural特别原因导致的变异Special Cause Variation y偶发性,不可预计的-特定的 sporadic, unexpected things - assignable SPC for Operators - Version 2.1 Mar. 10, 200147 NoIm
43、ageNoImageNoImage物料变异物料变异: : 化学品浓度化学品浓度, ,混合比例及补料份量混合比例及补料份量, ,不同供应商不同供应商 Material Variation: chemical concentration, mix ratios & dosing; different suppliers 设备变异设备变异: : 保养时间保养时间, ,损耗损耗, ,调校过程调校过程 Machine Variation: maintenance schedules, wear & tear, calibration process 人为变异人为变异: : 技巧及经验技巧及经验, ,身材
44、及体能身材及体能, ,态度及纪律态度及纪律 People Variation: skill & experience, size & strength, attitude & discipline NoImageNoImagea 方法变异方法变异: : 程序文件程序文件, ,文件维持文件维持 Method Variation: documentation of procedures, maintenance of documentation 环境变异环境变异: : 时间时间, ,天气天气, ,厂房位置厂房位置, ,空气污染空气污染 Environmental Variation: time,
45、weather, location of plant, air contamination SPC for Operators - Version 2.1 Mar. 10, 200148 流程中太多一般原因导致的变异 Too much common cause variation in a process 流程中出现特别原因导致的变异 An occurrence of special cause variation in a processy 影响工作 disruption in work y 生产力下降 lost productivity y 报废及返工 scrap and rework y
46、 甩交货期 missed delivery datesy 不稳定的产品表现 variable product performance y 不稳定的产量 erratic product yields y 不稳定的产品特性 variations in product featuresAAAAAAAA AAAAAAA SPC for Operators - Version 2.1 Mar. 10, 200149 Cpk z根据规格评估流程表现. Evaluate process performance against specs. z决定流程是否能满足新规格要求. Determine if proc
47、ess can meet new specs. z根据Cpk及Cp值的距离制订改善目标. Set improvement goal based on distance between Cpk and Cp. z评估供应商. Evaluate suppliers. z用于销售过程的一部分. Use as part of sales process.SPC for Operators - Version 2.1 Mar. 10, 200150 z比较流程中间值设定及目标. Compares setting of process center (average) to target. z z基本定律
48、: Cpk要求最少1.33 General rule - aim for Cpk of 1.33 minimumUSL LSL 目标 Target 流程中间值 Process Center 偏离中间值 Off Center! SPC for Operators - Version 2.1 Mar. 10, 200151 Cpk良好 Good CpkCpk差 Poor CpkLSL USL LSL USL SPC for Operators - Version 2.1 Mar. 10, 200152 z比较可接受变异(A)及实际变异(B) Compare allowable variation
49、(B) to actual variation (A). z zB = USL - LSL z zA=6标准差 A = 6 standard deviations z zCp = B/ALSL USL A B SPC for Operators - Version 2.1 Mar. 10, 200153 显示于实际变异少于可接受变异 Indicates actual variation is less than allowable variation.LSL USL LSL USL LSL USL Cp大于1 Cp greater than 1Cp = 1Cp小于1 Cp less than
50、1SPC for Operators - Version 2.1 Mar. 10, 200154 z数值越大越好数值越大越好! ! The bigger the number, the better! zCpCp的缺点的缺点-只能显示流程分布只能显示流程分布, ,但不能显示是否符合规格但不能显示是否符合规格 Deficiency of Cp - It shows only the spread of process, but not the compliance to specification. z例子：两流程的Cp一样。 Example: Two processes with same C