【正文】 的指標(biāo) The %RR is the best measure for the Black Belt ?對于整體流程變差而言，此估計(jì)可以用于評估測量系統(tǒng)的表現(xiàn) This estimates how well the Measurement System performs with respect to the overall process variation ? %RR是進(jìn)行流程改進(jìn)研究的一個(gè)很好的估計(jì) %RR is the best estimate when performing process improvement studies 當(dāng)使用樣本來反映完整的，具有代表性的流程變差時(shí)，務(wù)必要小心 Care must be taken to use samples representing full, but typical, process variation 019 MSA17 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. ? 通常需要 2－ 3名操作員 Generally 2 to 3 operators ? 通常需要對 10個(gè)單位的對象進(jìn)行測量 Generally 10 units to measure ? 每一個(gè)單位被每一個(gè)操作者測量 2－ 3次 Each unit is measured 23 times by each operator Gage RR 分析的設(shè)定 Gage RR Study Setup 019 MSA18 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. 研究樣本的選擇 Study Sample Selection ? 樣本應(yīng)從流程正常變差的范圍中取得 Samples that span the normal variation of the process should be pulled from the process 舉例： 如果生產(chǎn)一種平均厚度為 “ sigma為”的材料，則樣本厚度范圍應(yīng)為 （ 95％ 的范圍延伸） Example: If you produce a material with a mean thickness of ” and a sigma of ”, use samples that go from ” thick (95% of the range) 019 MSA19 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. 研究樣本的選擇 Study Sample Selection ? 注意 BE CAREFUL! ?如果同一流程生產(chǎn)不同厚度的材料，那就將他們分組并進(jìn)行RR研究 If you produce different thickness’ of material with the same process, subgroup them and perform the RR study ?舉例 Example: ? 一個(gè)流程生產(chǎn)厚度為 ”, ”, and ” 的材料并且公差是 +/ ，所有數(shù)值均由同一系統(tǒng)測量所得 A process produces ”, ”, and ” materials all with a +/ ” tolerance, all measured with the same system ? 執(zhí)行 3次研究 每個(gè)厚度一次 Perform 3 studies one for each thickness ?如果你將上述樣本混合在一起， GRR的值就被人為的調(diào)低了 If you lumped the above samples together, the GRR value will be artificially low 019 MSA20 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. 執(zhí)行 RR 研究的步驟 Procedure for Performing RR Study ? 校準(zhǔn)量具或者確認(rèn)它們已經(jīng)被校準(zhǔn)過 Calibrate the gage, or ensure that it has been calibrated ? 在 Minitab中建立數(shù)據(jù)矩陣，并確認(rèn)它們已經(jīng)被隨機(jī)化 Set up your data matrix in Minitab. Be sure to randomize ? 讓操作員以隨機(jī)順序測量樣本 Have operators measure the samples in random order ? 在 Minitab中分析數(shù)據(jù)來確定 RR的統(tǒng)計(jì)值 Analyze data in Minitab to determine the statistics of RR study ? 重復(fù)性 Repeatability ? 再現(xiàn)性 Reproducibility ? 上述之標(biāo)準(zhǔn)差 Standard deviations of each of the above ? % RR ? % P/T ? 分析結(jié)果并確定進(jìn)一步行動 Analyze results and determine followup action, if any 019 MSA21 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. 樣本大小的問題 Sample Size Issues ? 操作員的數(shù)量 Number of operators ? 如果流程使用多個(gè)操作員，隨機(jī)選擇 2－ 4個(gè) If process uses multiple operators, choose 24 at random ? 如果流程只有一個(gè)操作員或者沒有操作員，則分析時(shí)忽略操作員的影 響（即忽略再現(xiàn)性的影響） If process uses only one operator, or no operators, perform study without operator effects (reproducibility effects ignored) ? 樣本的數(shù)量 Number of samples ? 選擇足夠的樣本，使得（樣本數(shù)量） x （操作員數(shù)量） 15 Select enough samples so that (number of samples) X (number of operators) 15 ? 如果不可行，選擇測試的次數(shù)使得： If not practical or possible, choose number of trials so that: S 樣本數(shù)量 O 操作者數(shù)量 ? 測試次數(shù)＝ 3 if S x O 15, trials = 3 ? 測試次數(shù)＝ 3 到 4 if S x O 8, trials = 3 to 4 ? 測試次數(shù)＝ 4 到 5 if S x O 5, trials = 4 to 5 ? 測試次數(shù)＝ 6 到 8 if S x O 4, trials = 6 to 8 019 MSA22 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. Open the worksheet Gageaiag Part Operator Response 1 1 1 1 2 1 2 1 3 1 3 1 … … ... 測量系統(tǒng)分析練習(xí) MSA Exercise 019 MSA23 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. Minitab – 量具重復(fù)性和再現(xiàn)性研究 Minitab Gage RR Studies Stat Quality Tools Gage Study Gage RR Study (Crossed) 019 MSA24 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc. Used with permission. Copyright 2023 Cummins, Inc. All Rights Reserved. Minitab – 量具重復(fù)性和再現(xiàn)性研究 ?Minitab Gage RR Studies 量具的選項(xiàng)： 如果有公差， 填入此處 GAGE Options If There Is A Tolerance For This Part Put It Here 研究的概要信息 General Info On Your Study 放入 (AIAG 標(biāo)準(zhǔn)） Set to (AIAG Standard) 019 MSA25 Effective Date: 2/2023 Copyright 20232023 Sigma Breakthrough Technologies, Inc.