6160sigma160bb黑帶培訓(xùn)資料-160160booster_wk1_03_msa(存儲(chǔ)版)
【正文】 IT help desk is investigating how well staff can recognise different types of calls for assistance. This exercise is based on a real case study in TRW. The case study is included as an appendix to this section. Data: C:\SixSigma\Data\ Conduct an attribute Gage Ramp。 Case Study . Measurement Systems Analysis: Reference Manual. ., 1990 (Available from . (313)3583570) Wheeler, Donald J. and Lyday, Richard W. Evaluating the Measurement Process, Second Edition. Knoxville, TN: SPC Press Inc. 1989. Discrete Measurement System Case Study A TRW case Study is included as an Appendix to this section. 53 APPENDIX: Attribute Gauge Ramp。R study. ?The RSquared value es from correlating the average Gage measurement for each part against the master gage measurement (note it is the average, not the individual values) ?The accuracy (or bias) of the gage is calculated as the mean deviation from the master measurement for all of the parts. Dividing this figure by the process variation gives %Bias. 41 Using Minitab to Look for Degree of Accuracy Use your notebook to follow along. 42 Data for a Gage Linearity Study Use your notebook to follow along. 43 Entering the Gage Linearity Information Use your notebook to follow along. 44 Gage Linearity Output Use your notebook to follow along. 45 Exercise 2: Responding to Measurement Problems Objective: To gain a better understanding of how to evaluate a Gage Ramp。R ANOVA Method Focus on the following: A) p value for the operator term p .05 implies that the operators get significantly different average results. B) p value for the operator unit number term p .05 implies that the operator to operator differences are not consistent across parts. C) Number of distinct categories If the number is 5 it implies that the measurement variation is too large to adequately distinguish the part to part variation. D) The R chart by operator If it is stable, this tells is that there are no special causes in the measurement process that could be throwing off our calculations 29 Interpreting the Gage Ramp。R ? Describes the variation of the measurement system in parison to the part variation of the process ? ? %P/T ? Describes the variation of the measurement system in parison to the part tolerances ? t ot als y s t e mtm e as ur e m e nSSRR _amp。R study pares several repeated measurements by different operators on several parts. ?Then Minitab uses Analysis of Variance (see Hypothesis Testing module) to separate out measurement Repeatability amp。Measurement System Validation Gage Ramp。R Problem ? We cannot measure p directly with the measurement system we have. ? The process variation is spread by the measurement error to t , where t = ( p 2 + m 2 ) (See Normal Module) ? m needs to be small pared with t or p . ?If we could measure exactly, each part would always look the same and the process variation would have a standard deviation p ?But if we measure any part with the system available, measurements on the same part are different. The standard error is m. 16 0 50 1000 . 0 00 . 0 10 . 0 20 . 0 30 . 0 40 . 0 50 . 0 6
教學(xué)課件相關(guān)推薦
鄂ICP備17016276號(hào)-1