Minitab常用操作说明

ConfidentialPage:1/43Minitab常用操作說明Revision:APreparedby:蔡培奪Date:12/1‘2006ConfidentialPage:2/43內容：1.產品尺寸評估方式說明2.客訴報告中常用操作(2-samplet）3.GaugeR&R評估說明ConfidentialPage:3/431.以Boxplot確認量測值內是否有離散值2.以NormalityTest確認量測值是否呈常態分配3.以1-sampleT檢定量測值是否與目標值相符4.以CapabilityAnalysis來確認Cpk&SigmaLevel1.產品尺寸評估方式說明ConfidentialPage:4/431.1.Boxplot離散值確認Measure3.1623.1603.1583.1563.1543.1523.150BoxplotofMeasuremeasure4.2304.2254.2204.2154.2104.205Boxplotofmeasure有離散值無離散值•GraghBoxplotOneY/SimpleConfidentialPage:5/431.1.Boxplot離散值確認從Boxplot確認是否有Outlier,若有的話需再確認其是否已outspec(有無異常).ConfidentialPage:6/431.2.量測值是否呈常態分配確認•StatBasicStatisticNormalitytestConfidentialPage:7/43MeasurePercent3.1643.1623.1603.1583.1563.1543.1523.150999590807060504030201051Mean0.1253.156StDev0.002893N32AD0.574P-ValueProbabilityPlotofMeasureNormalmeasurePercent4.2304.2254.2204.2154.2104.2054.200999590807060504030201051Mean0.0054.217StDev0.005622N32AD1.572P-ValueProbabilityPlotofmeasureNormalP-value0.05,量測值呈“常態分配”P-value0.05,量測值非“常態分配”通過P-value是否大於0.05來判定是否為常態分布.ConfidentialPage:8/431.3.以1-sampleT檢定量測值是否與目標值相符此部分填入目標值(Nominalvalue)•StatBasicStatistics1-SampletConfidentialPage:9/43確認P-value是否0.05,若其0.05即表示其接受虛無假設Ho,也就是量測值等於目標值;若P-value0.05即表示拒絕虛無假設而接受對立假設,也就是量測值不等於目標值.ConfidentialPage:10/431.4.以CapabilityAnalysis來確認Cpk&SigmaLevel首先需先確認其是否為常態分配(同1.2量測值是否呈常態分配確認）MeasurePercent3.1643.1623.1603.1583.1563.1543.1523.150999590807060504030201051Mean0.1253.156StDev0.002893N32AD0.574P-ValueProbabilityPlotofMeasureNormalmeasurePercent4.2304.2254.2204.2154.2104.2054.200999590807060504030201051Mean0.0054.217StDev0.005622N32AD1.572P-ValueProbabilityPlotofmeasureNormalConfidentialPage:11/431.4.1.當量測值為常態分配•StatQualityToolsCapabilityAnalysisNormalConfidentialPage:12/43由於我們為一個lot收集32個量測值而非不同lot進行收集,因此於Subgroupsize部分填入32.分別填入既定的尺寸上下限值ConfidentialPage:13/43ConfidentialPage:14/430.7200.7080.6960.6840.6720.6600.648LSLUSLProcessDataSampleN32StDev(Within)0.00895598StDev(Overall)0.00895598LSL0.64Target*USL0.72SampleMean0.683906Potential(Within)CapabilityOverallCapabilityZ.Bench4.03Z.LSL4.90Z.USL4.03Ppk1.34CpmZ.Bench*4.03Z.LSL4.90Z.USL4.03Cpk1.34ObservedPerformancePPMLSL0.00PPMUSL0.00PPMTotal0.00Exp.WithinPerformancePPMLSL0.47PPMUSL27.87PPMTotal28.35Exp.OverallPerformancePPMLSL0.47PPMUSL27.87PPMTotal28.35WithinOverallProcessCapabilityofC1可分別得到Cpk&ZBench•若數據為短期：CPK=1.34sigmalever=Z.Bench=4.93•若數據為長期：CPK=1.34sigmalever=Z.Bench+1.5=6.43ConfidentialPage:15/43ConfidentialPage:16/431.4.2.當量測值為非常態分配填入組內量測值數量•StatControlChartsBox-CoxTransformticConfidentialPage:17/43LambdaStDev5.02.50.0-2.5-5.00.0059750.0059500.0059250.0059000.0058750.005850Lambda-5.00(using95.0%confidence)Estimate-5.00LowerCL*UpperCL*RoundedValueBox-CoxPlotofC1正態轉換係數Lambda=-5.00ConfidentialPage:18/43將之前得到的正態轉換係數填入即可,其他步驟則與量測值為常態分配時一樣•StatQualityToolsCapabilityAnalysisNormalBoxCoxConfidentialPage:19/430.04680.04560.04440.04320.04200.0408LSL*USL*transformeddataProcessDataSampleN32StDev(Within)0.00591975StDev(Overall)0.00591975AfterTransformationLSL*0.0474145Target*LSL*USL*0.0403861SampleMean*0.0430389StDev(Within)*0.0006708StDev(Overall)*0.00067081.84Target*USL1.9SampleMean1.87603Potential(Within)CapabilityOverallCapabilityZ.Bench3.95Z.LSL6.52Z.USL3.95Ppk1.32CpmZ.Bench*3.95Z.LSL6.52Z.USL3.95Cpk1.32ObservedPerformancePPMLSL0.00PPMUSL0.00PPMTotal0.00Exp.WithinPerformancePPMLSL*0.00PPMUSL*38.33PPMTotal38.33Exp.OverallPerformancePPMLSL*0.00PPMUSL*38.33PPMTotal38.33WithinOverallProcessCapabilityofMeaUsingBox-CoxTransformationWithLambda=-5•若數據為短期：CPK=1.32sigmalever=Z.Bench=3.95•若數據為長期：CPK=1.32sigmalever=Z.Bench+1.5=5.45ConfidentialPage:20/43分別填入Cpk&SigmaLevel(OverallCapabilityZBench+1.5)最後將minitab分析結果做綜合判斷ConfidentialPage:21/432.CAR報告中操作說明.2.11-sample的數據1-sample的數據分析同item1Boxplot離散性確認NormaltestCPK確認ConfidentialPage:22/432.22-sample的數據2-sample的數據分析如下Boxplot離散性確認Normaltest標准差檢定標准差檢定差異分析ConfidentialPage:23/432.2.1samplesize確定ConfidentialPage:24/432.2.2Boxplot離散性確認Measure3.1623.1603.1583.1563.1543.1523.150BoxplotofMeasuremeasure4.2304.2254.2204.2154.2104.205Boxplotofmeasure有離散值無離散值•GraghBoxplotOneY/withgroupConfidentialPage:25/432.2.2NormalTest(同item1.2)•StatBasicStatisticNormalitytestConfidentialPage:26/432.2.3標准差檢定ConfidentialPage:27/43ConfidentialPage:28/432.2.4平均值檢定ConfidentialPage:29/43ConfidentialPage:30/433GaugeR&R-計數型3.1基本概念ConfidentialPage:31/433.2准確性及線結評估ConfidentialPage:32/43ConfidentialPage:33/43ConfidentialPage:34/433.3重復性及再現性評估ConfidentialPage:35/43ConfidentialPage:36/43ConfidentialPage:37/43ConfidentialPage:38/43ConfidentialPage:39/43◆Keymessage:量測人員之間無明顯量測差異,不同sample之間有明顯差異.顯示量測系統無異常.3.4GaugeR&R實例例:A,B,C3個量測員分別針對10個產品進行量測，量測設備為QV，重復量測次數2次，所得數據如下，針對量測系統進行分析:3.4.1准確性及線性評估ConfidentialPage:40/433.4.2GageR&R--重復性及再現性分析StaticQualitytoolsGaugestudyGaugeR&R(crossed)ConfidentialPage:41/43PARTP-value小於0.05,顯示樣品為影響量測變異的顯著因子.量測系統造成之變異占總變異的3.29%9%,判定為可接受.辯別率=75此量測系統可接受%SV為18.13%30%,,判定為可接受.3.4.2GageR&R--重復性及再現性分析ConfidentialPage:42/433.4.2GageR&R--重復性及再現性分析→從GageR&R和ParttoPart所占比例可以看出主要變異來自樣品本身.→從Chart