Minitab使用教学

2019/8/2012019/8/202Minitab使用教學2019/8/203目錄0.平均數與標準差圖示1.平均數與標準差2.常態檢定3.直方圖4.箱型圖6.RunChart5.二項分配2019/8/204(u)MEAN(群體平均)TIGHTDISTRIBUTION(緊密分配)(σ很小)BROADDISTRIBUTION(σ很大)變異-3σ-2σ-1σTARGET+1σ+2σ+3σσ：是測量離勢或變異(變異情形通常以平均值μ和標準差σ來表示品質特性值之位置及分散情形)其值愈小愈趨中心分配，品質變異越小越穩定，亦即是平均值愈接近目標值。反之，σ愈大分配離中心值愈遠品質愈壞。NXΣXXnxΣNX2)(μΣ1)(2nXXΣ+∞-∞μ:母群體平均數(趨中勢)(=):樣本平均數(品質特性平均值μ之估計值)(=)σ:(離中趨勢)標準差(母群體)(σ=)S:樣本標準差(品質特性σ之估計值)(S=)常態分配機率平均數及標準差2019/8/2051.平均數&標準差A.選擇StatBasicStatisticsDisplayDescriptiveStatistics.2019/8/2061.平均數&標準差B.在Variable選入”C1”。檢查ByVariable選入C2C.選擇OK2019/8/207平均數標準差2019/8/2082.常態分配檢定A.選擇StatBasicstatisticsNormalityTest.B.在Variable選入”Height”。C.選擇OKHeightPercent807570656099.99995908070605040302010510.1Mean0.05068.72StDev3.659N92AD0.748P-ValueProbabilityPlotofHeightNormalPValue需大於0.052019/8/2093.直方圖-1A.選擇GraphHistorgramWithFit.B.在GraphVariable選入”Height”。C.選擇OKHeightFrequency7572696663601086420Mean68.72StDev3.659N92HistogramofHeightNormal可看到平均數、標準差2019/8/20103.直方圖-2-1A.選擇StatQualityToolsCapabilityAnalysisNormal.2019/8/20113.直方圖-2-2B.如下圖輸入。輸入要計算的數據行輸入Group的大小值(建議值：5)2019/8/20123.直方圖-2-3C.在Lowerspec/UpperspecKey入規格值。輸入上下限規格值2019/8/20133.直方圖-2-4D.進入Option在”Key入中心規格按OKOK規格中心值或目標值2019/8/2014360350340330320310300LSLTargetUSLProcessDataSample?N39StDev(Within)5.12353StDev(Overall)4.97411LSL300Target330USL360SampleMean309.282Potential(Within)CapabilityCCpk1.95OverallCapabilityPp2.01PPL0.62PPU3.40PpkCp0.62Cpm0.461.95CPL0.60CPU3.30Cpk0.60ObservedPerformancePPM??LSL0.00PPM??USL0.00PPM?Total0.00Exp.WithinPerformancePPM??LSL35020.06PPM??USL0.00PPM?Total35020.06Exp.OverallPerformancePPM??LSL31015.49PPM??USL0.00PPM?Total31015.49WithinOverallProcessCapabilityofC13.直方圖-2-5估計驗收組內變異整個系統有持續性系統改善組間變異2019/8/20154.箱型圖A.選擇GraphBoxplot2019/8/20164.箱型圖B.依需求選擇所要的箱型圖模式2019/8/20174.箱型圖C.在GraphVariables輸入要分析的數據列OK2019/8/20184.箱型圖D.圖形如下，有一異常點C1340330320310300BoxplotofC1異常點2019/8/20195.二項式分配異常點A.選擇CalcProbabilityDistributionBinomial.2019/8/20205.二項式分配B.如為機率分配選擇Probability輸入樣本數輸入機率輸入要求數2019/8/20215.二項式分配C.答案如下：2019/8/20225.二項式分配D.如為累積分配選擇Cumulativeprobability2019/8/20235.二項式分配E.結果如下：2019/8/20246.RunChartA.選擇SataQualitytoolRunchart2019/8/20256.RunChartB.如下輸入2019/8/20266.RunChartC.結果如下圖：SampleC135302520151051340330320310300Numberofrunsaboutmedian:0.2866820Expectednumberofruns:18.37143Longestrunaboutmedian:4ApproxP-ValueforClustering:0.71332ApproxP-ValueforMixtures:Numberofrunsupordown:0.7948621Expectednumberofruns:23.00000Longestrunupordown:3ApproxP-ValueforTrends:0.20514ApproxP-ValueforOscillation:RunChartofC12019/8/20277.GageR&R