1、JMP TRAINING JMP是由SAS公司开发的统计分析软件。SAS公司是一家提供软件和服务的公司,总部设于美国北卡罗莱纳州。目前JMP的最新版本是:.。我司现用的版本是5.1。JMP 软件简介:JMP功能简介:JMP可以提供一个图形界面来显示与分析数据,它具备以下功能:在数据表单上可以查看、编辑、输入与处理数据;强大而图形化的统计技术数据分析功能 表单的排序和合并;快捷的数据分组与统计计算功能;质量提升技术如过程能力等;分析结果输出.JMP操作界面:关于JMP基本介绍请参阅JMP自带的帮助文件。JMP基本操作:操作界面JMP基本操作:操作界面定义数据列名称与类型:双击列上部或选择col/
2、col info Continuous 定义可连续的数值型数据,如铜厚、介质层厚度等。Ordinal 定义有顺序的数值型或字符型数据,如lot no。Nominal 定义分类(分级)的数值型或字符型数据,如缺陷名称。数据导入:从EXCEL表格中COPY即可。JMP应用一:柱状图和饼图。应用一:柱状图和饼图。JMP应用二:柏拉图。应用二:柏拉图。JMP应用三:直方图,计算应用三:直方图,计算Cpk。JMP应用四:控制图。应用四:控制图。JMP应用五:应用五:Analysis of variance常用工具:柱状图和饼图:柏拉图:1,导入数据3,导入相应位置2,选择图表4,生成图表,适当编辑Wei
3、ght:quantiry(p)0255075100Percent42.8553.9164.8471.7877.3781.7285.8489.7593.3995.0796.5497.4198.2598.6598.9999.3499.6099.6999.7799.8699.93绝缘层缺陷超标切片分析资料设计错误电镀折板渗镀翘曲度漏镀偏位玻纤电镀铜厚超标蚀刻不净残铜夹膜图形折板退膜折板板面金属掉线路折板漏镍棕化不良defect0255075100Cum PercentPareto PlotPareto Plot柏拉图:Weight:quantiry(p)0255075100Percent42.855
4、3.9164.8471.7877.3781.7285.8489.7593.3995.0796.5497.4198.2598.6598.9999.3499.6099.6999.7799.8699.93绝缘层缺陷超标切片分析资料设计错误电镀折板渗镀翘曲度漏镀偏位玻纤电镀铜厚超标蚀刻不净残铜夹膜图形折板退膜折板板面金属掉线路折板漏镍棕化不良defect0255075100Cum PercentP Pa ar re et to o P Pl lo ot t单个缺陷占总缺陷的百分比。直方图:1,导入数据2,选择图表3,生成图表3,STACK 功能2530354045505560100.0%99.5%97
5、.5%90.0%75.0%50.0%25.0%10.0%2.5%0.5%0.0%maximumquartilemedianquartileminimum 57.900 57.900 57.900 51.100 47.400 42.600 37.700 32.980 29.200 29.200 29.200Q Qu ua an nt ti il le es sMeanStd DevStd Err Meanupper 95%Meanlower 95%MeanN42.7891896.69082221.0999644 45.0200240.558358 37M Mo om me en nt ts s直
6、方图:Cpk计算1,右键选择2,输入目标值3,算出CPKLower Spec LimitUpper Spec LimitSpec TargetSpecification 30 50 40Value Below LSLAbove USLTotal OutsidePortion 2.7027 18.9189 21.6216%ActualLSLUSLTarget-3s+3sMean2030405060 CPCPKCPMCPLCPUCapability 0.498 0.359 0.460 0.637 0.359Index 0.384 0.224 0.362 0.454 0.222Lower CI 0.
7、613 0.495 0.558 0.817 0.493Upper CIBelow LSLAbove USLTotal OutsidePortion 2.7973 14.0580 16.8553Percent27973.225 140580.1168553.33PPM O Ov ve er ra al ll l,S Si ig gm ma a =6 6.6 69 90 08 82 2C Ca ap pa ab bi il li it ty y A An na al ly ys si is s直方图:多组数据同时分析直方图:正态分布拟合曲线正态分布函数样本数量直方图:Lower Spec Limi
8、tUpper Spec LimitSpec TargetSpecification 30 50 40Value Below LSLAbove USLTotal OutsidePortion 2.7027 18.9189 21.6216%ActualLSLUSLTarget-3s+3sMean2030405060 CPCPKCPMCPLCPUCapability 0.498 0.359 0.460 0.637 0.359Index 0.384 0.224 0.362 0.454 0.222Lower CI 0.613 0.495 0.558 0.817 0.493Upper CIBelow LS
9、LAbove USLTotal OutsidePortion 2.7973 14.0580 16.8553Percent27973.225 140580.1168553.33PPM O Ov ve er ra al ll l,S Si ig gm ma a =6 6.6 69 90 08 82 2C Ca ap pa ab bi il li it ty y A An na al ly ys si is s规格上下限和目标值控制图:75808590Mean of D136912151821242730SampleAvg=80.67LCL=75.09UCL=86.26Note:Sigma used
10、 for limits based on range.X X B B a a r r o o f f D D 1 1-50510152025Range of D136912151821242730SampleAvg=9.67LCL=0.00UCL=20.46R R o o f f D D 1 1V V a a r r i i a a b b l l e e s s C C o o n n t t r r o o l l C C h h a a r r t t控制图:75808590Mean of D1CBACBA36912151821242730SampleAvg=80.67LCL=75.09
11、UCL=86.26Note:Sigma used for limits based on range.X XB Ba ar r o of f D D1 1-50510152025Range of D136912151821242730SampleAvg=9.67LCL=0.00UCL=20.46R R o of f D D1 1V Va ar ri ia ab bl le es s C Co on nt tr ro ol l C Ch ha ar rt t控制图分区:A,B,C区右键单击此处:控制图:CpkSelect:capability输入目标值CPK控制图:控制图:75808590Mea
12、n of D1CBACBA36912151821242730SampleAvg=80.67LCL=75.09UCL=86.26Note:Sigma used for limits based on range.X XB Ba ar r o of f D D1 1-50510152025Range of D136912151821242730SampleAvg=9.67LCL=0.00UCL=20.46R R o of f D D1 1根据SPC规定判定过程是否受控。方差分析:One way ANOVA输入数据选择stackStack OK数据STACK方差分析:One way ANOVA数据STACK方差分析:One way ANOVA数据STACK方差分析:One way ANOVA数据STACKP 值很重要,当P值大于0.05时,在统计学上我们认为两者没有显著性差异,当P值小于0.05时,统计学上认为有显著性差异;结合本例,因为P值小于0.0001,所以我们认为两者有显著性差异,即棕化药水30克/L的CU 含量对WB 影响很大。Thank You!