MSA-SPC

DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016MSA及SPC应用知识培训DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016一.什么是MSAMSA是测量系统分析DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•MSA包含哪些因素：•人：测量人•机：测量仪器•物：被测对象•法：测量方法•环：测量环境DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016二.MSA与仪器校正的区别与联系区别:校正只针对仪器这个变量MSA针对五个变量(测量仪器\测量人\被测对象\测量方法\测量环境)DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016联系:仪器只有经过校正合格后才可以做MSA分析(校正不合格的仪器是不可以使用的,不使用的仪器也就无价值做MSA)DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:20161.测量汽车产品仪器所构成的测量系统----MSAISO/TS169492.测量非汽车产品的仪器-----校正ISO9001DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•测量系统（人、机、物、法、环）中的任何一因素改变，需重新进行MSA分析DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:20161。测量特殊特性的测量系统：一定要做MSA分析。2。测量其它特性（控制计划）的测量系统：可以视资源进行MSA分析DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016三.MSA分析方法1.计数型的测量系统MSA的方法:小样法2.计量型的测量系统MSA的方法:五大性分析(重复性\再现性\线性\稳定性\偏倚性)DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016四.MSA的基本概念Bias(偏倚)-測量的觀察平均值和基準值之間的差異DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•Repeatability(重覆性)•由一位評价人多次用一種測量儀器測量同一零件的同一特性時所得的測量變差DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•Reproducibility(再現性)•-由不同的評价人使用同一個量具測量一個零件的一個特性的產生的測量平均值的變差DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016Stability(穩定性)是测量系统在某持续时间内测量同一基准或者零件的单一特性时获得的测量值总变差。DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•Linearity(線性)•是在量具预期的工作量程内，偏倚值的差值DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016五.MSA计算方法详见软件,无需详述DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016六.MSA结论判断•oStability穩定性•-沒有點子超過控制圖的界限•oBias偏倚•-如果’0’落在置信區間以內(-x•0x),所得出的偏倚是可接•受DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016oLinearity線性-偏倚=0’線必須完全在拟合线和置信帶以內Y=KX+BoRepeatability&Reproducibility重复性及再現性-少於10%完全可以接受-10%到30%之间,估计可行后接受-大于30%,不可以接受-ndc大于等于5o小样法GO&NG完全一至DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•1。只有五大性都是OK的，才能证明这个测量系统是OK的。•2。MSA做完后应MSA标签（明确人机物法环）•3。MSA的研究频次通常与仪校一至（在人机物法环没有变动的情况下）DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016七.什么是SPC统计过程控制DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016八.在什么情况下应用SPC1.有特殊特性的位置一定要用SPC控制2.其它位置也可以视资源情况应用SPC控制,但没有强制要求DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016九.SPC控制图的分类1.对于计数型的特殊特性的SPC控制控制图种类用途P图用以监视过程不良品的比率P图用以监视过程不良品的数目U图用以监视每个单位产品的平均缺陷数C图用以监视过程缺陷的数目nDQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:20162.对于计量型特殊特性的SPC控制2.1单值移动极差图主要用于监测过程这种特殊特性(一次只能取一个样本,而且只对样本做一次测量,不可以重复测量的)2.2均值极差图主要用于监测产品这种特殊特性(一次取几个样本,而且可对组样本做测量,算出均值)DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016十.SPC的统计SPC的统计详见软件,无需讲述DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016常见的术语•USL：规格上限（客户给的）•LSL：规格下限（客户给的）•SL：规格中值（客户给的）•UCL：控制上限（统计出来的）•LCL：控制下限（统计出来的）•CL：控制中值（统计出来的）DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•X均值：一般一个子组取2-5个样本（值），计算平均数。•计算CPK、PPK的时候，至少要有100个样本（值）。DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11QualityManagementSystem–IATF16949:2016•UCL一定在USL的下方•LCL一定在LSL的上方•也就是说UCL与LCL一定要在USL与LSL的中间。•超出UCL但没有超出USL的点（值）的产品，依然是合格品，但该点要分析。•超出LCL但没有超出LSL的点（值）的产品，依然是合格品，但该点要分析。DQS-ULManagementSystemsSolutions©ULManagementSystemsSolutionsTrainingIATF16949:20162016.11Qualit