摘要
多元指数加权移动平均(MEWMA)控制图常用来监测多元质量过程中存在的小偏移,对其性能评价时多采用平均运行链长。在马尔可夫链的MEWMA控制图平均运行链长计算方法的基础上,分析了MEWMA控制图运行链长的概率分布及累积概率分布。基于MATLAB计算平台,编制了相应的计算程序,给出了MEWMA控制图运行链长的概率分布曲线及累积概率曲线。提出了MEWMA控制图参数优化方法,对电子产品装配中的表面贴片工艺过程的焊点质量控制进行了研究。
Multivariate exponentially weighted moving average （MEWMA） control chart was designed to detect small shifts in the mean vector of a multivariate quality process; its performance was evaluated by the average run length （ARL）. Based on the method for ARL computation proposed by Runger and Prabhu, this paper studied the run length distributions of MEWMA control chart, including the run length probability distributions and cumulative probability distributions. Based on the procedure for parametrical optimization of MEWMA control chart, a case study of its application in surface mount process for the solder joint quality control was presented.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2007年第7期820-824,共5页
China Mechanical Engineering
基金
国家自然科学基金资助项目(50475005
70321001)
关键词
质量控制
统计过程控制
多变量控制系统
马尔可夫过程
优化控制系统
quality control
statistical process control
multivariable control system
Markov process
optimal control system
