用于监控均值和方差的非参数变点模型  

Nonparametric Change Point Model Monitoring Mean and Variance

作　　者：周茂袁[1] 刘志敏 罗晓 Zhou Maoyuan;Liu Zhimin;Luo Xiao(Science College,Civil Aviation University of China, Tianjin 300300, China;Second Research Institute of Civil Aviation Administration of China,Chengdu 610041, China)

机构地区：[1]中国民航大学理学院,天津300300 [2]中国民航局第二研究所,四川成都610041

出　　处：《南开大学学报（自然科学版）》2019年第4期81-85,共5页Acta Scientiarum Naturalium Universitatis Nankaiensis

基　　金：国家自然科学基金(U1533203);中央高校基本科研业务费项目中国民航大学专项(3122017082);中国民航大学教育教学研究项目重点课题(CAUC-2016-C1-01)

摘　　要：统计过程控制被广泛应用于工业过程,服务领域和其他行业.尽管在特定过程的情况下参数控制图是有效的,但通常情况下都没有足够的信息确定过程分布,所以在此情况下非参数控制图是必要的.但大多非参数控制图要么是监控位置参数的,要么是监控尺度参数的,而不是既可以监控位置参数又可以监控尺度参数的.通过整合Wilcoxon秩和检验和Ansari-Bradley检验到变点模型,提出了一个新的非参数控制图,并且模拟结果也展现了该控制图在探测位置和尺度漂移时,是优于其他的非参数控制图的.Statistical process control are being widely used in many industrial processes,and healthcare problems.While parametric charts are useful in certain processes,there is often a lack of enough knowledge about the process distribution.So nonparametric charts are needed in such situations.Most nonparametric control charts are for monitoring mean or variance,rather than both mean and variance.This paper develops a new nonparametric control chart by integrating the Wilcoxon rank-sum test and AnsariBradley test into the change-point model.Simulation results show that the proposed control chart is superior to other nonparametric charts in detecting both mean and variance.

关 键 词：位置参数 尺度参数 统计过程控制 非参数 

分 类 号：O213.1[理学—概率论与数理统计]