基于似然比的指数加权滑动平均动态控制图  

The Adaptive EWMA Control Chart Based Likelihood Ratio

作　　者：尚云艳[1] Shang Yunyan(Department of Science,Xijing University,Xi'an 710123,China)

机构地区：[1]西京学院理学院,陕西西安710123

出　　处：《南开大学学报（自然科学版）》2021年第1期41-47,共7页Acta Scientiarum Naturalium Universitatis Nankaiensis

基　　金：陕西省教育厅专项科研项(20JK0967)。

摘　　要：基于似然比的控制图是质量控制的一种常用方法,为进一步增强对中小漂移的检测敏感度,运用变化样本容量的动态控制图的思想计算警告线,设计了基于似然比的变化样本容量的指数加权滑动平均动态控制图,用于检测过程均值.通过R软件模拟过程开始受控和失控两种情况下平均运行长度,并与似然比控制图和基于似然比的指数加权滑动平均控制图比较,模拟结果显示良好.该方法应用于实践,不仅可以提高检测的速度,还可以提高企业统计决策的精确度和可靠度.The control chart based on likelihood ratio test is always used in Quality Control.In order to increase the monitoring sensitivity for middle-small shift,the threshold limits are computed,and the EWMA control chart is imposed based likelihood ratio with adaptive sample sizes,which is used to moni-tor process mean.Then the average run length is simulated by Monte Carlo method for process in control and out control.In addition,LRT control chart,ELR control chart are compared with the EWMA control chart based likelihood ratio with adaptive sample sizes,the result turned better for VSS-ELR control chart.This method could increase the speed of the monitoring,and also could increase the accuracy and reliability for dcision,when it is applied to practice.

关 键 词：似然比 马尔科夫链 指数加权滑动平均控制图 

分 类 号：O213.1[理学—概率论与数理统计]