A Comparative Analysis of Robust Moving Average Control Charts for Process Dispersion

Abstract

In this paper, a comparative analysis of alternative methods for the moving average (MA) control chart for dispersion is developed using robust estimators. To compare the ability and performance of the existing moving average (MA) control charts for dispersion based on the sample standard deviation (S) and the proposed alternative methods based on robust estimators to detect shifts in a process, a Monte Carlo simulation study is used. It is observed from the results of the simulation study that the proposed robust alternative methods are effective in determining small shifts in the process and gives better performance as compared to the existing moving average (MA) control charts for dispersion, i.e. it provides swift indication about shifts in a process. An application numerical example with a real data set is used to illustrate the application and implementation of the control charts considered in this study which also supported the findings of the simulation study to some extent.