An EWMA for Monitoring a Process Standard Deviation

TLDR

EWMA monitoring has mainly focused on mean shifts, yet detecting increases in process variability is arguably more critical for product quality. This study proposes an EWMA based on the log transformation of the sample variance to monitor process standard deviation. The authors present properties of this log‑variance EWMA and an optimal design strategy. The log‑variance EWMA outperforms traditional range or s² charts by detecting small increases in normal process standard deviation more quickly.

Abstract

Most applications of the exponentially weighted moving average (EWMA) for process monitoring have concentrated on the problem of detecting shifts in the mean level of a process. Perhaps more important is the problem of detecting increases in process variability, which can also have a major impact on product quality. In this paper we propose using an EWMA based on the log transformation of the sample variance. Properties of this EWMA are given and an optimal design strategy is presented. It is shown that the EWMA is superior to the usual range chart or s2 chart in terms of its ability to quickly detect small increases in the standard deviation of a normal process.