A critique of the use of modified and moving average‐based EWMA control charts

Abstract

We show that the so-called modified exponentially weighted moving average (EWMA) control chart is very closely related to a moving average-based EWMA chart. These charts have ill-advised weighting functions that assign more weight to some past data values than to the most recent data values. We show that these weighting functions adversely affect their run length properties and that one can obtain better overall performance with an ordinary EWMA chart.

A critical note on memory‐type charts with the Brownian statistic

“…In addition, these charts offer no significant advantages over standard methods. In similar vein, Haq and Woodall 9 have shown that the modified EWMA, EWMA and DEWMA based on MA, and MA‐based EWMA and DEWMA charts also have poor SS ARL performances. In addition, an ordinary EWMA chart may outperform these charts when detecting mean shifts that have more practical importance 10 …”

On updating the results of the control chart using a modified EWMA statistic

“…The chart proposed by Khan et al., 1 in fact, has better performance than the EWMA chart as measured by zero‐state ARL values. Haq and Woodall 4 showed, however, that if the process change occurs at any time other than at the very onset of monitoring, the EWMA chart performs better. The modified EWMA chart is a weighted average method that gives higher weights to some past data values than to the most recent data value.…”

Beyond the EWMA chart: Critiquing hybrid approaches

“…Importantly, they do not offer significant advantages over standard control methods. In similar vein, Haq and Woodall 15 showed that the classical EWMA chart can outperform the mEWMA, MA–EWMA (similar to mEWMA), EWMA–MA (MA chart with EWMA statistic as an input), DMA–EWMA and EWMA–DMA charts as the latter charts assign more weights to past data values than the current ones.…”

“…In this article, getting motivation from the work of Haq and Woodall, 15 for the EWMA, DmEWMA, mEWMA–MA, MA–mEWMA, mEWMA–EWMA, and EWMA–mEWMA charts, the weighting patterns are derived. Unlike the EWMA chart, these charts employ less favorable weighting patterns.…”