Increasing the sensitivity of variability EWMA control charts

Abstract

Control chart is the most important statistical process control (SPC) tool used to monitor reliability and performance of manufacturing processes. Variability EWMA charts are widely used for the detection of small shifts in process dispersion. For ease in computation all the variability EWMA charts proposed so far are based on asymptotic nature of control limits. It has been shown in this study that quick detection of initial out-of-control conditions can be achieved by using exact or time varying control limits. Moreover the effect of fast initial response (FIR) feature, to further increase the sensitivity of variability EWMA charts for detecting process shifts, has not been studied so far in SPC literature. It has been observed that FIR based variability EWMA chart is more sensitive to detect process shifts than the variability charts based on time varying or asymptotic control limits. © 2011 Springer Science+Business Media B.V.