Design of exponentially weighted moving average chart for monitoring standardized process variance

Abstract

Control charts for monitoring of process variance are developed based on Shewhart, exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) control charts for mean. In all these variance control charts, log transformation of the sample variance is used. The design procedure of this chart is complex and it is poorly understood by the industry. In this paper a EWMA chart for monitoring standardized variance is developed which is having several advantages over the existing charts such as sample number free design, use in the joint monitoring scheme of process mean and variance and fit to multivariate monitoring. In industrial application this chart can be used to monitor few variables in one display simultaneously.