Impact of measurement error on mixed EWMA-CUSUM control chart

Abstract. In statistical process control, measurement error plays a key role that is usually ignored. Measurement error can yield incorrect conclusions about the performance of the process. This study examined the effect of measurement error on the shift detection ability of the mixed Exponentially Weighted Moving Average-Cumulative Sum (EWMA-CUSUM) control chart. Then, it investigated the performance of the mixed EWMA-CUSUM chart in case of mean shift through (i) covariate method, (ii) multiple measurement method, and (iii) linearly increasing variance method. The performance measurement tools such as Average Run Length (ARL) and Standard Deviation of Run Length (SDRL) were estimated using the Monte-Carlo simulation method. It was concluded that the performance of the mixed EWMA-CUSUM control chart was adversely affected by considering the measurement error. It was revealed from the comparative study that the mixed EWMA-CUSUM control chart outperformed the EWMA and CUSUM control charts in the presence of measurement error. An illustrative example was presented to demonstrate the performance of control charts in case of measurement error.