An Improved Multivariate Short Run Control Chart Based On The CUSUM Statistic

Abstract

Short run control charting in necessary at the start-up of a process and at the initiation of a new process for which prior information is unavailable. Numerous univariate short run control charts have been introduced to overcome the problems faced by conventional charts in a short run environment. Most practical scenarios involve several related variables. the multivariate short run charts for individual measurements and subgrouped data were proposed so that a joint monitoring of several correlated variables can be made simultaneously in a short run environment (see[5]). This paper aims at improving the performance of the multivariate short run chart for individual measurements by using its statistics to construct a cumulative sum (CUSUM) chart.