A Control Chart Based on A Nonparametric Multivariate Change-Point Model

Phase-II statistical process control (SPC) procedures are designed to detect a change in distribution when a possibly never-ending stream of observations is collected. Several techniques have been proposed to detect a shift in location vector when each observation consists of multiple measurements. These procedures require the user to make assumptions about the distribution of the process readings, to assume that process parameters are known, or to collect a large training sample before monitoring the ongoing process for a change in distribution. We propose a nonparametric procedure for multivariate phase-II statistical process control designed to detect shifts in location vector that relaxes these requirements based on an approximately distribution free multivariate test statistic. This procedure may not be appropriate for some multivariate distributions with unusual dependence structure between vector components. A diagnostic tool that can be used if a historical sample of data is available is provided to assist user in determining if the proposed procedure is appropriate for a given application.