Nonparametric EWMA chart for simultaneous monitoring of event frequency and magnitude

摘要

Traditionally, control charts for joint monitoring of the event frequency and magnitude are designed on the assumption of a parametric distribution. However, when there is a lack of knowledge about the underlying distribution, the parametric control charts may be not suitable in some real applications. Therefore, distribution free control charts have attached much attention in recent years. In this paper, we propose a nonparametric EWMA chart based on Mathur statistic. The control limits can be obtained via simulation method. Some in control and out of control performances are discussed and the results indicate the proposed chart can successfully detect shifts in different general bivariate processes. Finally, an example is given to illustrate the implementation of the proposed method.