A change-point model for monitoring the coefficient of variation based on squared ranks test

摘要

Monitoring the coefficient of variation tends to be established as an efficient approach of the statistical process control in case of simultaneously fluctuating mean and variance. However, the recent research is constricted to develop control charts for monitoring the coefficient of variance based on normality assumption. This paper presents a new efficient method to monitor the coefficient of variation implementing the well known distribution-free square ranks test through a change-point model. Extensive numerical analysis is presented to evaluate the performance of the proposed change-point model alongside with tables showing its statistical run length properties and comparison examples with conventional control charts.