An efficient nonparametric double progressive mean chart for monitoring of the process location

Abstract

A control chart has become a choice of quality practitioners for monitoring the output of industrial and production processes. It is a common practice to develop control charts under normality assumption or known distribution of the quality characteristic(s). These control charts are known as parametric charts. These charts may provide misleading results when the normality assumption of the process distribution is doubtful or unknown. In such situations, the nonparametric (NP) charts serve as an effective alternative for efficient monitoring of the process parameter(s). In this study, we have proposed a new NP double progressive mean chart based on sign statistic for detecting small shifts in the process location. The run-length (RL) properties of the proposed chart have been computed and compared with the existing competitor charts. We have used the most popular performance measure namely average RL (ARL) as an evaluation criterion. The comparisons revealed that the proposed chart performs better than its existing competitor’s charts. We have also considered a real-life data set related to a piston ring manufacturing process for the practical implementation of the proposed and existing charts.