Mixed Control Charts Using EWMA Statistics

Abstract

In this paper, two mixed control charts are designed for process monitoring when the quality characteristic of interest follows a normal distribution. The mixed control chart starts with monitoring the number of non-conforming items but switches to monitoring using exponentially weighted moving average (EWMA) statistic or hybrid EWMA statistic when the decision is indeterminate with the attribute data. The average run lengths are calculated to evaluate the performance of the proposed control charts according to the mean shift. The performance of both control charts is compared with each other and with the existing control chart. Simulation study is given to demonstrate the efficiency of the proposed control charts.

Keywords

• Exponentially weighted moving average
• hybrid exponentially weighted moving average
• normal distribution
• average run length
• mixed chart