Abstract: The traditional Shewhart-p control chart only monitors the percent defective of a single property and has a certain lag in the process offset. A multivariate exponentially weighted moving average p (MEWMA-p) control chart is proposed to improve the accuracy of the percent defective control chart. The control chart applies the percent defective of multiple properties to the multivariate exponentially weighted moving average (MEWMA) control chart, which can monitor multiple properties at the same time and is more sensitive to small ranges of shift. The average run length (ARL) table of exponentially weighted moving average p (EWMA-p) control chart and MEWMA-p control chart is studied and compared at the same offset degree, and the validity of this method is verified by simulation.

Key words: Shewhart-p control chart, multivariate exponentially weighted moving average control chart, average run length