Design and Analysis of Universal Exponentially Weighted Moving Average Control Chart

In order to improve the sensitivity of quality control chart in monitoring small to medium process deviation, a new universal exponentially weighted moving average (UEMWA) control chart for process mean monitoring is proposed. The control chart is a general generalization of EWMA control chart. To optimize the control effect, the smoothing coefficient λ1, λ2, … λs are selected according to the data characteristics; The calculation method of mean and control limit of UEWMA control chart are given, and the average run length (ARL) and standard deviation run length (SDRL) are derived. Finally, the influence of smoothing coefficient on the performance of the UEWMA control chart is studied, and is compared with some existing control charts for monitoring small to moderate shifts. The results show that UEWMA control chart has high flexibility and sensitivity, strong expansibility and excellent control effect by designing the smoothing coefficient according to the data characteristics, and has high research value.