In this era, manufacturing sectors should ensure the quality of their production process and products. They must reduce the variability that occurs in their operation. Coefficient variation control charts have become important statistical Process Control (SPC) tools for monitoring processes when the process mean linear function with the standard deviation. In recent years, auxiliary information-based-CV control charts using memory type structure have been investigated to enhance the sensitivity of control charts. Auxiliary information is selected when the variable remains stable during the monitoring period. Nevertheless, the AIB statistic is constructed based on lognormal transformation, and no research investigated the memory type CV chart using estimator of AIB-CV from the combination of ratio and regression form called hybrid form. This research proposes a hybrid auxiliary information-based exponentially weighted moving coefficient of variation (Hybrid AIB-EWMCV) control chart for detecting small to moderate shifts in the CV process. The Average Run Length (ARL) simulation shows that increasing the level of correlation and sample sizes enhances the detection ability of the control chart. Also, the proposed chart performs well than existing chart. A real dataset from fertilizer manufacturing is implemented to explain the condition of the process by using a Hybrid AIB-EWMCV control chart.
Read full abstract