طراحي نمودار كنترل تغييرپذيري فرآيندهاي دو متغيره نرمال

Designning A Control Chart to Monitor Variability of Bivariate Normal Processes

Efficient monitoring of correlated process via control charts has seriously been on attention of many researchers to follow the performance of a process along time; the main goal of this study is to examine if the variation of the process is in the control when in the perturbation factors doesnʹt change the variation of process. In this article, through presenting the quantilies of the relevant test statistic and using PearsonWilks statistic, we propose a simplified method to design the L chart that concentrates on changes in the covariance matrix. This control chart enables the user to determine the out of control points when variations occur in the covariance matrix. Like the standard deviation charts for univariate processes, this chart and multivariate cumulative sum control chart can be considered as suitable tools to detect variations. In addition, general guidelines for the proper implementation of the proposed method are provided. A numerical example is provided to illustrate the procedure of designing the proposed control chart. Also, the performance of the control chart is examined by simulating some significant variations in the process.