A Bayesian Control Chart for Monitoring Process Variance

Automation in the service industry is emerging as a new wave of industrial revolution. Standardization and consistency quality an important part automation process. The control methods widely used manufacturing can provide measurement process monitoring. In particular, chart online monitoring technique be to quickly detect whether out control. However, more difficult than that because variability comes from widespread complex factors. First all, distribution usually non-normal or unknown. Moreover, skewness time-varying, even if this study, Bayesian procedure applied construct Phase II exponential weighted moving average (EWMA) for variance distribution-free We explore sampling properties statistic, which suitable time-varying distribution. run lengths (ARLs) proposed EWMA are calculated, they show performs well. simulation studies normal process, mixed prove our any shift variance. Finally, numerical example bank time illustrate application confirm performance