Bayesian monitoring of linear profile monitoring using DEWMA charts

Process monitoring is an essential element for an improved quality of final products. A variety of tools are used for it; control charts are one of these choices. Classical and Bayesian thoughts are 2 main aspects of statistics used in different areas of application. This study introduces an approach to existing theories in applied quality control: Bayesian double exponentially weighted moving average (DEWMA) control charts for monitoring the profiles of products and processes. Three novel univariate Bayesian DEWMA charting structures for the Y intercepts, slopes, and error variances are designed under phase 2 procedures. The performance of the designed structures of control charts is evaluated based on different run length measures. The comparative analysis revealed that Bayesian DEWMA control charts are efficient at identifying the sustainable shifts in the process parameters. Moreover, DEWMA control charts are more effective under classical and Bayesian methodologies for detecting smaller value shifts compared with exponentially weighted moving average charts. We have examined that acquiring extra information in the form of prior's about process parameters comes up with tangible benefits and enhances the detection potential of DEWMA charts for profiles monitoring. An example and case studies are provided to justify the above findings.