A Bayesian way of monitoring the linear profiles using CUSUM control charts. Issue 1 (2nd January 2019)

• Record Type: Journal Article
• Authors: Abbas, Tahir; Ahmad, Shabbir; Riaz, Muhammad; Qian, Zhengming
• Is Part Of: Communications in statistics. Volume 48:Issue 1(2019)
• Page Start: 126
• Page End: 149
• Publication Date: 2019-01-02
• Subjects: Average run length (ARL) -- Extra quadratic loss (EQL) -- Linear profiles -- Phase I and II -- Posterior distribution; 60E05 -- 60K99 -- 62E10 -- 62E15 -- 62P30 -- 62Q05 -- 97K50 -- 97K80; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2017.1375520
• Languages: English
• ISSNs: 0361-0918

Abstract

ABSTRACT: In statistical process control applications, profiles functions are considered an efficient way of representing quality of products or processes. Classical and Bayesian thoughts are two chief sources of defining control charting structures for profiles monitoring. This Study introduces novel Bayesian CUSUM control structures for profiles monitoring. The comprehensive comparative study identifies that the proposed Bayesian CUSUM control charts under conjugate priors has better expected performance than competing methods. The implementation of Bayesian structures requires detailed information about process parameters which come up with considerable benefits. In addition, simulative example and case study further justified the superiority of proposed techniques.