A Bayesian Approach to Capability Testing Based on Cpk with Multiple Samples. (19th April 2013)

• Record Type: Journal Article
• Authors: Kargar, M.; Mashinchi, M.; Parchami, A.
• Is Part Of: Quality and reliability engineering international. Volume 30:Number 5(2014:Sep.)
• Page Start: 615
• Page End: 621
• Publication Date: 2013-04-19
• Subjects: Reliability (Engineering) -- Periodicals; Quality control -- Periodicals; High technology -- Periodicals; 620.00452
• DOI: 10.1002/qre.1512
• Languages: English
• ISSNs: 0748-8017

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>Process capability indices provide numerical measures to compare the output of a process to client's expectations. However, most of the existing researches have used traditional distribution frequency method by using a single sample due to assess process capability. An alternative to this approach is to use the Bayesian method. In this paper, we utilize a Bayesian approach based on subsamples to check process capability via capability index C_pk. As a new suggestion, we used the informative normal prior distribution and the characteristics of sufficient statistic of the parameter to drive the posterior distribution. The capability test is done, and the posterior probability <italic>p</italic>, for which the process under investigation is capable, is derived both based on the most popular index <italic>C_pk</italic>. Finally, a numerical example is given to clarify the method. Copyright © 2013 John Wiley &amp; Sons, Ltd.</p> </abstract>