A Study on Evaluation of Stability in Process Mean Using Bayesian Updating

Yasuhiko Takemoto, Ikuo Arizono

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

It is common that a manufacturing process is unstable at the beginning of operation. Then, a process condition is brought into being stable with the lapse of time. The evaluation of stability is very important to process management in order to shift to mass production. In this study, we propose a method of evaluating the stability in a process. In a detail, we investigate a conjugate distribution of the process mean based on Bayesian theory first. In particular, we consider the difference between both of the prior and posterior probability distributions in the process mean to be a criterion for the stability of a process. Hence, we evaluate the difference between the both distributions using information theory. Then, a numerical example in the method of evaluating the stability in a process is shown.

Original languageEnglish
Title of host publication2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling, APARM 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728171029
DOIs
Publication statusPublished - Aug 2020
Event2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling, APARM 2020 - Vancouver, Canada
Duration: Aug 20 2020Aug 23 2020

Publication series

Name2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling, APARM 2020

Conference

Conference2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling, APARM 2020
Country/TerritoryCanada
CityVancouver
Period8/20/208/23/20

Keywords

  • Kullback-Leibler divergence
  • Prior and posterior distribution
  • Statistical process control
  • component; Baysian statistics

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Mechanical Engineering
  • Safety, Risk, Reliability and Quality
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'A Study on Evaluation of Stability in Process Mean Using Bayesian Updating'. Together they form a unique fingerprint.

Cite this