A Bayesian reliability evaluation method with different types of data from multiple sources

Lizhi Wang, Rong Pan, Xiaohong Wang, Wenhui Fan, Jinquan Xuan

Research output: Contribution to journalArticle

15 Scopus citations

Abstract

Bernoulli data (pass/fail), lifetime data, and degradation data are commonly encountered in product reliability assessment. Oftentimes these data are collected from different sources (such as field use, accelerated tests, history, and so on), and it is desirable to utilize these heterogeneous data within one computational framework to provide a comprehensive evaluation of product reliability. In this paper, three Bayesian inference models are proposed to establish the relationship among pass/fail-type Bernoulli data, lifetime data, and degradation data, and to integrate them to solve relevant problems and improve the accuracy of reliability prediction. The proposed methods are demonstrated by a synthetic example and two real examples. The evaluation results can be used for formulating product development strategies.

Original languageEnglish (US)
Pages (from-to)128-135
Number of pages8
JournalReliability Engineering and System Safety
Volume167
DOIs
StatePublished - Nov 1 2017

    Fingerprint

Keywords

  • Bayesian method
  • Bernoulli data
  • Data from multiple sources
  • Data integration
  • Reliability evaluation

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Industrial and Manufacturing Engineering
  • Applied Mathematics

Cite this