The Use of Bayesian Methods in Reliability Data Analyzes and Modeling

• Activity Number: 90
• Type: Roundtables
• Date/Time: Monday, August 4, 2014 : 7:00 AM to 8:15 AM
• Sponsor: Quality and Productivity Section
• Author(s): William Q. Meeker*+
• Companies: Iowa State University
• Keywords: Monte Carlo Markov Chain ; physics of failure ; random effects ; censored data

Abstract

The development of theory and application of Monte Carlo Markov Chain methods, vast improvements in computational capabilities, and emerging software alternatives have made it possible for the wide use of Bayesian methods in reliability applications. Bayesian methods, however, remain controversial in reliability because of the concern about where the needed prior distributions should come from. On the other hand, there are many applications in which engineers have solid prior information based on physics of failure or previous experience with the same failure mechanism (e.g., imprecise knowledge about the activation energy in a temperature-accelerated life test or about the Weibull shape parameter in analysis of fatigue failure data). Frequently, there are gaps in reliability modeling inputs that cannot be filled with traditional reliability data sources. These and other applications make a compelling case for the use of Bayesian methods in reliability analysis and modeling. Bayesian methods also provide a powerful approach to handling complicated data/model combinations involving, for example, censoring and random effects. This roundtable will allow sharing of such applications.