Bayesian Reliability Analysis With Evolving, Insufficient, and Subjective Data Sets

Abstract

This paper presents a new paradigm of system reliability prediction that enables the use of evolving, insufficient, and subjective data sets. The data sets can be acquired from expert knowledge, customer survey, inspection and testing, and field data throughout a product life-cycle. In order to handle such data sets, this research integrates probability encoding methods to a Bayesian updating mechanism. The integrated tool is called Bayesian Information Toolkit. Subsequently, Bayesian Reliability Toolkit is presented by incorporating reliability analysis to the Bayesian updating mechanism. A generic definition of Bayesian reliability is introduced as a function of a predefined confidence level. This paper also finds that there is no data-sequence effect on the updating results. It is demonstrated that the proposed Bayesian reliability analysis can predict the reliability of door closing performance in a vehicle body-door subsystem, where available data sets are insufficient, subjective, and evolving.