Bayesian models and ageing indicators for analysing random changes in failure occurrence

The paper introduces ageing models of repairable components based on Bayesi an approach. Models for the development of both failure rate and the probab ility of failure on demand are presented. The models are based on the assum ption that the failure probability or rate has random changes at certain ti me points. This is modelled by assuming that the successive transformed fai lure probabilities (or rates) follow a Gaussian random walk. The model is c ompared with a constant increment model, in which the possible ageing trend is monotone. Monte-Carlo Markov Chain sampling is applied in the determina tion of the posterior distributions. Ageing indicators based on the model p arameters are introduced, and the application of these models is illustrate d with case studies. (C) 2000 Elsevier Science Ltd. All rights reserved.