Bayesian analysis of incomplete data from accelerated life testing with competing failure modes

A Bayesian method is proposed for analyzing incomplete data obtained from constant stress accelerated life testing when there are two or more failure modes, or competing failure modes. The incompleteness is mainly due to censoring, as well as masking which might be the case that the failure time is observed, but its corresponding failure mode is not identified. Furthermore, sensitivity analysis is performed to check model sensitivity. The method is illustrated by an example for validation, which also shows that the method provides better estimators than MLE by incorporating available prior information in the case of small sample sizes.