Bayesian accelerated life testing: A generalized Eyring-Birnbaum-Saunders model

In this paper, an accelerated life testing model using the Birnbaum-Saunders distribution and the generalized Eyring model is explored. A Bayesian approach is proposed, considering the combined effect of one thermal stressor and one non-thermal stressor. The model is demonstrated via an application to an electronics epoxy packaging ALT data set, with temperature and humidity as accelerated stressors. Markov chain Monte Carlo methods are employed to generate posterior samples for inference. A sensitivity analysis is performed via the deviance information criterion. The predictive reliability is also investigated.