Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 343 - Contributed Poster Presentations: Section on Bayesian Statistical Science
Type: Contributed
Date/Time: Tuesday, August 9, 2022 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #322901
Title: A Comparison of Bayesian Accelerated Life Testing Models in the Presence of Dual Stresses
Author(s): Neill Smit* and Lizanne Raubenheimer
Companies: North-West University and Rhodes University
Keywords: Accelerated life testing; Bayesian methods; Deviance information criterion; Generalised Eyring

In this paper, Bayesian accelerated life testing models that make use of the generalised Eyring time transformation function are compared. In accelerated life testing, products are tested under more severe than their normal operating conditions. This is done to obtain failure data in a much shorter timeframe, where the life characteristics of the product under the normal operating conditions can then be extrapolated. The models compared incorporate two stress variables, one thermal and one non-thermal. The Weibull, Birnbaum-Saunders and log-normal distributions are used as the life distributions. The models are applied to a real dataset, where various prior settings are considered as part of a sensitivity analysis. Due to the intractable nature of the posterior distributions for these models, Markov chain Monte Carlo methods are employed to generate posterior samples for inference. The deviance information criterion is used to compare the fit of the models and the predictive reliability is also calculated.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program