Bayesian reliability analysis using the Dirichlet prior distribution with emphasis on accelerated life testing run in random order Original Research Article

This paper examines Bayesian analysis as a tool for predicting both reliability growth and reliability of censored multi-level, multi-stress accelerated life tests. The methodology is examined using the Dirichlet prior distribution in reliability growth applications and in randomly ordered accelerated life tests occurring in designed experiments (DOE). A simulation is used to examine the utility and accuracy of the approach both graphically and statistically. The results are both enlightening and conclusive