A BAYES RANKING OF SURVIVAL DISTRIBUTIONS USING ACCELERATED OR CORRELATED DATA

Summ. & Conclusions - The high reliability of modern components requir es accelerated testing to compare or predict survival or failure rates in the use condition, If the testing of highly reliable components is done in the use condition, the times to failure are too long to obser ve, Therefore, it is often required to compare or predict mean times t o failure in the use condition when the only available data are from a highly accelerated condition. Comparison of failure rates in this sit uation is possible using frequentist methods but estimation of the ind ividual failure rates Is not, Using Bayes methods, both comparison and prediction results are easily possible & computable. The accelerated model in this paper is similar to the model used in health-related res earch when the data are from a paired experiment, This use of the mode l to compare & predict survival rates using paired data is also handle d easily by the Bayes approach. Let the two failure rates be lambda(1) , lambda(2); the results are presented as the posterior probability Pr {lambda(1) < c .lambda(2)\data}, 0 < c less than or equal to 1. The pr edictions are the posterior means, E{lambda(1)\data}, E{lambda(2)\data }.