PROJECT TITLE :

A Bayesian Method for Planning Accelerated Life Testing

ABSTRACT:

During this paper, a Bayesian criterion is proposed based mostly on the expected Kullback-Leibler divergence between the posterior and the previous distributions of the parameters of interest. We have a tendency to decision the Bayesian criterion the reference optimality criterion, that is to find an optimal arrange to maximise the amount of information from the data. A giant-sample approximation is used to simplify the formula to get optimal plans numerically. As a result of optimal plans based mostly on reference optimality criterion don't rely on the sample size, a modified reference optimality criterion is proposed. We offer numerical examples using the Weibull distribution with type I censoring to illustrate the strategies, and to look at the influence of the prior distribution, censoring time, and sample size. We have a tendency to additionally compare our strategies with other criteria through Monte Carlo simulation.