Bayesian reliability acceptance sampling plans under adaptive simple step-stress partial accelerated life test

In the traditional simple step-stress partial accelerated life test, items are placed under normal operating conditions up to a specified time, after which the stress is increased to obtain failure-time information earlier. However, increasing the stress level introduces an additional cost, which raises the overall cost of the life test. To address this issue, we consider an adaptive step-stress scenario in which the stress is increased at a predetermined time only if the number of failures observed up to that time point is less than a pre-specified number. In this work, we develop Bayesian reliability acceptance sampling plans based on an adaptive simple step-stress partial accelerated life test conducted under type-I censoring. The Bayes decision function and Bayes risk are derived for a general loss function, and the optimal Bayesian reliability acceptance sampling plan under this adaptive setting is obtained for the quadratic loss function by minimizing the Bayes risk. An algorithm is provided to determine the optimal plan. Finally, we compare the proposed Bayesian reliability acceptance sampling plan with those obtained under conventional non-accelerated and conventional accelerated life tests to illustrate the improvement achieved under the adaptive scenario.