The lay abstract featured today (for Optimal design of constant-stress accelerated degradation tests based on the Wiener process with manufacturing batches heterogeneity and individual differences by Nianhuan Li, Dongwei Gu, Bingkun Chen, Weiliang Li, Xilu Zhao and Pengfei Chen) is from Quality and Reliability Engineering International with the full article now available to read here.

Abstract

The application of constant-stress accelerated degradation tests modeling proves to be a highly effective and valuable approach for evaluating the reliability of a diverse range of products across industries. Nevertheless, during the execution of reliability experiments, one may encounter the challenge that data acquisition originates from diverse sources, including distinct manufacturing batches, equipment, and operators. This phenomenon may lead to a diminished precision in the evaluations of the tests, posing a potential obstacle to obtaining highly reliable results. In this comprehensive research endeavor, optimal designs for the Wiener constant-stress accelerated degradation model are introduced, strategically considering the heterogeneity present in manufacturing batches and individual differences. The overarching objective is to minimize the variance of the mean time to failure, ensuring optimal reliability, all while meticulously adhering to predetermined constraints such as budget, test duration, and the number of available test units. To attain this objective, particle swarm optimization is employed to identify the optimal solution based on a multifaceted objective function. Additionally, a comparison is made between our proposed model and alternative methodologies using degradation data from LEDs, demonstrating its superior goodness-of-fit. In the conclusive phase, exemplary instances of optimal constant-stress accelerated degradation test plans under varying constraints are presented, and a sensitivity analysis is conducted.