Reliability inference for dual stress factors accelerated degradation test based on the nonlinear Wiener process with three-source variability

Accelerated degradation test plays a prominent role in reliability assessment and lifetime prediction for highly reliable products. The literature on accelerated degradation modeling primarily focuses on single stress factor situations. Therefore, this article proposes a nonlinear Wiener process-based dual stress factors accelerated degradation model with interaction, which simultaneously accounts for temporal variability, individual variability, and measurement variability. The maximum likelihood estimates (MLEs) of the model parameters are obtained using the profile likelihood approach and the Nelder-Mead algorithm, along with the MLEs for the reliability metrics of interest under normal operating conditions. We then provide bootstrap confidence intervals of the model parameters using the parametric percentile bootstrap method. The performance of the proposed method is assessed through the Monte Carlo simulation. Finally, a real-world example is presented to illustrate the application of our method.