Competing Risk Analysis in Constant Stress Partially Accelerated Life Tests Under Censored Information

This article presents a comprehensive study on competing risk analysis in constant stress partially accelerated life tests (CSPALT) under censored information. The focus is on the Fréchet distribution, a widely used model in extreme value theory, under Type-I and Type-II censoring schemes. The study introduces a novel framework for estimating parameters in CSPALT under competing risk plans, including the construction of likelihood functions, Fisher Information Matrix, and variance-covariance matrix. The authors conduct a simulation study to evaluate the performance of the estimators and highlight the practical applications of CSPALT in industrial settings. The article concludes by emphasizing the robustness and stability of the proposed test design, encouraging further research in different failure models and censoring plans.