Analyzing Copula Dependent Competing Risks Using Step‐Stress Accelerated Life Testing

Abstract

Through this paper, statistical analysis for the dependent competing risks (CRs) model is introduced in simple step‐stress accelerated life testing (SSALT) under progressive Type‐II censoring (PT‐IIC). Weibull distribution (WD) and Tampered random variable (TRV) are considered to study the failure time of test units. Maximum likelihood estimators (MLEs) of the unknown parameters, acceleration coefficients and the reliability of units are obtained by using the Copula function (CF) to construct dependent correlations among the marginal distributions of a number of CRs factors. Also, CF is applied to utilize the copula characteristics to ascertain how close the reliance is. Bayesian estimators (BEs) by Markov Chain Monte Carlo (MCMC) method is computed under PT‐IIC. Approximate and credible confidence intervals (CIs) of the parameters are computed. Finally, the precision of MLEs and BEs for the parameters is demonstrated by simulation studies.