Inference for Unit Inverse Weibull Distribution Under Block Progressive Type-II Censoring

In this paper, estimation is considered for model parameters and reliability assessment is also discussed under a block progressive censoring scheme, which enhances experimental efficiency by conducting tests within multiple testing facilities. Under the assumption that failure times of units follow a unit inverse Weibull distribution, various point and interval estimates of parameters and reliability indices are obtained with different techniques. The maximum product spacing technique is developed for block progressive censoring and compared with maximum likelihood and Bayesian approaches. Further, existence and uniqueness of maximum likelihood estimators are established. To evaluate complex posterior estimates, the Metropolis–Hastings sampling algorithm is also applied for computation. Extensive simulation studies are conducted for investigating the performance of different methods. A real data set is analyzed in support of the findings under block progressive censoring.