Inferential Survival Analysis for Inverted NH Distribution Under Adaptive Progressive Hybrid Censoring with Application of Transformer Insulation

In this paper, the reliability analysis of the inverted Nadarajah–Haghighi (INH) distribution under an adaptive type-I progressive hybrid censoring scheme (AT-I PHCS) has been investigated. The unknown parameters of the INH distribution based on AT-I PHCS have been estimated using Bayesian and non-Bayesian methods. The asymptotic and two bootstrap confidence intervals are also calculated, as well as maximum likelihood estimates of the unknown parameters. The maximum product spacing estimation method based on AT-I PHCS has been introduced. Bayesian estimates of the unknown parameters are obtained based on symmetric (squared error) loss function. Furthermore, the Markov chain Monte Carlo (MCMC) technique is used to compute the Bayesian estimators and the associated credible intervals. Bootstrap confidence intervals have been discussed for parameters of INH based on AT-I PHCS. A real-life data set is used of progressively censored samples for transformer insulation life testing and compare tests included three levels of constant voltage, which were 35:4, 42:4, and 46:7kv, respectively. Finally, a simulation study is conducted to evaluate the estimators' performance.