Bayesian Reliability Estimation for Series System Based on The Inverse Ramos – Louzada Distribution Under Different Loss Functions

الملخص

In this paper, we apply Bayesian estimation of a series system based on the Inverse Ramos–Louzada (IRL) distribution on doubly censored data. The three prior distributions, namely Flat, Gamma, and Jeffrey's priors, are taken with the three loss functions, namely LINEX, Huber, and Log-Cosh. Mathematical representation of the reliability function of the series system is made and the estimators are obtained using Bayesian methods. A Monte Carlo simulation with sixteen cases is performed to estimate the estimator based on Mean Squared Error (MSE), Pitman Closeness (PC), and Relative Efficiency (RE). In actuality, the Bayes-Gamma estimator yields reliability values that are closer to the true system reliability, and the best estimates can be found using accuracy criteria. Such results support that Bayesian estimation is efficient for modelling reliability with doubly censored samples