Stress-strength reliability of a non-identical-component- strengths system under the progressive censoring sample from the two parameter Rayleigh distribution

This paper deals with the statistical inference of the stress-strength reliability of a multi-component system with non-identical-component strengths based on the progressively Type-II censored sample from the two-parameter Rayleigh distribution‎. ‎Both stress and strength are assumed to have a Rayleigh distribution with different scale parameters yet similar location parameters‎. ‎Its maximum likelihood estimate‎, ‎asymptotic confidence interval‎, ‎Bayes estimates‎, ‎and highest posterior density are derived‎. ‎The uniformly minimum-variance unbiased estimator and Bayes estimations for the reliability are obtained when the common location parameter is known‎. ‎Different methods are compared using Monte Carlo simulations‎. ‎{The results demonstrate that Bayes estimates outperform maximum likelihood estimates‎, ‎highest posterior density intervals outperform asymptotic intervals‎, ‎and in Bayes estimates‎, ‎informative priors outperform non-informative priors.} Finally‎, ‎a dataset is analyzed for illustrative purposes.

کلیدواژه ها:

Lindley's approximation ، Markov Chain Monte Carlo method ، Multi-component stress-strength reliability ، Progressively censored