Estimation of Multi-Component Reliability Parameter in a Non-identical-Component Strengths System Under Dependency of Stress and Strength Components
محورهای موضوعی : فصلنامه ریاضی
1 - Imam Khomeini International University, Iran
کلید واژه: Stress-strength reliability, Kumaraswamy generalized distributions, Copula theory, MCMC method, Bayesian inference,
چکیده مقاله :
Generating more realistic stress-strength model is main attempt, in this paper. For this aim, inference on stress-strength parameter was considered in a multi-component system with the non-identical-component strengths, based on the Kumaraswamy generalized distribution, when the stress and strength variables are dependent. The dependency assumption is studied by Copula theory, one of the most important concept in dependent variables. The maximum likelihood estimation (MLE), bootstrap confidence interval, Bayesian approximation and highest posterior density (HPD) interval are obtained, for the multi-component stress-strength parameter. Employing Monte Carlo simulations, the performance of different estimations are compared together. Finally, one real data set is analyzed for illustrative purposes.
Generating more realistic stress-strength model is main attempt, in this paper. For this aim, inference on stress-strength parameter was considered in a multi-component system with the non-identical-component strengths, based on the Kumaraswamy generalized distribution, when the stress and strength variables are dependent. The dependency assumption is studied by Copula theory, one of the most important concept in dependent variables. The maximum likelihood estimation (MLE), bootstrap confidence interval, Bayesian approximation and highest posterior density (HPD) interval are obtained, for the multi-component stress-strength parameter. Employing Monte Carlo simulations, the performance of different estimations are compared together. Finally, one real data set is analyzed for illustrative purposes.
