Estimation of multi-component stress-strength model based on geometric upper record values

Stress-strength models have special importance in reliability literature and engineering applications. This paper consists of the estimation problem of a stress-strength model with a multi-component system, i.e., a system that can be regarded to be alive if at least s out of k (s ≤ k) strength components exceed the stress component. The reliability of such a system has been obtained when both the stress and the strength variables have Geometric distributions. UMVUE of Rs,k is obtained based on upper record values. Bayesian estimators under the squared error loss function using the conjugate beta prior distributions have been obtained. A simulation study has been implemented to assess the performance of estimates. KEYWORDS : Stress-strength, Multi-component, Upper record values, UMVUE, Loss function