Robust Estimation of Component Reliability Based on System Lifetime Data with Known Signature

Abstract

This paper considers the estimation of component reliability based on system lifetime data with known system signature using the minimum density divergence estimation method. Different estimation procedures based on the minimum density divergence estimation method are proposed. Standard error estimation and interval estimation procedures are also studied. Then, a Monte Carlo simulation study is used to evaluate the performance of those proposed procedures and compare those procedures with the maximum likelihood estimation method under different contaminated models. A numerical example is presented to illustrate the effectiveness of the proposed minimum density divergence estimation method. We have shown that the proposed estimation procedures are robust to contamination and model misspecification. Finally, concluding remarks with some possible future research directions are provided.

Keywords

• censoring
• maximum likelihood estimation
• minimum density divergence
• Monte Carlo simulation
• Weibull distribution