Failure Analysis of Coupler Knuckle Considering Truncated and Censored Lifetime Data

Abstract

Increased axle load, heavy load, and increased speed of railway wagons need high reliability of coupler knuckle. However, the failure analysis of coupler knuckle is troublesome under incomplete lifetime data due to extensive management. In this paper, maximum likelihood estimation method is employed to with a mixture data of left truncation, interval censoring, and right censoring. The parameter estimations are done by Newton-Raphson method and bootstrap resampling. The results of case study shows that the proposed method is reasonable and could reflect the reality. These results have important application value in reality for further refinement of the maintenance strategy according to the remaining life of components, thereby reducing the maintenance cost and improving the reliability of components.