Estimating MTTF of a Component Based on Spare Parts Consumption Data

Abstract

In this paper we consider the situation where the OEM of a repairable system wants to estimate the mean time to failure (MTTF) of a given component of the system and the component’s lifetime distribution so as to forecast the demand of spare parts and to optimize the maintenance policy of the component. The OEM does not have exact field failure time data of the component but has the installed base information of the systems and the spare parts consumption information of the component. Due to lack of failure time data, the failure-time-based approach is no longer applicable. To overcome this difficulty, a novel approach is proposed to estimate MTTF and lifetime distribution of the component based on its spare parts consumption data. The proposed approach is based on the assumption of Weibull renewal process and a modification of the asymptotic renewal function. It provides a much more accurate estimate of MTTF than the one obtained from the exponential distribution assumption. A numerical example is included to illustrate the appropriateness and usefulness of the proposed approach.