Dependent Competing Risks Modeling for Mechanical Systems under Component and Performance Failure

Jingyi Liu, Yugang Zhang, Bifeng Song

Research output: Contribution to journalArticlepeer-review

Abstract

This paper establishes a competing failure analysis model for complex mechanical system under component failure and performance failure considering degradation. Traditionally, mechanical system is composed by a number of components. Meanwhile, mechanical system has the ability to accomplish its specific performance. Therefore, mechanism may fail because of two kinds of failure modes, the component failure due to degradation (such as component wear) and the performance failure (system couldn't complete performance). The two failure modes compete with each other because as soon as one mode occurs the system just fails. The component will degrade with time as system operates as well as the system performance. In this paper, Brownian motion (BM) with nonlinear drift is used to model the degradation of components based on which component failure is analyzed. The function of performance measurement is built by surrogate and performance failure is analyzed by it at different working circulation. Farlie-Gumbel-Morgenstern (FGM) copula is introduced to describe the dependence. The system reliability is analyzed by FGM copula as well as competing failure probability for each failure mode. Finally, a numerical example and an engineering case study are used to illustrate the proposed model.

Original languageEnglish
Article number021001
JournalASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering
Volume5
Issue number2
DOIs
StatePublished - 1 Jun 2019

Keywords

  • BM with nonlinear drift
  • competing failure probability
  • degradation
  • FGM copula
  • mechanical system

Fingerprint

Dive into the research topics of 'Dependent Competing Risks Modeling for Mechanical Systems under Component and Performance Failure'. Together they form a unique fingerprint.

Cite this