Importance-measure based methods for component reassignment problem of degrading components. (October 2019)

• Record Type: Journal Article
• Authors: Fu, Yuqiang; Yuan, Tao; Zhu, Xiaoyan
• Is Part Of: Reliability engineering & system safety. Volume 190(2019)
• Publication Date: 2019-10
• Subjects: Component reassignment -- Degradation mode -- Functionally exchangeable components -- Time-dependent importance measure -- Linear consecutive degradation-θs-out-of-n systems; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2019.106501
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Modeling of a general component reassignment problem of degrading components. Proposal of a time-dependent importance measure of degrading components (TID). A TID-based greedy local search for the component reassignment problem. A TID-based decomposition method with nonlinear optimization & genetic algorithm. Exemplification of exponential degradation & consecutive degradation systems etc. Abstract: This paper studies a component reassignment problem of degrading components under two degradation termination modes: hot degradation and cold degradation. Consider a system in which components are functionally exchangeable but their degradation characteristics are different and the operational loads or environmental conditions or both that the components are exposed to are different. To improve overall performance of the system, it is critical to determine the time and plan of reassigning the components among the positions in the system by means of reliability and lifetime modeling. To address such the component reassignment problem, component degradation models incorporating the component reassignment and a system-lifetime maximization model are established, a new time-dependent importance measure for degrading components is proposed, and a decomposition method is developed for solving the component reassignment problem. The decomposition method uses the proposed importance measure and interactively uses a continuous nonlinear optimization algorithm and a heuristic combinatorial optimization method. Further, the system-lifetime model is exemplified for four systems with exponentially degrading components, including the k out of n systems, linear consecutive- k -out-of- n systems, and their derivative degrading systems. Finally, numerical experiments demonstrate efficiency of the method and benefits of component reassignment.