A new multi-criteria importance measure and its applications to risk reduction and safety enhancement

Abstract

A wide variety of importance measures have been proposed to rank components for different purposes and most of them are based on a single criterion like reliability, maintenance cost, or risk. In many practical situations, however, people need to evaluate alternatives based on multiple criteria. Thus, the single-criterion importance measure is no longer sufficient for such evaluation. Fewer research has been reported on importance measures applicable to multi-criteria engineering systems, which however fail to consider the correlation among different criteria leading to inaccurate and even misleading importance results. In this paper, a new multi-criteria importance measure (MCIM) is proposed to identify the weakest components in complex engineering systems and it objectively considers the correlation among different criteria. The proposed MCIM is then extended to two common types of complex engineering systems, series hierarchical systems and modular systems. At last, the applications and effectiveness of the proposed MCIM are illustrated by two case studies, including reducing risks of a hierarchical supply chain system and improving the safety of a train bogie modular system.