Title: Multi-state system importance analysis method of fuzzy Bayesian networks

Authors: Rui-Jun Zhang; Lu-Lu Zhang; Ming-Xiao Dong

Addresses: School of Mechanical and Electrical Engineering, Shandong Jianzhu University, Lingang Development Zone 250101, Jinan, China ' School of Mechanical and Electrical Engineering, Shandong Jianzhu University, Lingang Development Zone 250101, Jinan, China ' School of Mechanical and Electrical Engineering, Shandong Jianzhu University, Lingang Development Zone 250101, Jinan, China

Abstract: In order to quantify the reliability index of the real system and identify the key events affecting the reliability of the system, fuzzy importance analysis method which can be applied to multi-state system is proposed on the basis of Bayesian network targeting the fuzziness and uncertainties of information. The fuzzy set theory is introduced into the Bayesian network analysis. The failure likelihood of the various components of the system is represented by fuzzy subset, and the fault states of components and system are described by fuzzy numbers. Considering the uncertainty of the fault logical relationship among components, the fuzzy conditional probability tables are used to describe the fault logical relationship among components. Two kinds of fuzzy Bayesian network importance are proposed on the basis of fuzzy Bayesian network analysis algorithms, which describe the contribution of various components to system failure clearly. At last, it is proved that the methods are feasible in the important analysis of the accident of crane rope-breaking.

Keywords: fuzzy set theory; Bayesian networks; crane accidents; crane rope breaking; reliability; uncertainty; fuzzy importance analysis; multi-state systems; component failure; failure probability; system failure; fuzzy logic.

DOI: 10.1504/IJISE.2015.072272

International Journal of Industrial and Systems Engineering, 2015 Vol.21 No.3, pp.395 - 414

Received: 16 Sep 2013
Accepted: 14 Mar 2014
Published online: 08 Oct 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article