An Improved Modular Approach for System Reliability Analysis
A modular method, based on Binary Decision Diagrams and Bayesian Networks, is proposed to analyze system reliability. A fault tree is divided into independent subtrees using a very efficient linear time algorithm. Different solution techniques are applied to each subtree depending on the latters characteristics, Binary Decision Diagrams solution for static fault trees, and Bayesian Networks solution for dynamic fault trees. The solutions of various independent subtrees are integrated to obtain the reliability of urban mass transit vehicles. The modular method, based on based on Binary Decision Diagrams and Bayesian Networks, can overcome main shortcomings of Markov Chains, and is useful for assessing the reliability of large and complex systems.
modular Binary Decision Diagram system reliability Bayesian Networks
TAO Yongjian DONG Decun
School of Transportation Engineering Tongji University Shanghai, China
国际会议
2009 WASE International Conference on Information Engineering(2009年国际信息工程会议)(ICIE 2009)
太原
英文
111-113
2009-07-10（万方平台首次上网日期，不代表论文的发表时间）