基于模糊贝叶斯网络的铁路危险货物运输过程风险评估  

Risk Assessment of Railway Dangerous Goods Transport Process Based on Fuzzy Bayesian Network

作　　者：杨能普 杨月芳[1] 冯伟 

机构地区：[1]北京交通大学交通运输学院,北京100044 [2]中铁集装箱运输有限责任公司乌鲁木齐分公司,新疆维吾尔自治区乌鲁木齐830011

出　　处：《铁道学报》2014年第7期8-15,共8页Journal of the China Railway Society

基　　金：铁道部科技研究开发计划(2012X007-G)

摘　　要：基于Buckley决策方法,标定基本事件之间相对概率大小模糊权值,以此作为各基本事件的概率值;同时以事故树模型为基础,映射为贝叶斯网络;通过正、反向推理分析各基本事件的结构重要度、概率重要度、关键重要度,并对此进行排序,有效地定量评估运输过程的风险性,找出最薄弱的工作环节,有针对性地提出改进措施。对比仅采用事故树的结构重要度排序分析,本方法提高评估的可信度和客观性,对铁路危险货物运输风险评估具有一定的理性论和实际应用价值。With the method of Buckley Decision Making,the fuzzy weight value of relative probability between an elementary event and another one was calibrated.The fuzzy weight value was taken as the probability of each elementary event.The FTA model was mapped into the Bayesian Network （BN）.According to forward and reverse reasoning of the BN,the structural importance degree,probability importance degree and critical importance degree of each elementary event were analyzed and sorted.In the light of sequencing of the three importance degrees,risks of transportation processes can be assessed quantitatively,the weakest link of transportation can be identified effectively and directional measures can be taken for improvement.Compared to sorting only by the structural importance degree of FTA,the reliability and obj ectivity of risk assessment were greatly improved.The proposed method is of theoretical and practical values to risk assessment of railway dangerous goods transport.

关 键 词：危险货物运输 贝叶斯网络 事故树模型 Buckley决策 风险评估 

分 类 号：U294.83[交通运输工程—交通运输规划与管理]