基于关联规则的铁路事故致因网络构建与分析

Construction and Analysis of Railway Accident Causation Network Based on Association Rules

为高效、精准地处理大量铁路事故数据、提升铁路运输系统安全性,结合关联规则挖掘和复杂网络理论,对大量铁路事故调查报告进行预处理,提取致因因素并建立铁路事故致因数据集,应用Apriori算法挖掘铁路事故致因关联规则,在此基础上,构建铁路事故致因网络,分析网络的社团结构和拓扑特征。案例分析表明:铁路事故致因网络存在着安全培训不到位、安全检查不到位、司机操作不当等问题,人因层和设备层节点与其他节点的关系更为紧密,管理层节点更容易对其他节点产生间接影响。

In order to deal with large amounts of railway accident data efficiently and accurately and improve the safety of the railway transportation system, combined with association rule mining and complex network theory, a large number of railway accident investigation reports were preprocessed, the causes were extracted, and the data set of railway accident causes was established. Apriori algorithm was applied to mine the association rules of railway accident causation, furthermore, the railway accident causation network was constructed, and the community structure and topological characteristics of the network were analyzed. The case analysis shows that there are some problems in railway accident causative network, such as inadequate safety education, the lack of safety inspection and improper operation of train drivers. Nodes of human layer and equipment layer are more closely related to other nodes, while nodes of management layer are more likely to have an indirect impact on other nodes.