基于复杂网络的民航跑道侵入事件分析

Analysis of civil aviation runway incursion incident based on complex network

为了明晰跑道侵入事件致因之间的关系及其发生演变规律,收集整理了国内2011—2021年的跑道侵入事件案例,通过对案例进行解析,建立了跑道侵入事件致因网络模型,随后对该网络划分社团;利用复杂网络的各指标分析致因网络的结构特征,并找寻需要防控的关键致因;采用相继故障中的负载容量模型,构建跑道侵入事件致因关联网络模型,挖掘出事故致因最可能形成的致因链。结果表明:网络的直径为5,平均路径长度为2.231,平均聚类系数为0.303;得到培训和经验不足A12、注意力分配不当A6等9项关键致因;挖掘出109条致因链且由注意力分配不当A6引发的跑道侵入事件占比较大。可见,该网络具有较高的聚类系数和较短的传播距离;负载容量模型可对关键致因进行客观分析,避免了只依靠事件数量判断关键致因的偶然性。

To clarify the relationship between the causative factors and explore the law of its evolution,the domestic runway incursion cases from 2011 to 2021 were collected and sorted out.The cases of these events were analyzed from the four aspects of man,machine,environment,and management,and 36 causative factors were identified;Through the correlation analysis between the causative factors of the event,the relationship between 199 causative factors is determined,the causation factors as nodes and the causation relationship between them as the edge in this paper,and the causal model of the network is established and the community is divided;Using degree,clustering coefficient,betweenness centrality,network diameter and average path length of complex networks,the structural characteristics of the causal network are analyzed and the key factors for control are determined;Taking into account the characteristics of the chain incident based on the structural analysis of the causal model,to further revealed the influence of causality among incident factors on the formation process of incidents.The causal correlation network model for runway incursion is established based on the load-capacity model of the theory of cascading failures,and the foremost causation chains of accidents were found.The results show that the diameter of the network is 5,the average path length is 2.231,and the average clustering coefficient is 0.303;9 key causes of runway incursion such as insufficient training and experience A12,improper attention distribution A6,blind operation or action A8,inappropriate scheduling D6 and other factors are obtained;109 causal chains were excavated and runway incursions caused by improper attention distribution A6 accounted for a large proportion.It can be seen that the network has a higher clustering coefficient and shorter causation propagation distance;The key causes are mainly concentrated in three categories:human,machine,and management;The load-capacity model can objectively analyze the key causes and avoid the contingency of relying only on the number of events to determine the key causes.