Journal of Construction Automation and Robotics ISSN:2800-0552(Print) 2951-116X(Online) 건설자동화·로보틱스 논문집

The spatial characteristics of underground facilities of limit ventilation capabilities, the constrained direction of escape due to ascending structures make it highly likely for secondary accidents or large-scale casualties to occur in the event of a disaster within subway stations. Instead of fixed disaster response manuals, which have limitations in real-time disaster response, there is a demand for intelligent disaster response systems that can actively guide disaster responses. Also, disaster vulnerability analysis during evacuation should be backed to derive optimal disaster evacuation routes and establish response measures. In this paper, we propose a framework that models subway stations as a network model and evaluates disaster vulnerability and evacuation convenience using network centrality theory. Focusing on the Gwanggyo Jungang Station of the Seoul Shinbundang Line, we divided the station space into equal sections and set them as nodes, and directly connectible nodes were linked with edges to construct a network model. Using betweenness centrality, one of the centrality theories, as an indicator, we analyzed the disaster vulnerability and evacuation convenience and verified the effectiveness of the proposed framework. From the analysis, we confirmed that the betweenness centrality is concentrated in specific areas, enabling the identification of disaster-vulnerable (important) spaces that control the entire disaster evacuation flow.