Multimodal-Transport Collaborative Evacuation Strategies for Urban Serious Emergency Incidents Based on Multi-Sources Spatiotemporal Data (Short Paper)

When serious emergency events happen in metropolitan cities where pedestrians and vehicles are in high-density, single modal-transport cannot meet the requirements of quick evacuations. Existing mixed modes of transportation lacks spatiotemporal collaborative ability, which cannot work together to accomplish evacuation tasks in a safe and efficient way. It is of great scientific significance and application value for emergency response to adopt multimodal-transport evacuations and improve their spatial-temporal collaboration ability. However, multimodal-transport evacuation strategies for urban serious emergency event are great challenge to be solved. The reasons lie in that: (1) large-scale urban emergency environment are extremely complicated involving many geographical elements (e.g., road, buildings, over-pass, square, hydrographic net, etc.); (2) Evacuated objects are dynamic and hard to be predicted. (3) the distributions of pedestrians and vehicles are unknown. To such issues, this paper reveals both collaborative and competitive mechanisms of multimodal-transport, and further makes global optimal evacuation strategies from the macro-optimization perspective. Considering detailed geographical environment, pedestrian, vehicle and urban rail transit, a multi-objective multi-dynamic-constraints optimization model for multimodal-transport collaborative emergency evacuation is constructed. Take crowd incidents in Shenzhen as example, empirical experiments with real-world data are conducted to evaluate the evacuation strategies and path planning. It is expected to obtain innovative research achievements on theory and method of urban emergency evacuation in serious emergency events. Moreover, this research results provide spatial-temporal decision support for urban emergency response, which is benefit to constructing smart and safe cities.