Impact of Water Depth on Subway Emergency Evacuation in Flood Scenarios

[Objective] Evacuation bottlenecks for subway stations in flood scenarios are investigated to enhance subway station personnel evacuation efficiency under water inundation and backflow conditions, and provide insights for developing subway evacuation emergency plans. [Method] Pedestrian walking tests are conducted using drone aerial footage. OpenCV (open source computer vision) library image processing algorithms are employed to analyze pedestrians walking paths and calculate their walking speeds in carious water depth scenarios. Based on the social force model and in line with subway safety evacuation standards, an emergency evacuation simulation model for subway stations in flood scenarios is established and then simulated using Anylogic. [Result & Conclusion] The test results show that the deeper the water depth, the slower the pedestrian passage speed, and the greater the amplitude of the left-right swing of the pedestrians to maintain balance. If support structures are added in critical areas, pedestrian walking efficiency can be effectively improved. Taking a sunken plaza subway station in a coastal city as an example, different degrees of inundation and backflow are analyzed through simulation. Results show that the water depth has a significant impact on the walking speed and evacuation efficiency of pedestrians. When water depth is over 5 cm, the evacuation time may exceed the requirements of evacuation specifications. When water depth reaches 20 cm, the proportion of people safely evacuated within 6 minutes is 26.38%. Failure to organize timely and effective evacuation during the flood event will seriously threaten the evacuation safety of the subway station.