Simulating the Evacuation of a Subway Station after a Sarin Release

Chemical terrorist attacks in closed areas are a serious threat to national security. In order to deduct the best reponse with regard to treatment and protective measures, simulation is the best tool, either in-silico or as a real-life exercise. The difficulty of the latter is the great number of resources necessary to set up a large-scale realistic exercise while the limitation of the former is that the model needs to be as realistic as possible to draw relevant conclusions. The complexity in designing and running a realistic and large-scale computer simulation model with pedestrian movement and air/gas flows renders studies near-impossible with current computers. To address the complexity issue, agent-based simulation is usually preferred over a Computational Fluid Dynamics (CFD) approach. The aim of this contribution is to present a method to generate a list of victims with both inhaled doses of the chemical warfare agent sarin (NATO designation GB) and traumatic injuries in an emergency subway evacuation scenario. The method couples a simple gas diffusion model with a crowd dynamics model. The interaction of the crowd follows a social force model (SFM) where the overlaps are used to determine a physical injury distribution.