Nonlinear effects in examples of crowd evacuation scenarios

Severe accidents with many fatalities have occurred when too many pedestrians had to maneuver in too tight surroundings, as during evacuations of mass events. This demonstrates the importance of a better general understanding of pedestrians and emergent complex behavior in crowds. To this end, we develop both a new microscopic agent-based pedestrian model and also study simplified evacuation scenarios which permit the isolation of relevant nonlinear effects and their systematic investigation. We concentrate on two effects: First, the influence of the position and size of an obstacle in front of an emergency exit on the flux through the exit, and second, the influence of other pedestrians on the route choice of an individual. The first investigation demonstrates the possibility of improving substantially the flow through an exit by placing an obstacle in a suitable way in front of it. The latter shows clearly bistable states and hysteresis effects, indicating the existence of unstable pedestrian flow states in addition to the stable states. Furthermore, this set-up is an example of a radical change of the pedestrian flux by only a small change in the geometry of the evacuation scenario. The results motivate further investigation and eventually engineering use by optimizing the design of large buildings, stations, airports and stadiums for mass events.