A fine discrete floor field cellular automaton model with natural step length for pedestrian dynamics

Abstract

This paper proposes an extended floor field cellular automaton (FFCA) model for evacuation simulations, which integrates the natural step length into pedestrian movement. The model enables pedestrians to occupy multiple grids and extends the interaction area to a disk with a radius determined by the natural step length. These modifications facilitate pedestrian movement in various directions, limited only by the chosen numerical resolution. The model is calibrated to reproduce known density-velocity relations by simulating different evacuation scenarios. The flow rate through the bottleneck scenario matches the experimental results very well. Interestingly, the evacuation time, crowd shape near the exit and average conflict frequency are found to be independent of the discretisation degree, which is different from previous work. Compared with the traditional FFCA model, the proposed model yields larger and more reasonable velocity variation and evacuation path. The main parameters have a similar influence on the evacuation time, indicating that the proposed model retains the robustness of the traditional FFCA model.