Modeling following behavior and right-side-preference in multidirectional pedestrian flows by modified FFCA

Abstract

Pedestrian movement modeling is a popular out-of-equilibrium problem in statistical and computational physics. As a kind of typical pedestrian movement, multidirectional flow is quite common in real-life, and examples include the bidirectional flow in corridors and cross flow at intersections. In the discrepancies of the multidirectional flows, the behaviors of pedestrians should play a crucial role. Therefore, in this paper, the following behavior and the right-side-preference are investigated in three different types of multidirectional flow. By the floor field cellular automaton (FFCA), the dynamic floor field is redefined, and the right-preferred floor field and the order parameter for lanes formation are formulated. Then, the fundamental diagram, lanes formation, density distribution and passing time are analyzed in the multidirectional flows considering the influence of following behavior and right-side-preference. Finally, the simulation is compared with the empirical data, indicating that the proposed FFCA models the multidirectional flow well.