Simulations of bi-direction pedestrian flow using kinetic Monte Carlo methods

Abstract We present a two-dimensional (2D) lattice model based on the exclusion principle and Arrhenius microscopic dynamics to study bi-direction pedestrian flows. This model implements stochastic rules for pedestrians’ movements based on the configuration of the surrounding conditions of each pedestrian, pedestrian–pedestrian interactions, and their walking preference. Our rules simplify tactically the decision-making process of pedestrians in their movements and can effectively reflect the behaviors of pedestrians at the microscale while attaining realistic emergent macroscale activity. Our computational approach is an agent-based method, which uses an efficient list-based kinetic Monte Carlo (KMC) algorithm to evolve the pedestrian system. The simulations focus on two aspects: different directional splits of the pedestrians and different strengths of walking preference for the right-hand side. Both results exhibit a phase transition from freely flowing to fully jammed, as a function of initial density of pedestrians. At different phases the relationships of density–flow and density–velocity are different from each other. The KMC simulations show some pedestrian flow self organization phenomena including lane formation phases and reflect transition trends of the corresponding empirical data from real traffic.