Relationship between behavior aggressiveness and pedestrian dynamics using behavior-based cellular automata model

Abstract Crowds in public spaces such as subway station are generally composed by diverse types of pedestrians with various motion features and behavior preferences. This paper aims to propose a behavior-based cellular automata model that can express the heterogeneous structure of crowds and explore the influence of different crowd composition on pedestrian dynamics, especially evacuation efficiency. The degree of aggressiveness is first introduced into cellular automata model as the internal state, which affects the motion characteristics and behavior preference of pedestrians. It is closely related to the motion characteristics including free walk velocity, target cells selection and conflict resolution mechanism. In addition, it determines the balance selection between keeping shorter paths and avoiding congestion. Simulation experiments of unidirectional pedestrian flow is conducted in the corridor with single exit. Simulation results indicate that a higher shortest path tendency leads to more time spent on blocking, and a higher detours tendency will cause longer detours time. A good balance between the two tendencies can make the individual's evacuation more efficient. A higher proportion of aggressive people and moderate degree of aggressiveness results in a higher average evacuation efficiency. In the crowd, the evacuation of conservative groups (i.e. pedestrians with zero aggressiveness) will be the most advantageous when the proportion of aggressive groups and degree of aggressiveness are moderate. For aggressive groups (i.e. pedestrians with the degree of aggressiveness greater than zero), evacuation efficiency will be the highest when the proportion of aggressive groups is moderate and the degree of aggressiveness is high. The proposed model provides a new perspective for the evaluation and analysis of heterogeneous pedestrian flow.