Restoring subway station emergency evacuation with improved social force model from the perspective of small groups

Currently, most research on subway evacuation focuses on individual pedestrians, neglecting the small group of pedestrians. This paper introduces the probability of small group composition, the attraction of small group members, and the relationship between groups based on the traditional social force model. It analyzes the different force conditions of the members in different small groups and examines the influence of small groups on their members. Based on this, the social force emergency evacuation model was established, considering the behavioral influence of small groups, and simulating the Puhe subway station in Shenyang by Anylogic. Using the shortest path algorithm, a logical model for pedestrian evacuation was constructed. The results show that a mixed evacuation strategy involving both individual pedestrians and small groups better replicates real-life scenarios in metro station emergencies. According to the simulation results, the evacuation efficiency of the evacuation mode combined with the concept of small group is obviously more in line with reality, and the evacuation efficiency of the stranger group is the fastest, followed by the group of friends and the group of family members is slower. This provides valuable insights for evacuation studies based on simulations.