Pedestrian Evacuation Modeling and Simulation Under Small-Group and Multi-Exit Conditions: Presenting a Novel Framework

Pedestrian evacuation modeling and simulation (PEMS) plays an important role in analyzing and understanding individual and crowd behaviors in disastrous scenarios such as fires, floods, terrorism, and national emergencies. Based on observations and statistical analysis from PEMS, appropriate evacuation plans can be formulated so as to significantly reduce casualty numbers in such situations. Recent work on small-group-based pedestrian evacuation investigates how people often move in social groups during evacuations. However, there is a lack of comprehensive studies to understand the impact of various factors occurring in small-group evacuations, such as the group size, whether there is a leader in the group or not, and the influence of other conditions (e.g., multi-exit evacuations). In this article, we present a novel PEMS framework to understand the effect of small-group scenarios for pedestrian evacuation. In particular, we introduce an evacuation model with multiple conditions, including small-group forces and multi-exit conditions, under a realistic scenario. Furthermore, we perform a comprehensive small-group-based pedestrian evacuation simulation to gain insights into how various factors, such as group size and leadership, impact evacuation efficiency. Based on our experimental analysis, it is found that through heuristic settings on evacuation groups, e.g., group size and assignment of leadership, it is possible to effectively improve evacuation efficiency, thus providing useful guidance to evacuation plans.