Experiment and simulation study of emergency evacuation during violent attack in classrooms

Abstract

In recent years, as a disaster, terrorist attacks have occurred throughout the world. However, emergency evacuation behaviors during these incidents were not clear, and the traditional emergency plans were not suitable for such incidents. In this paper, evacuation behaviors under armed assault attack in a classroom were studied based on evacuation experiments. A total of 103 participants took part in three experiments. In each experiment, the attacker's attacking route was set differently to study the impact of the attacking route on evacuation behaviors. Pre-evacuation delay, panic of the evacuees, exit choices, evacuation time, and evacuees’ trajectories in the experiments were all analyzed. The results of the experiments showed that when a terrorist attack occurs, there is a long delay before evacuation, and most of the evacuees were in the state of “observation” before they moved. When one of the participants started to evacuate or shout, other participants would begin to recognize the danger and escape quickly. These three experiments showed that the route of the attacker had a significant impact on the routes and exit choices of the evacuees. Rather than searching for the nearest exit, the primary purpose of evacuees was to keep a safe distance from the attacker. The average speed of the evacuees in these three experiments was 1.07m/s, 0.81 m/s, and 0.84 m/s, respectively. The density distribution during the crowd evacuation process was uneven, with the highest density occurring at the area from the seats to the aisles. The research can provide data support for the design of emergency plans and the computer simulation of the armed assault attack.

Keywords

• Emergency evacuation
• Armed assault attack
• Attacker's route
• Experiment