Pedestrian Evacuation Simulation Considering Hiding Behavior and Obstacle Configurations Under Violent Attacks
摘要
Analyzing pedestrian evacuation dynamics under violent attacks is critical for understanding the risk evolution of this specific disaster scenario. One particularity of pedestrian movement under violent attacks is that pedestrians would intend to hide toward obstacles for protection when they cannot escape from exits immediately. However, the hiding behavior and obstacles have attracted little attention in existing simulation studies. To fill this gap, an extended cellular automaton model for pedestrian evacuation coupling interactions between attackers, pedestrians, and obstacles was constructed. In the model, the motion and assaultive behaviors of the attacker are modeled. The movement rules of three types of pedestrians are elaborated respectively. In particular, the hiding behaviors are taken into account, the decision-making process of Type-I pedestrians to move towards obstacle hiding zones or exits is characterized, and a novel exit selection strategy for Type-I pedestrians is proposed. The effects of critical factors on pedestrian evacuation dynamics are fully analyzed. Results indicate that an optimal individual hiding intensity is essential for reducing fatalities and enhancing evacuation efficiency. There is an optimal length of the obstacle to maximize the evacuation performance, while the location and number of obstacles, as well as the exit configuration, have considerable impacts on pedestrian evacuation time and death tolls. Placing a single obstacle perpendicular to the exit achieves the most effective evacuation outcomes. Additionally, placing two exits on opposite walls or a single exit in the corner is more useful. Increasing exit width and numbers can significantly shorten evacuation time for crowds.