Pedestrian Evacuation Simulation Considering Hiding Behavior and Obstacle Configurations Under Violent Attacks

Ma, Yaping; Chen, Jiaxi; Li, Mengle; Chen, Zhiwei; Tong, Yunhe; Deng, Qing; Huo, Feizhou · 2025 · DOAJ

DOI: 10.1007/s13753-025-00687-6

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical gap in pedestrian evacuation simulation regarding violent attacks, specifically focusing on the under-researched dynamics of hiding behavior and obstacle interactions. While existing models often treat crowds as fluids or focus solely on attacker-pedestrian interactions, they frequently overlook how pedestrians actively seek shelter behind obstacles when immediate escape is impossible. Motivated by the increasing frequency of knife attacks in public spaces, the authors aim to understand how these protective behaviors and environmental configurations influence evacuation efficiency and fatality rates. To achieve this, the researchers developed an extended cellular automaton model that couples interactions between attackers, pedestrians, and static obstacles. The simulation space is discretized into 0.4 m × 0.4 m cells, utilizing a Moore neighborhood for movement decisions. The model distinguishes three types of pedestrians based on their visual field and memory of the attacker: Type-I (attacker visible), Type-II (attacker in memory but currently obscured), and Type-III (attacker never seen). A key innovation is the explicit modeling of "hiding behavior," where Type-I pedestrians decide between moving toward exits or obstacle hiding zones based on a dynamic decision-making process. This process incorporates a "hiding intensity" parameter, the risk of waiting at crowded exits, and the shortest path distances calculated via the A* algorithm. Attackers are modeled to move toward pedestrians within their visual field, with obstacles creating visual blind spots that affect both attacker targeting and pedestrian safety perceptions. The simulation results, derived from 100 trials in a 10 m × 10 m room, reveal that an optimal individual hiding intensity is essential for minimizing fatalities and enhancing evacuation efficiency; both insufficient and excessive hiding intensity lead to worse outcomes. The study found that obstacle configuration significantly impacts evacuation performance. Specifically, there is an optimal obstacle length that maximizes safety, and placing a single obstacle perpendicular to the exit yields the most effective results. Furthermore, exit configuration plays a crucial role: placing two exits on opposite walls or a single exit in a corner is more effective than other arrangements. Increasing exit width and the number of exits significantly shortens evacuation time. The spatial distribution of fatalities also shifts with hiding intensity; low intensity concentrates deaths near exits, while high intensity shifts fatalities toward the obstacles as pedestrians cluster there for protection. The significance of this work lies in its detailed characterization of pedestrian decision-making under threat, moving beyond simple obstacle avoidance to model active shelter-seeking. The findings provide actionable insights for emergency planning and architectural design, suggesting that strategic placement of obstacles and exits can substantially reduce casualties during violent attacks. By quantifying the trade-offs between hiding and fleeing, the model offers a more realistic framework for simulating disaster scenarios involving human adversaries, contributing to improved public safety strategies in crowded environments.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• evacuation egress