@article{ZHAO2025, 
author = {Dongyue ZHAO and Xuan PENG and Shanning HU and Zhifang ZHU and Mengni LIU and Shihua REN},
title = {Impact of security patroller numbers on accident warning sign detection efficiency under visual obstruction},
year = {2025},
journal = {Journal of Tsinghua University (Science and Technology)},
volume = {65},
number = {6},
pages = {1070-1078},
keywords = {visual obstruction, security patroller, accident warning sign, optimization of allocation},
url = {https://www.sciopen.com/article/10.16511/j.cnki.qhdxxb.2025.22.023},
doi = {10.16511/j.cnki.qhdxxb.2025.22.023},
abstract = {ObjectiveIn densely populated venues, early warning signs of safety accidents triggered by portable flammable items (e.g., power banks) are often difficult to detect. This challenge necessitates effective security patrol inspections. Although recent research has focused on optimizing patrol routes, resource allocation, and deploying unmanned aerial vehicles, manual inspections remain the primary method. Moreover, visual obstructions caused by high crowd density and structural limitations further complicate the detection process. Consequently, the optimal allocation of security patrol personnel is critical for enhancing early warning sign detection. This paper addresses the impact of visual obstructions and investigates how the number of security patrol personnel influences detection efficiency in densely populated venues.MethodsThis study examined the effect of the number of security patrol personnel on the efficiency of early warning sign detection and explored the underlying mechanisms. The investigation was conducted in three steps. First, behavior rules for pedestrians and security patrol personnel were established using the classical social force model. These rules were tailored to reflect the characteristics of densely populated venues and incorporated key factors such as interpersonal interactions and environmental constraints. Second, a dynamic visual field model for security patrol personnel was developed under three conditions: unobstructed vision, obstructed vision with non-overlapping blind spots, and obstructed vision with overlapping blind spots. By dynamically calculating the visual coverage area in real time, the model enabled adaptive adjustments of the visual field in response to changes in crowd density and obstructions. Third, eight sets of scenarios were designed based on variations in crowd density and typical spatial factors of densely populated venues. MATLAB was used to simulate detection times when one to eight security patrol personnel patrolled along random paths with early warning signs placed at random locations and their visual field trajectories recorded. Two hundred simulation runs were performed to ensure the robustness and reliability of the results.ResultsThe simulation shows that the dynamic visual field model effectively monitors crowd density within a specified range along the patrol direction. The model dynamically calculates the visual coverage area while accounting for both unobstructed vision and overlapping blind spots, thereby enabling adaptive adjustments to maintain optimal monitoring conditions. Moreover, the simulation data reveals that increasing the number of patrol personnel significantly reduces the time required to detect early warning signs. However, the improvement effects exhibit a trend of diminishing effects, following a negative power-law relationship as personnel numbers increase. In larger spaces or high-density environments, a fixed number of patrol personnel require substantially longer detection times, whereas a moderate increase in personnel yields more effective reductions in detection time. Furthermore, the simulations demonstrates that as the number of patrol personnel increases, the time required for the average visual coverage to reach its upper limit gradually decreases and eventually stabilizes. This finding suggests that the marginal benefit of adding extra personnel declines beyond a certain point, thereby limiting further reductions in detection time. The diminishing marginal effect is especially pronounced in smaller venues or environments with lower crowd density.ConclusionsThis study clarifies how the number of security patrol personnel affects the efficiency of early warning sign detection under visual obstruction conditions and provides insights into the underlying mechanisms. The findings help improve detection efficiency in densely populated venues and offer theoretical and methodological support for developing more effective patrol strategies.}
}