Analysis of Fire Lane Obstruction Detection and Resource Dispatch Bottlenecks in High-Density Residential Fire Incidents: A Case Study

DOI: https://doi.org/10.62517/jsse.202608205

Author(s)

Yunwei Wang*

Affiliation(s)

Shenzhen High School of Mathematics and Physics Shenzhen, Guangdong, China *Corresponding Author

Abstract

A severe fire broke out in a high-density residential area of Hong Kong in November 2025. The fire spread rapidly from scaffolding on the exterior walls and escalated to a Level 5 fire within a short period [1]. This incident demonstrates that the efficiency of fire rescue operations in high-density residential areas is not only influenced by the scale of the fire but is also closely related to the accessibility of fire lanes, the timeliness of on-site detection, and the rationality of resource dispatch. Taking this fire incident as a case study, this paper constructs a research framework of “incident reconstruction–obstruction identification–access modeling–dispatch analysis” [2][3], drawing on research approaches from the intelligent security field regarding lightweight detection, edge-cloud collaboration, and dynamic dispatch. First, the fire response process is reconstructed in phases to identify key influencing factors in residential rescue operations. Second, an obstruction detection approach focused on fire lanes is proposed to identify obstacles such as illegally parked vehicles, accumulated debris, construction barriers, and gathered crowds. Third, a fire truck travel time model is established, integrating road width, obstruction index, pedestrian density, and on-site transit time into a unified framework. Finally, the resource scheduling bottlenecks in high-density residential fires are summarized and targeted optimization recommendations are proposed. The study concludes that traditional emergency response methods relying on manual patrols and static contingency plans are no longer sufficient to meet the rapid response demands of complex residential fires. Instead, real-time detection, status assessment, and coordinated scheduling should be integrated to form a more proactive smart firefighting support mechanism.

Keywords

High-Density Residential Fire; Fire Lane Obstruction; Object Detection; Edge-Cloud Collaboration; Resource Dispatch

References

[1]Government of the Hong Kong Special Administrative Region. (November 28, 2025). No. 5 alarm fire in Tai Po. [2] Wei, Q., Li, Y., Lou, P., Yan, J., & Hu, J. (2022). A Study on Face Recognition Methods Based onEdge-Cloud Collaboration. Computer Science, 49(5), 71–77. [3] Wu Di, Zhang Zhe, & Li Qiang. (2024). A lightweight method for small-object detection integrating machine vision and unsupervised domain adaptation. Journal of Heilongjiang University of Science and Technology, 34(2), 329–334. [4] Wu, Yuhang, & Sang, Nong. (2023). Unsupervised Domain Adaptation for Pedestrian Re-identification Based on Consistency Constraints and Label Optimization. Chinese Journal of Image and Graphics, 28(5), 1372–1383. [5] Chen Renfei, Peng Yong, & Li Zhongwen. (2023). A Method for Detecting Floating Objects on Water Based on a Continuous Unsupervised Domain Adaptation Strategy. Systems Engineering and Electronics, 45(11), 3391–3401. [6] Wang, Y. Y., Sun, G. W., Zhao, G. X., & Xue, H. (2022). Unsupervised domain adaptation learning based on self-supervised knowledge. Journal of Software, 33(4), 1170–1182. [7] Jiang Shaozhong, Yao Keming, Chen Lei, Wang Zhongzhou, & Guo Fu’ao. (2023). A Mask-Based Face Recognition Method Using a Hybrid CNN-Transformer Model. Sensors and Microsystems, 42(1), 144–148. [8] Hong Kong Housing Authority. (n.d.). Estate locator: Wang Fuk Court, Tai Po, New Territories. [9] Buildings Department, Hong Kong. (2004). Code of Practice for the Provision of Means of Access for Firefighting and Rescue Purposes 2004. [10] Hong Kong Fire Services Department. (2024). Fire Safety Requirements for Non-Designated Use of Venues.