Optimizing emergency evacuation: An in-depth review of multi-agent pathfinding (MAPF) for escape routes

View at publisher Summarise with VIRAT
Authors: Nurul Ishafira Ismail, Siti Normaziah Ihsan, Tuty Asmawaty Abdul Kadir
Cited by: 0
References: 11

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Emergency escape in enclosed environments, such as buildings, poses obstacles in ensuring fast and safe exits. Multi-agent pathfinding (MAPF) is a computational technique that involves coordinating agents to identify collision-free pathways. However, determining appropriate pathways during an emergency remains challenging, particularly in challenging situations. This study reviews various MAPF algorithms for optimizing escape routes in different scenarios, with an emphasis on speed, safety, and closeness to exits. It evaluates algorithms such as A*, PRM, RVO, and reinforcement learning in real-world scenarios with crowd density, dynamic obstacles, and restricted space. The review highlights MAPF’s flexibility across various scenarios, including fires, floods, and ship evacuations. The results demonstrate that hybrid algorithms significantly improve evacuation efficiency by guiding agents to exits quickly and safely. Future work should integrate MAPF with real-time decision-making systems for better emergency responses. In conclusion, MAPF holds promise for improving emergency evacuations, with potential for further advancements in both algorithmic efficiency and adaptability.

Fields of Study & Subjects

Evacuation and Crowd Dynamics Robotic Path Planning Algorithms Maritime Navigation and Safety

Publication details

Journal / source
AIP conference proceedings
Volume
3462
Pages
150001–150001
Publication date
1 January 2026
Language
EN