A Review of AI-Based Fire Evacuation Automation Systems: Current Practices, Challenges, and Future Enhancements

This paper explores the evolving landscape of AI-powered fire evacuation automation systems, analyzing their benefits, drawbacks, and potential for improvement. These systems utilize diverse AI methodologies, including machine learning, computer vision, and reinforcement learning, to analyze real-time data and guide occupants to safety during fires. Examples like SFES-HB (Smart Fire Evacuation System in High-Rise Buildings), ARAE (Autonomous Robot-Assisted Evacuation), and WFEAS (Warehouse Fire Evacuation Automation System) showcase the various applications of these systems across different settings. While AI-powered systems offer advantages like real-time adaptability, reduced human error, and scalability, they also present challenges such as complexity, cost, and dependence on data quality. To address these limitations, the paper proposes advanced AI techniques like deep learning for behavior prediction, reinforcement learning for dynamic decision-making, and computer vision for enhanced situational awareness. Implementation strategies include hybrid systems that combine AI with human decision-making, simulations for training emergency personnel, and integration with IoT devices for more precise data collection. The paper concludes that the future of fire evacuation lies in the continued development and refinement of AI systems, coupled with human supervision and robust support structures, to ensure effective and safe evacuation procedures in complex and dynamic environments.