Artificial Intelligence and Data Analytics in Fire Science: Detection, Modeling, and Suppression

摘要

Fire science has advanced rapidly in recent decades, yet fire events continue to cause significant loss of life, infrastructure, and natural ecosystems. Traditional approaches to fire detection, dynamics modeling, and suppression strategies often face limitations in speed, accuracy, and scalability. The emergence of artificial intelligence (AI), machine learning (ML), and data analytics is transforming fire science by enabling early detection, predictive modeling, and optimization of suppression strategies. This review highlights state-of-the-art applications of AI-driven methods in fire detection, modeling fire growth and smoke propagation, and improving suppression techniques. Key challenges, opportunities, and future research directions are discussed, focusing on how data-driven fire science can improve public safety and resilience against wildfires and structural fires. Fire remains one of the most destructive hazards worldwide, causing extensive human, economic, and ecological losses. Despite advances in fire science, traditional approaches to fire detection, dynamics modeling, and suppression are often constrained by slow response times, computational intensity, and limited adaptability to complex environments. The rapid development of AI, ML, and data analytics offers new opportunities to overcome these challenges by enabling early warning systems, predictive modeling, and data-driven suppression strategies. This review synthesizes recent research on AI-enhanced fire science, focusing on three key areas: detection, modeling, and suppression. In detection, computer vision methods, IoT sensor networks, and predictive analytics improve early identification of ignitions while reducing false alarms. In modeling, surrogate AI frameworks and physics-informed neural networks accelerate simulations of fire dynamics, heat transfer, and smoke propagation, supporting real-time decision-making. In suppression, intelligent resource allocation, autonomous firefighting robots, and smart suppression systems highlight the role of AI in optimizing operations and reducing risks to human responders. Overall, AI-driven fire science has the potential to reduce casualties, mitigate economic losses, and build fire-resilient societies by complementing traditional methods with adaptive, data-driven intelligence.