Survey of AI-Enabled IoT for Wildfire Prediction, Detection, and Post-Operations

Wildfires are destructive natural disasters that pose severe risks to ecological systems and human lives. This survey explores the integration of Artificial Intelligence (AI) and Internet of Things (IoT) technologies, collectively known as AIoT, in the management of wildfires across three critical phases: prediction, detection, and post-operations. By leveraging real-time data collection, robust communication protocols, and advanced machine learning (ML) models, AIoT systems address the complex challenges associated with wildfire management. We examine AIoT applications across these phases: prediction, where traditional machine-learning and fuzzy-logic methods have evolved into deep-learning approaches; detection, which combines advanced sensors and communication networks with real-time algorithms; and post-operations, covering damage assessment, vegetation recovery, and ecosystem restoration. We present an integrated operational framework showing how these traditionally isolated phases could function as interconnected components with automated data pipelines and feedback mechanisms. The survey identifies seven critical challenges—including hardware constraints, false alarm rates, and network reliability—along with eight future research directions aimed at advancing the field toward practical, scalable deployment.