The Evolution of AI in Air Traffic Management

This book provides a comprehensive, structured journey through how AI is reshaping ATM, beginning with the historical evolution of the system and the urgency for innovation, followed by a detailed grounding in the data sources that enable AI applications and the core machine learning methods used to extract insights from them. Readers will find accessible explanations of key techniques—including supervised learning, reinforcement learning for real-time decision-making, and deep learning and large language models—paired with concrete aviation use cases such as delay prediction, conflict detection and resolution, rerouting, and procedural automation. The book places strong emphasis on human-centered design, with dedicated chapters exploring how air traffic controllers interact with AI, how trust and cognition are affected, and how effective human-AI hybrid teams can be designed for safety-critical environments. It then looks ahead to emerging paradigms, including agentic AI and increasingly autonomous ATM systems, while maintaining a clear focus on the human role as supervisor and ethical anchor. Concluding with an integrated discussion of operational, technical, legal, and ethical challenges, the book equips readers with both practical insights into current capabilities and a forward-looking roadmap for achieving safe, efficient, and trustworthy AI-enabled air traffic management.