Real-time predictive modeling of flight delays using distributed systems and machine learning

This study presents a real-time system for flight delay prediction using distributed systems and machine learning. By integrating flight data from ADS-B signals, METARs, and TAF weather reports, the system processes the data streams via a reactive architecture. The applied predictive models, including random forests and linear regression, were validated and evaluated for their accuracy and scalability. The developed system offers a practical solution for improving decision-making in air traffic management.