Graph-based Complexity Forecasts in UK En Route Airspace Using Relevant Aircraft Interactions
Researchers have developed a graph-based system to forecast air traffic complexity and predict Air Traffic Control Officer (ATCO) workload up to 45 minutes in advance. This probabilistic approach uses the number of relevant aircraft pairs as a proxy for workload, adapting an existing algorithm for London Middle Sector (LMS). The refined algorithm demonstrated improved performance with an F1-score of 0.84, outperforming the original by 0.15, and showed a stronger correlation with actual interactions than standard traffic volume predictions. AI
IMPACT This AI-driven tool could enhance air traffic safety and efficiency by providing controllers with advance workload predictions.