Predicting Airport Runway Configurations for Decision-Support Using Supervised Learning

One of the most critical tasks in air traffic management is runway configuration management (RCM). It deals with the optimal selection of runways for arrivals and departures based on current and forecast traffic, surface wind speed, wind direction, noise abatement and other airport-specific considerations. In this paper, a methodology using supervised learning is developed to build a predictive model for RCM decision-support using historical data. Related U.S. National Airspace System (NAS) data from 2018 and 2019 (two full years) on current and forecast weather, demand/capacity, etc. is collected, analyzed, and fused together. A variety of supervised learning algorithms are tested for predicting runway configuration, and hyperparameter tuning is carried out to select the best performing model. We validate the models on two airports, CLT (low complexity in RCM operations) and DEN (high complexity in RCM operations). The results show significant promise for the two airports with test accuracy of 93% (CLT) and 73% (DEN). We are confident that the proposed methodology is scalable and generalizable to other airports across the NAS.