Optimizing Flight Departure Delay and Route Selection under En Route Convective Weather

Abstract

This paper presents an optimization model and a priority-based scheduling algorithm for managing air traffic flow in the United States. The models assign departure delays and pre-departure reroutes to aircraft whose trajectories are predicted to cross weather-impacted regions of the National Airspace System. It is ensured that the 4-dimensional trajectories of flights remain free of conflict with weather when the forecasts are deterministic. The optimization model and the scheduling algorithm are applied to solve a large-scale traffic flow management problem with realistic weather data and flight schedules. Experimental results indicate that allowing rerouting can reduce departure delays by nearly 67%, but it is associated with an increase in total airborne time due to longer routes flown by aircraft. The performances of the optimization model and the priority-based scheduling algorithm were compared. Delays obtained from the optimization model were 5% lower than those from the scheduling algorithm. However, the priority-based scheduling algorithm was about 5-times faster in generating solutions than the optimization model. This paper also discusses how airline rerouting preferences and some of the latest concepts in Collaborative Decision Making can be incorporated into the proposed models.