Extrapolating the Sustainability Benefits of Collaborative Digital Departure Reroute

A method for extrapolating the benefits and delay savings of NASA’s Collaborative Digital Departure Reroute (CDDR) technology to major terminal areas across the National Airspace System is developed and initial results presented for a year’s worth of flight departures for 10 of the busiest terminal areas in the US. Features related to route options, delay, and weather impact are derived from Coded Departure Routes (CDR) and Aviation System Performance Metrics (ASPM) and used to predict the counts of CDDR proposed flight reroutes. Machine Learning (ML) and Linear Regression models are independently developed and trained on a year of data from CDDR deployed at the Dallas Fort-Worth terminal. Similarity in yearly count prediction trends between these two models when applied across 10 terminal areas lends confidence to the methodology. Estimates of delay per flight for each terminal area are then combined with the CDDR reroute count predictions to derive yearly benefits in terms of delay and fuel savings.