No abstract

This paper describes the development of an automated mechanism to alert aviation traffic managers of the need to take action to adjust the rate of aircraft arriving into airports. When rates are too high, air traffic controllers are forced to do costly maneuvering and to hold aircraft to maintain required spacing. When arrival rates are too low, valuable airport landing capacity goes unused. In today's operations, mismatches between the planned and actual arrival rates often occur gradually and may not even be noticed until too late, after significant problems have materialized. This paper proposes an alerting mechanism that uses realtime signal metrics based on actual airspace operations to alert controllers of impending problems. Alerts would be triggered when signal metrics crossed their respective threshold values, which would be tailored for specific airspaces and generated with sufficient lead time to allow for mitigating actions. The alerting mechanism would reduce reliance on manual monitoring and thus reduce traffic manager workload. With historical flight data from airspace surrounding Atlanta International Airport an initial predictive model was developed and validated for one possible signal metric. Through discussions with subject matter experts, an analysis of various metric threshold values was also performed.