Factors for predicting airport surface characteristics and prediction accuracy of the surface management system

NASA Ames Research Center, in cooperation with the FAA, is developing the surface management system (SMS). SMS is an air traffic decision support tool that aids both air traffic controllers and air carriers in collaboratively managing surface movements of aircraft. One of the benefits of SMS is the accurate prediction of future airport surface characteristics. For example, SMS predicts the take-off time for each departure. The accurate estimation of take-off time requires first predicting taxi time which, in turn, requires knowledge of the departure runway to which the controller will assign the flight. This paper focuses on the factors that air traffic control tower (ATCT) controllers currently consider in their allocation of departure runways. Using a logistic regression technique, the potential predictors for departure runway assignments were investigated at Memphis International Airport. The factors that were studied were airport configuration, flight path, ramp area, aircraft type, airline, and weight class. Based on the results of the logistic regression, the strongest predictors for runway assignments were incorporated into the SMS prediction algorithm. The paper reports the take-off time prediction accuracy and shows its sensitivity to the departure runway prediction accuracy.