Taxi-out Time Prediction Model of Departure Aircraft Based on BP Neural Network

Abstract

Accurately predicting the departure aircraft slip out time can effectively improve the airport surface operation efficiency and reduce the operation cost. A prediction model of departure and departure time based on BP neural network is established. The key factors influencing the actual slip out time are analyzed,and the correlation is tested. The model is verified by two weeks' actual operation data of an airport in central and southern China,and the root mean square error(RMSE),mean absolute error(MAE)and MAE percentage are analyzed. The results show that there is a strong correlation between the number of aircraft launched in the same period,the number of aircraft taking off in the same period,the number of aircraft in the same period,the average taxiing time within one hour and the taxiing time of departing aircraft. The taxiing distance,the number of turns and the delay time are related to the taxiing time,but not significant. The time period at which the aircraft takes off is not related to the slip out time. The average slip out time within 1 hour plays an important role in improving the prediction accuracy of the model. The introduction of relevant but insignificant influencing factors plays a certain role in improving the accuracy of prediction results. After the introduction of irrelevant factors,the prediction accuracy of the model will decrease significantly.

Keywords

• taxi-out time prediction
• bp neural network
• one hour average taxi-out time
• departure aircraft
• influencing factors of taxi-out time