为提升机场运行性能、支撑航空运输决策，提出了基于提升树模型的航空器离场滑行时间预测方法。考虑进场航空器场面运行对离场航空器滑行时间的影响，建立了涵盖四大类、八小类的滑行影响因素特征指标体系，采用提升树方法对离场滑行时间进行了机器学习建模，从多维视角建立了预测性能评价指标。选取上海浦东国际机场进行实例验证表明，所提方法具有较高的预测精度，可显著增强离场航空器滑行性能，并有效提升复杂机场的场面运行效率。
In order to improve airport operation performance, and support air transportation decision-making, we propose a method for aircraft taxi-out time prediction based on boosting tree model. Considering the impact of arrivals on departure taxi-out time, an index system covering four categories and eight sub-categories is proposed to reflect the main factors influencing aircraft taxiing activities. Boosting tree method is applied to traina machine learning model for taxi-out time prediction, and then some prediction performance indices are established from a multi-dimensional perspective. A case study of Shanghai Pudong International Airport shows that the proposed method has a high prediction accuracy, which can significantly improve departure taxiing performance and surface operational efficiency at complex airport systems.