
科学技术与工程

中国技术经济学会
Departure Flights' Taxi-out Time Prediction Based on DDA-SIM-ATT-CatBoost Model: Shenzhen Bao'an International Airport
Zheng-hong XIA , Yan-zhi LI , Xin-lei JIA , Jun-chuan HUANG , Tian LUO
Science Technology and Engineering ›› 2026, Vol. 26 ›› Issue (31) : 13678 -13688.
To address the challenges of insufficient accuracy and error control in predicting departure taxi-out time, a DDA-SIM-ATT-CatBoost prediction model integrating dynamic data augmentation, similarity theory, and an attention mechanism was proposed. First, based on historical operational data, key influencing factors were selected through correlation analysis. Subsequently, dynamic data augmentation technology was employed to expand the training sample distribution, similarity theory was applied to achieve multi-dimensional feature space matching of samples, and an attention mechanism was introduced to adaptively calibrate feature weights. Finally, the CatBoost algorithm was utilized for regression prediction, leveraging its advantages in handling categorical features and complex nonlinear relationships. Comparative and ablation experiments conducted on actual operational data from Shenzhen Bao'an International Airport show that the proposed model achieves prediction accuracies of 74.57%, 89.12%, and 97.76% within error margins of ±120 s, ±180 s, and ±300 s, respectively, with MAPE(mean absolute percentage error), MAE(mean absolute error), and RMSE(root mean square error) values of 10.34%, 87.55 s, and 125.61 s. The model significantly outperforms comparative models in performance. Each improved module contributes positively to enhancing model performance, and the integrated DDA-SIM-ATT-CatBoost model, through synergistic optimization, achieves optimal performance in both prediction accuracy and stability, providing reliable decision support for airport ground operational scheduling.
air transportation / taxi-out time / similarity theory / attention mechanism / dynamic data augmentation
/
| 〈 |
|
〉 |