由於客運量大的機場於尖峰時段中，仍然必須做好機場流量管制，因此航空管制人員不僅要迅速安排進場航空器的降落順序，還必須讓航空器間維持一定的隔離，以遵守國際民航組織規定的隔離規範。此篇論文主要探討如何利用機器學習工具實現航空器的降落排程。此研究主要分成兩個部分，第一部分是選出各項供機器學習演算法辨識的特徵，特徵包含航空器的機型以及航空器間的隔離時間，接著篩選將特徵編譯成電腦能判讀的編碼方式，本篇論文使用的編碼方式，是把整組航班拆解成一架一架的航空器，再將每架航空器依照機型，分別由數值取代，篩選編碼方式的目的在於考量預測結果在排程功能上的應用。最後以最佳化排程數據為參考解答，透過隨機森林演算法建構出機器學習模型，此模型能夠在短時間內，對尚未排程且規模龐大的航班進行預測，找出理想的排程結果。第二部分為透過循序二次規劃法調整航空器在時間視窗內的飛行速度，時間視窗的結束時間點為航空器抵達機場前50分鐘，長度則由航班路線決定，當航空器在時間視窗內以調整後的飛行速度飛行，離開時間視窗時就能自動以機器學習預測的降落排序飛行，同時縮小預計到場時間與調整速度後的更新到場時間的變動量，並能符合國際民航組織的隔離時間規範。最後透過桃園國際機場的歷史資料，驗證此研究的成果，模擬的結果顯示機器學習預測的降落排序，能減小航班完成降落所需的時間。

Flow control is one of the major challenges for busy airports during rush hours. Therefore, air traffic controllers not only have to provide advisory services to approaching and departing aircraft but also make each two aircraft keep a safe separation in order to follow the rule of International Civil Organization (ICAO). The main purpose of this research is to discuss how to use machine learning techniques to schedule arriving aircraft. This research is divided into two parts. The first part is building a machine learning model based on flight data. The features of flight information created for training model include the classification of the aircraft and the separation time between all aircraft. And we select the encoding method for string data type. Using these encoded features can make the program tell the difference between predictions and decide the priority of aircraft landing. The result of mixed-integer programming is considered as the expected output result. The expected output result is for machine learning model to verify the result of the training and measure the accuracy. Then we build machine learning model through random forest algorithm to predict the landing sequence of unknown flight. The second part is calculating the speed of approaching aircraft within time window. Time window was set in the en-route section and end 50 minutes before arriving the airport. This speed can reduce the difference between estimated arrival time and suggested arrival time. And the flight will follow the landing sequence of arriving aircraft predicted by machine learning when leaving the time window. Meanwhile, the separation of each two aircraft comply with ICAO regulations. Finally, historical data for the Taiwan Taoyuan international airport is used to verify the consequent of our approach.

[1]蔡昆哲，2007，飛航管制人員值勤疲勞影響因素之探討，碩士論文，國立
台灣海洋大學。
[2]許悅玲，鄭永安，楊啟良，2018，我國國籍航空公司疲勞管理之研究：從
理論到實務，第5卷第4期，第331-351頁。
[3] ACI. "Annual World Airport Traffic Forecasts 2017-2040," Airports Council International, 2017.
[4] OAG. "On-time performance for airlines and airports and TOP 20 busiest routes," OAG Punctuality League, 2018.
[5] FAA. "UPDATE to The Business Case for the Next Generation Air Transportation System based on the Future of the NAS Report," U.S. Department of Transportation, 2016.
[6] FAA. "Future of the NAS," U.S. Department of Transportation, 2016.
[7] Caprı̀, S., and Ignaccolo, M. "Genetic algorithms for solving the aircraft-
sequencing problem: the introduction of departures into the dynamic model,"
Journal of Air Transport Management Vol. 10, No. 5, 2004, pp. 345-351.
[8] Beasley, J. E., Krishnamoorthy, M., Sharaiha, Y. M., and Abramson, D.
"Scheduling aircraft landings—the static case," Transportation science Vol. 34,
No. 2, 2000, pp. 180-197.
[9] Zhang, X., Zhand, X., Zhang, J., and Liu, B. "Optimization of Sequencing for
Airport Arrival Based on Approach Routes," 2007 IEEE Intelligent
Transportation Systems Conference, 2007, pp. 592-596.
[10] Dear, R. G. "The dynamic scheduling of aircraft in the near terminal area." Cambridge, Mass.: Flight Transportation Laboratory, Massachusetts Institute, 1976.
[11] Dear, R. G., and Sherif, Y. S. "The dynamic scheduling of aircraft in high density
terminal areas," Microelectronics Reliability Vol. 29, No. 5, 1989, pp. 743-749.
[12] Wang, T.-C., and Li, Y.-J. "Optimal Scheduling and Speed Adjustment in En Route Sector for Arriving Airplanes," Journal of Aircraft Vol. 48, No. 2, 2011, pp. 673-682.
[13] Wang, T. C., and Tsao, C. H. "Time-Based Separation for aircraft Landing Using Danger Value Distribution Flow Model, " Mathematical Problems in Engineering Vol. 2012, 2012, pp. 16.
[14] Wang, T.-C., and Chen, T.-C. "Arrival and Departure Aircraft Scheduling with
Turbulence Interaction Concept," Journal of Aircraft Vol. 53, No. 5, 2016, pp.
1201-1210.
[15] McCulloch, W. S., and Pitts, W. "A logical calculus of the ideas immanent in
nervous activity," The bulletin of mathematical biophysics Vol. 5, No. 4, 1943,
pp. 115-133.
[16] Rumelhart, D. E., Hinton, G. E., and Williams, R. J. "Learning representations by
back-propagating errors," nature Vol. 323, No. 6088, 1986, pp. 533-536.
[17] LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W.,
and Jackel, L. D. "Backpropagation applied to handwritten zip code recognition,"
Neural computation Vol. 1, No. 4, 1989, pp. 541-551.
[18] Yu, H., and Liang, W. "Neural network and genetic algorithm-based hybrid approach to expanded job-shop scheduling," Computers & Industrial Engineering Vol. 39, No. 3-4, 2001, pp. 337-356.
[19] Freund, Y., and Schapire, R. E. "Experiments with a new boosting algorithm," icml. Vol. 96, Citeseer, 1996, pp. 148-156.
[20] Rätsch, G., Onoda, T., and Müller, K.-R. "Soft margins for AdaBoost," Machine learning Vol. 42, No. 3, 2001, pp. 287-320.
[21] Polikar, R. "Ensemble learning," Ensemble machine learning. Springer, 2012, pp. 1-34.
[22] Zhou, L., Pan, S., Wang, J., and Vasilakos, A. V. "Machine learning on big data:
Opportunities and challenges," Neurocomputing Vol. 237, 2017, pp. 350-361.
[23] Cruz, J. A., and Wishart, D. S. "Applications of machine learning in cancer
prediction and prognosis," Cancer informatics Vol. 2, 2006, p. 19.
[24] Rosten, E., and Drummond, T. "Machine learning for High-Speed Corner Detection, " Computer Vision Vol. 3951, 2006, pp. 430-443.
[25] Maxwell, A. E., Warner, T. A., and Fang, F. "Implementation of machine-learning
classification in remote sensing: An applied review," International Journal of
Remote Sensing Vol. 39, No. 9, 2018, pp. 2784-2817.
[26] Verrelst, J., Muñoz, J., Alonso, L., Delegido, J., Rivera, J. P., Camps-Valls, G., and Moreno, J. "Machine learning regression algorithms for biophysical parameter retrieval: Opportunities for Sentinel-2 and-3," Remote Sensing of Environment Vol. 118, 2012, pp. 127-139.
[27] Li, Y.-J. "Optimal Scheduling and Speed Adjustment for Arriving Airplanes," master, Institute of Civil Aviation, Cheng Kung University, Taiwan, 2010.
[28] FAA. "Air traffic control," U.S. Department of Transportation, 2018.
[29] FAA. "Pilot and Air Traffic Controller Guide to Wake Turbulence," U.S. Department of Transportation, 2020.
[30] Aviation Stack Exchange (2017). Which Terminal Arrival Area depiction is correct between NACO and Jeppesen plates? Available：https://aviation.stackexchange.com/questions/45548/which-terminal-arrival-area-depiction-is-correct-between-naco-and-jeppesen-plate
[31] FAA. "Instrument Flying Handbook," U.S. Department of Transportation, 2001.
[32] ICAO. "Air Traffic Management for Air Navigation Services Doc 4444," 2016.
[33] Elsa, F., Jean-Pierre, N., Antoine, V., and Peter, C. “Potential Benefits of a Time-based Separation Procedure to maintain the Arrival Capacity of an Airport in Strong- Head-Wind Conditions,” Proceedings of the 5th USA/Europe Airfic Traf Management Research and Development Seminar (ATM 2003), 2003.
[34] Seber, G. A., and Lee, A. J. "Straight-line regression," Linear Regression Analysis. New York, John Wiley & Sons, 1977, pp. 177-213.
[35] Nash, S. G. Linear and nonlinear programming: McGraw-Hill Science,
Engineering & Mathematics, 1996.
[36] Peng, C.-Y. J., Lee, K.-L., and Ingersoll, G. M. "An introduction to logistic
regression analysis and reporting," The journal of educational research Vol. 96,
No. 1, 2002, pp. 3-14.
[37] Zhang, S., Li, X., Zong, M., Zhu, X., and Wang, R. "Efficient knn classification with different numbers of nearest neighbors," IEEE transactions on neural networks and learning systems Vol. 29, No. 5, 2017, pp. 1774-1785.
[38] Pal, M., and Foody, G. M. "Feature selection for classification of hyperspectral
data by SVM," IEEE Transactions on Geoscience and Remote Sensing Vol. 48,
No. 5, 2010, pp. 2297-2307.
[39] Kulkarni, V. Y., and Sinha, P. K. "Pruning of random forest classifiers: A survey and future directions," 2012 International Conference on Data Science & Engineering (ICDSE). IEEE, 2012, pp. 64-68.
[40] Segal, M. R. "Machine Learning Benchmarks and Random Forest Regression," Division of Biostatistics, University of California, San Francisco, 2003.
[41] Liaw, A., and Wiener, M. "Classification and regression by randomForest," R news Vol. 2, No. 3, 2002, pp. 18-22.
[42] Biau, G., and Scornet, E. "A random forest guided tour," Test Vol. 25, No. 2, 2016, pp. 197-227.
[43] Boyd, S., and Vandenberghe, L. Convex Optimization: Cambridge University Press, 2004.
[44] Boggs, P. T., and Tolle, J. W. "Sequential quadratic programming," Acta numerica Vol. 4, No. 1, 1995, pp. 1-51.
[45] Breiman, L. "Random forests," Machine learning Vol. 45, No. 1, 2001, pp. 5-32.
[46] Altman, N., and Krzywinski, M. "Ensemble methods: bagging and random forests, " Nat Methods Vol. 14, 2017, pp. 933–934.
[47] Kohavi, R. " A study of cross-validation and bootstrap for accuracy estimation and model selection," Proceedings of the 14th International Joint Conference on Artificial Intelligence Vol. 2, No.1, 1995, pp. 1137-1143.