Forecasting Airport Surface Traffic Congestion based on Decision Tree

doi:10.23940/ijpe.20.05.p7.738746

Abstract

Abstract:

To improve the operational efficiency of airport surfaces, this paper studies the air traffic congestion prediction of airport surfaces, demonstrates the limitations of traffic congestion prediction, and proposes a prediction method for airport surface traffic congestion based on decision tree. Firstly, the definition and measurement methods of traffic congestion in airport surfaces are promoted. Then, the key factors affecting traffic congestion are extracted, and a prediction model of traffic congestion is established. Finally, we verify the validity of the model based on actual operation data from Atlanta. The results show that the accuracy of the prediction is 70%.

Key words: air transportation, traffic congestion, decision tree, C4.5 algorithm

Zhaoyue Zhang, An Zhang, Cong Sun, and Shanmei Li. Forecasting Airport Surface Traffic Congestion based on Decision Tree [J]. Int J Performability Eng, 2020, 16(5): 738-746.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

Figures/Tables 6

Table 1

Figure 1.

Table 2

Table 3

Figure 2.

Table 4

References 12.

1.	T. Xu, J. Ding, B. Gu,J. Wang, “Flight Delay Warning based on Incremental Array Support Vector Machines,” Acta Aeronautica ET Astronautica Sinica, Vol. 30, No. 7, pp. 1256-1262, 2009
2.	B. M. He, “Study on Short-Term Prediction Model of Air Traffic Flow,” Journal of Wuhan University of Technology (Transportation Science and Engineering), Vol. 12, No. 1, pp. 334-356, 2012
3.	S. M. Li, “Research on Identification and Prediction Methods of Air Traffic Congestion,” Tianjin University Press, 2013
4.	Volpe National Transportation Systems Center, “Enhanced Traffic Management System (ETMS) Functional Description,” U.S. Dept. of Transportation, Cambridge, MA, 2002
5.	G. B.Chatterji and B. Sridhar, “Measures for Air Traffic Controller Workload Prediction,” inProceedings of 1st AIAA Aircraft Technology, Integration and Operations Forum, pp. 104-125, 2001
6.	P. T. R.Wang, N. Tene, and L. Wojcik, “Relationship Between Airport Congestion and at-Gate Delay,” inProceedings of 21st Digital Avionics Systems Conference, pp. 67-88, 2002
7.	C. R. Wanke, L. Song, S. Zobell, D. Greenbaum,S. Mulgund, “Probabilistic Congestion Management,” 6th USA/Europe Seminar of Air Traffic Management R&D, pp. 27-30, 2005
8.	Z. Zhao, “Research on Airspace Capacity Assessment and Forecast,” Nanjing University of Aeronautics and Astronautics Press, pp. 34-41, 2015
9.	X. N. Dong, “Sector Capacity Evaluation and Complexity Analysis,” Nanjing University of Aeronautics and Astronautics Press, pp. 65-80, 2017
10.	Y. X. Sun, C. F. Shao, D. Zhao,S. Ou, “Traffic Accident Severity Prediction Model based on C5.0 Decision Tree,” Journal of Chang'an University (Natural Science Edition), Vol. 34, No. 5, pp. 123-132, 2014
11.	X. W. Wang, C. Q. Yuan,M. Huang, “A Motion Prediction Mechanism based on Fuzzy Decision Tree,” Computer Science, Vol. 32, No. 9, pp. 1176-179, 2005
12.	R. Li, Y. Liu, J. H. Li, X. P. Gu, D. X. Niu,Y. Q. Liu, “Study on Daily Characteristic Load Prediction based on Improved Decision Tree Algorithm,” Proceedings of the CSEE, Vol. 25, No. 23, pp. 36-41, 2005

Attribute name	Attribute value
Weather conditions (rain, fog, snow, hail…)	0-No impact, 1-Impact
Time slot	0-Peak, 1-Low peak
Holiday and vacations	0-Y, 1-N
Traffic volume	0-Large, 1-Small
Airport infrastructure integrity	0-Complete, 1-Incomplete
Ability of controllers	0-Outstanding, 1-Average, 2-Poor
Aircraft size	0-Large, 1-Small
Air force restrictions	0-Yes, 1-No
Traffic volume in the previous period	0-Large, 1-Small
Working day	0-Yes, 1-No

Sample number	Weather	Time slot	Holiday or not	Traffic	Working day or not	Traffic volume in the previous period		Crowding state
1	no influence	peak	yes	small	yes	small	0.10	smooth
2	no influence	peak	yes	large	yes	small	0.17	moderate congestion
3	no influence	peak	yes	small	yes	large	0.10	smooth
4	influential	peak	yes	large	no	small	0.15	moderate congestion
5	influential	low peak	no	large	yes	small	0.16	moderate congestion

Smooth	Slight congestion	Moderate congestion	Severe congestion
no influence	232	influential	57
peak	154	low peak	154
holiday and vacations	132	non-holidays	176
traffic volume	212	small traffic volume	96
working day	220	non-working days	88
large traffic volume in the previous period	211	small traffic volume in the previous period	97

Sample number	Time	Time slot	Working day or not	Weather	Traffic	Holiday or not	Traffic volume in the previous period	Crowding state	Predictive state
1	2014/12/21	peak	no	influential	large	no	small	severe congestion	smooth
2	2014/12/21	low peak	yes	no influence	small	no	large	smooth	smooth
3	2014/12/22	peak	yes	no influence	large	no	large	smooth	smooth
4	2014/12/23	peak	yes	influential	small	no	large	slight congestion	slight congestion
5	2014/12/23	low peak	yes	influential	small	no	small	smooth	smooth
6	2014/12/24	low peak	yes	influential	small	yes	large	smooth	smooth
7	2014/12/25	peak	yes	no influence	large	yes	large	slight congestion	smooth
8	2014/12/25	low peak	yes	no influence	large	yes	small	smooth	smooth
9	2014/12/26	low peak	yes	no influence	large	no	large	smooth	smooth
10	2014/12/27	peak	no	no influence	small	no	large	smooth	moderate congestion

[1]	Minghua Jia, Xiaodong Wang, Yue Xu, Zhanqi Cui, and Ruilin Xie. Testing Machine Learning Classifiers based on Compositional Metamorphic Relations [J]. Int J Performability Eng, 2020, 16(1): 67-77.
[2]	Sizhe Huang, Huosheng Xu, and Xuezhi Xia. Imbalanced Remote Sensing Ship Image Classification [J]. Int J Performability Eng, 2019, 15(6): 1709-1715.
[3]	Hongbin Wang, Lei Hu, Xiaodong Xie, Lianke Zhou, and Huafeng Li. Incremental Integration Algorithm based on Incremental RLID3 [J]. Int J Performability Eng, 2019, 15(1): 252-260.
[4]	Xie Wei, Xu Huoxi, Peng Liping, and Lan Zhigao. Short-Range Wireless Network Communication and Application based on Decision Tree Algorithm [J]. Int J Performability Eng, 2018, 14(11): 2561-2573.