Dynamic Estimates of Sector Capacity Using the Combined SOM and K-means Method

DOI

10.2991/icmii-15.2015.101How to use a DOI?

Keywords

air transportation, sector capacity, traffic flow, K-means

Abstract

It is well known that the sector capacity is particularly important for the air traffic flow management. In order to improve the accuracy in dynamic estimates of the sector capacity, a combined SOM and K-means method of clustering is proposed for the estimates in this paper. Firstly, the traffic flow data in term of the number of aircrafts within a time interval and a specific sector are acquired by using the outputs of the surveillance radar. The data are refreshed by sliding a window. The traffic flow data are then clustered by using the combined SOM and K-means method. The clustered results are compared with the results obtained by using the pure K-means method. The clustered results with high accuracy are suggested to be used. The sector capacity for each pattern is hence obtained by calculating the maximum flow for each sector. In the similar way, the late-coming data are matched with the patterns. The dynamic estimates of the sector capacity are hence completed. Using the real radar data we have performed the dynamic estimates of the sector capacity by following the above processes. The results show that the combined SOM and K-means method is efficient in the dynamic estimates of the sector capacity and may be useful for air traffic management.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Shuxuan Xiong
AU  - Yifei Zhao
PY  - 2015/10
DA  - 2015/10
TI  - Dynamic Estimates of Sector Capacity Using the Combined SOM and K-means Method
BT  - Proceedings of the 3rd International Conference on Mechatronics and Industrial Informatics
PB  - Atlantis Press
SP  - 592
EP  - 599
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmii-15.2015.101
DO  - 10.2991/icmii-15.2015.101
ID  - Xiong2015/10
ER  -