Restricted Airspace Unit Identification Using Density-Based Spatial Clustering of Applications with Noise

Yong Tian,Bojia Ye,Dawei Xing,Lili Wan,Minhao Yang

doi:10.3390/su11215962

Abstract

This paper first calculates the departure delay and arrival delay of each flight by mining historical flight data. Then, a new method based on density clustering for identification and visualization of restricted airspace units that considers this activity is proposed. The main objective is to identify the restricted airspace units by calculating the average delay time according to the accumulative delay time of airspace units and the accumulative delay flight. Therefore, the density-based spatial clustering of applications with noise (DBSCAN) clustering method is utilized to match the latitude and longitude coordinates of each spatial domain unit with its delay time to construct a feature matrix, and then clustering analysis is conducted according to the time period. The method aims at identifying the most severe restricted units in each period. The reliability and applicability of the proposed method are validated through a real case study with flight information from Beijing Capital International Airport, Hongqiao International Airport, and Baiyun International Airport during a typical day. The investigation shows that the DBSCAN clustering method can identify the restricted spatial units intuitively on the six flight paths between Beijing Capital International Airport, Hongqiao International Airport, and Baiyun International Airport.

Highlights

With the continuous growth of China’s air traffic volume, the problem of flight delays has become a major bottleneck restricting the rapid development of civil aviation
Choi et al used data mining and supervised machine learning algorithms to predict airline flight delays caused by bad weather conditions [15]
Identification of the restricted airspace unit and the start and end times of the restriction can help to solve the randomness of events caused by bad weather, air traffic control, and so on

Summary

Introduction

With the continuous growth of China’s air traffic volume, the problem of flight delays has become a major bottleneck restricting the rapid development of civil aviation. There are many factors that cause flight delays in actual operation, such as airline operation management, bad weather, military activities and air traffic control, and so on These factors will cause airspace units to be restricted to varying degrees, resulting in flight delays. Choi et al used data mining and supervised machine learning algorithms (random forest, decision tree, etc.) to predict airline flight delays caused by bad weather conditions [15]. Identification of the restricted airspace unit and the start and end times of the restriction can help to solve the randomness of events caused by bad weather, air traffic control, and so on.

Identification Method

Construction of the Identification Model

Conclusion