Key flight segment identification based on flight segment correlation

Existing analyses of air traffic networks are often based on the static structural characteristics of flight routes, considering waypoints as nodes and flight segments as edges, or airports as nodes with routes between them as edges. These analyses rarely take into account the mutual influence between nodes and do not fully reflect the directional and purposeful flow characteristics of flights. This paper focuses on flight segments, which better represent flight information, and selects indicators that reflect the dynamic characteristics of flight segments to represent their attributes. The maximum mutual information coefficient is used for the correlation analysis of flight segments, identifying the relationships of influence between them. Flight segments are treated as nodes, and their correlations as edges, to construct a complex network of segment correlations. The importance of flight segments is then evaluated using feature vector centrality and the PageRank algorithm, which consider the dynamic characteristics of the network. Finally, a simulation experiment is conducted based on the actual operation data of 278 flight segments in the Central South region of China's civil aviation network from January 1, 2016, to May 31, 2016. The correlations between flight segments are calculated and a network of segment correlations is constructed. By evaluating the importance of flight segments in this region, it is possible to identify those that contribute significantly to the stability and connectivity of the route network. Moreover, this analytical method provides a new perspective on the study of flight delay propagation, which is of great significance for improving the efficiency of the route network and for route planning.