Evolution-based modelling of complex airport networks

Abstract

This paper presents a time-series model of the United States Airport Network as a directed, weighted network, with the weight representing the total number of passengers flying from an origin to a destination airport, in a two month time period. Six independent networks are built for a given year, in order to capture the seasonal variation of passengers. To explore the evolution of the network over the past two decades, three specific years are investigated: 1990, 2000, and 2010. The results highlight the growth of the network in terms of airports and connections, and suggest a scale-free, small-world topology. In addition, the ranked passenger distribution appears to follow a logarithmic trend, implying high heterogeneity in passengers on different connections.