Generative network models for approximating urban networks: the case of inter-city transportation networks in Southeast Asia

In recent years, the use of network perspectives has become increasingly popular in urban geography. This shift is not only visible in a range of theoretical, but also in matching methodological approaches where 'network thinking' is invoked to better understand the position of cities in urban networks. The purpose of this paper is to examine the driving forces underlying urban network formation through a geospatial simulation of observed networks. We begin with proposing a generative network model (GNM) combining spatial (topography) and network structures (topology), thus combining traditional spatial simulation models (e.g., gravity models) and topological simulation models (e.g., actor-oriented stochastic models). It is assumed that the probability of connections between cities emerges from competing forces. Stimulating factors are a measure of city size (e.g., population) and a topological rule favoring the formation of connections between cities sharing nearest neighbors (i.e., transitive effects). The hampering factors are physical distance between two cities as well as institutional distance (e.g., border effects). The model then is validated against empirical data on the transport network connecting 51 major cities in Southeast Asia. Our results show that (1) the generated networks closely approximate the observed ones in terms of average path length, clustering, modularity, efficiency and quadratic assignment procedure (QAP) correlation, and that (2) GNM performs best when topographical and topological factors are simultaneously present. Each factor contributes differently to network formation with transitive effects play the most importance role, showing the limitations of gravity-type models.