Emergence of hierarchy in cost driven growth of spatial networks

TL;DR

This paper introduces a generic growth model for spatial networks driven by cost-benefit analysis, revealing how hierarchical structures naturally emerge and are optimized for efficiency, with implications for real-world networks like railways.

Contribution

It presents a unified model linking local cost-benefit decisions to global hierarchical structures in spatial networks, bridging a gap in understanding their evolution.

Findings

Hierarchical spatial organization emerges from local cost-benefit considerations.

Optimal network structures balance detour minimization and diversity of link lengths.

Real-world railway networks operate in the intermediate regime, indicating evolutionary advantages.

Abstract

One of the most important features of spatial networks such as transportation networks, power grids, Internet, neural networks, is the existence of a cost associated with the length of links. Such a cost has a profound influence on the global structure of these networks which usually display a hierarchical spatial organization. The link between local constraints and large-scale structure is however not elucidated and we introduce here a generic model for the growth of spatial networks based on the general concept of cost benefit analysis. This model depends essentially on one single scale and produces a family of networks which range from the star-graph to the minimum spanning tree and which are characterised by a continuously varying exponent. We show that spatial hierarchy emerges naturally, with structures composed of various hubs controlling geographically separated service areas, and appears as a large-scale consequence of local cost-benefit considerations. Our model thus provides the first building blocks for a better understanding of the evolution of spatial networks and their properties. We also find that, surprisingly, the average detour is minimal in the intermediate regime, as a result of a large diversity in link lengths. Finally, we estimate the important parameters for various world railway networks and find that --remarkably-- they all fall in this intermediate regime, suggesting that spatial hierarchy is a crucial feature for these systems and probably possesses an important evolutionary advantage.