Complementarity of generative models for road networks

TL;DR

This paper systematically benchmarks various generative models of road network growth, revealing high complementarity among them and emphasizing the importance of using multiple models for urban system analysis.

Contribution

It introduces a comprehensive benchmark integrating diverse paradigms of road network models, highlighting their complementarity and supporting pluralistic urban modeling.

Findings

High complementarity between different models

Supports the necessity of multiple models for urban analysis

Confirms the diversity of feasible network measures

Abstract

Understanding the dynamics of road networks has theoretical implications for urban science and practical applications for sustainable long-term planning. Various generative models to explain road network growth have been introduced in the literature. We propose in this paper a systematic benchmark of such models integrating different paradigms (spatial interactions, cost-benefit compromises, biological network growth), focusing on the feasible space of generated network measures. We find a quantitatively high complementarity between the different models. This confirms the necessity of a plurality of urban models, in line with integrative approaches to urban systems.