Urban Sprawl and Routing: A Comparative Study on 156 European Cities

Abstract

• We propose a novel methodology that integrates operations research into urban studies. • We find a significant correlation between urban sprawl and routing times. • Addressing urban sprawl can reduce transportation’s economic, social, and environmental costs. • Shannon’s entropy emerges as the best predictor of the route length. • We gain insights about the geographical distribution of sprawl in Europe. To address the growing challenges urban sprawl poses, it is essential to understand its influence on urban transportation, a primary source of economic, social, and environmental impact. This study fills this gap by quantifying the consequences of sprawl on transportation efficiency, proposing an interdisciplinary methodology that integrates knowledge from operations research. Specifically, adopting a broad European perspective, we investigate how urban sprawl correlates with travel distances and optimal routes in 156 spatially heterogeneous cities across 28 European countries. We discover a significant correlation between five sprawl indicators (Land usage, Gini coefficient, Shannon entropy, Moran I index, and Bribiesca index) and both travel distances and routes by car and bicycle: transportation is inherently less efficient in cities with higher levels of sprawl. Among the considered indicators, Shannon entropy emerges as the best predictor of route efficiency. We offer insights into the geography of sprawl in Europe, finding that many Spanish cities stand out for their compactness and route efficiency, while hotspots of sprawl are present in many Western and Central European countries. Our results underline the underestimated importance of addressing urban sprawl to reduce transportation’s economic, social, and environmental costs and encourage policymakers and urban planners to prioritize compact city development to foster sustainable urban growth.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.