Commuter networks and community detection: a method for planning sub regional areas

TL;DR

This paper proposes a network-based community detection method to identify cohesive sub-regions in Sardinia based on commuter flows, aiding regional planning and administrative boundary definition.

Contribution

It introduces a modularity maximization algorithm applied to commuter networks to delineate sub-regions with similar commuting patterns, informing administrative boundary decisions.

Findings

New provinces in Sardinia align with commuter-based communities.

Commuter network analysis reveals natural sub-regions.

Method supports policy decisions for regional planning.

Abstract

A major issue for policy makers and planners is the definition of the "ideal" regional partition, i.e. the delimitation of sub-regional domains showing a sufficient level of homogeneity with respect to some specific territorial features. In Sardinia, the second major island in the Mediterranean sea, politicians and analysts have been involved in a 50 year process of identification of the correct pattern for the province, an intermediate administrative body in between the Regional and the municipal administration. In this paper, we compare some intermediate body partitions of Sardinia with the patterns of the communities of workers and students, by applying grouping methodologies based on the characterization of Sardinian commuters' system as a complex weighted network. We adopt an algorithm based on the maximization of the weighted modularity of this network to detect productive basins composed by municipalities showing a certain degree of cohesiveness in terms of commuter flows. The results obtained lead to conclude that new provinces in Sardinia seem to have been designed -even unconsciously- as labour basins of municipalities with similar commuting behaviour.