A centrality measure for urban networks based on the eigenvector centrality concept

A massive amount of information as geo-referenced data is now emerging from the digitization of contemporary cities. Urban streets networks are characterized by a fairly uniform degree distribution and a low degree range. Therefore, the analysis of the graph constructed from the topology of the urban layout does not provide significant information when studying topology-based centrality. On the other hand, we have collected geo-located data about the use of various buildings and facilities within the city. This does provide a rich source of information about the importance of various areas. Despite this, we still need to consider the influence of topology, as this determines the interaction between different areas. In this paper, we propose a new model of centrality for urban networks based on the concept of Eigenvector Centrality for urban street networks which incorporates information from both topology and data residing on the nodes. So, the centrality proposed is able to measure the influence of two factors, the topology of the network and the geo-referenced data extracted from the network and associated to the nodes. We detail how to compute the centrality measure and provide the rational behind it. Some numerical examples with small networks are performed to analyse the characteristics of the model. Finally, a detailed example of a real urban street network is discussed, taking a real set of data obtained from a fieldwork, regarding the commercial activity developed in the city