Distance as a variable of configurational models for understanding the distribution of urban economic activities

Abstract

Different configurational urban models can generate similar outcomes by changing the distance parameter used in their processing. The use of distance as adjustable variable is an alternative way of regulating configurational models for understanding urban phenomena. The distribution of urban economic activities is linearly analyzed by comparing their locations with the output of a configurational spatial model for urban networks. To test the proposed hypothesis, the economic activities of three cities were spatialized and urban configurational models were applied on the spatial network of streets segments of these cities, varying the models’ processing radius (i.e., the distance parameter). The correlation between values of centrality resulting from each model processing and the distribution of economic activities in the cities were tested with the Pearson's correlation coefficient (r). Similar results were obtained by applying the same configurational urban models for a set of different cities, changing only the distance parameter. We concluded that it seems to be more important to set the distance parameter than to figure out which configurational model can describe how the phenomenon of distribution of economic activity takes place. That is, depending on each case, there would be distances that, associated with certain models, would be more or less correlated with the way economic activities are distributed throughout the urban fabric. Thus, configurational models can be adjusted by an optimized distance, increasing the correlation with the urban phenomenon studied.

Keywords:
Centrality; Distance; Urban spatial networks; Economic activities; Pearson’s correlation coefficient