A new methodological approach for retail fabrics in the city

Résumé En

Urban retail distribution within the urban space contributes to the vitality and economic prosperity of streets and neighborhoods. Moreover, different retail patterns represent the specific outcome of the surrounding socioeconomic fabric. Several models of retail activity distribution have been proposed in the scientific literature of economic geography in order to study the spatial organization of these retail patterns. Despite the large debate about the validity of these models, few empirical based studied have been proposed. Micro retail location data has been the subject of studies from different disciplines; statistic point pattern analysis methodologies have recently been presented as a way to detect spatial concentrations of a point pattern. Yet, concentration is not the only significant aspect defining retail fabric. Form and function represent the two main characteristics identified from the economic geography literature participating simultaneously to the definition of retail districts. In this paper a new bottom-up data-based methodology is proposed in order to assess retail distribution starting from a store point pattern representation on a street network. Theory-based indicators are implemented with geoprocessing approaches, allowing the recognition of morpho-functional patterns. Bayesian Classification methods are then applied to these spatial patterns. Retail fabrics and their spatial organization are identified and described. This methodology is applied in this paper to the real case study of the French Riviera metropolitan region.