Automatically generated port hinterlands

Competition between ports depends on inland freight distribution and the spatial structure of the hinterland. With this, ports and port regions increasingly compete to serve distant hinterlands. In a European context, many researchers refer to the agglomeration of economic activity in the Rhine-Ruhr area and the ‘blue banana' to explain the concentration of port activity in the Hamburg-Le Havre port range. Besides this, the incorporation of new member states in the European market has changed the structure of port hinterlands. In this paper we attempt to reveal the spatial structure of the hinterland of the Hamburg-Le Havre ports using automated zoning techniques. These techniques aggregate geographical areas in homogeneous clusters using spatial as well as content-related constraints. We use both economic characteristics of hinterland regions and variables which express the link between these regions and ports to create a new map of the port hinterland. Besides an improved insight in the spatial structure of the hinterland, this analysis delivers a set of areas which can be used in economic models. Indeed, creating an ‘optimal' zoning is one of the strategies researchers employ to handle observational units with often arbitrary boundaries.