| Abstract: | We present a methodology to analyse high resolution population and transport data in order to assess cross-border connectivity within the European Union. Transport infrastructure can strongly influence cross-border interactions as well as regional, urban or local development. The analysis is carried out using a policy perspective, with network efficiency as the main indicator of accessibility. The aim is to allow the quantification of the quality of cross-border road connections and the identification of areas where infrastructure improvements can lead to higher benefits. We propose a machine learning approach that combines cell level route assignment and k-means clustering at a fine ?1 square km- population grid. The outputs cover all internal EU land borders and consist of sets of spatial clusters that meet user-defined policy criteria. The results can be used as input for investment decisions and can be easily combined with other policy support tools for tailored multi-criteria analysis. |
