Urban Forms Across Continents: A Data-Driven Comparison of Lausanne and Philadelphia

Abstract: Understanding urban form is crucial for sustainable urban planning and enhancing quality of life. This study presents a data-driven framework to systematically identify and compare urban typologies across geographically and culturally distinct cities. Using open-source geospatial data from OpenStreetMap, we extracted multidimensional features related to topography, multimodality, green spaces, and points of interest for the cities of Lausanne, Switzerland, and Philadelphia, USA. A grid-based approach was used to divide each city into Basic Spatial Units (BSU), and Gaussian Mixture Models (GMM) were applied to cluster BSUs based on their urban characteristics. The results reveal coherent and interpretable urban typologies within each city, with some cluster types emerging across both cities despite their differences in scale, density, and cultural context. Comparative analysis showed that adapting the grid size to each city's morphology improves the detection of shared typologies. Simplified clustering based solely on network degree centrality further demonstrated that meaningful structural patterns can be captured even with minimal feature sets. Our findings suggest the presence of functionally convergent urban forms across continents and highlight the importance of spatial scale in cross-city comparisons. The framework offers a scalable and transferable approach for urban analysis, providing valuable insights for planners and policymakers aiming to enhance walkability, accessibility, and well-being. Limitations related to data completeness and feature selection are discussed, and directions for future work -- including the integration of additional data sources and human-centered validation -- are proposed.

Urban Forms Across Continents: A Data-Driven Comparison of Lausanne and Philadelphia

This paper investigates the urban forms of two cities—Lausanne, Switzerland, and Philadelphia, USA—through a data-driven analysis using clustering approaches. The objective is to identify and compare urban typologies across geographically distinct and culturally divergent cities, utilizing multidimensional geospatial features derived from OpenStreetMap (OSM). This research seeks to provide insights that can be instrumental for urban planners and policymakers aiming to enhance sustainable urban environments.

Methodology

The authors employ a grid-based approach, segmenting each city into Basic Spatial Units (BSUs). Geospatial features related to topography, multimodality, green spaces, and points of interest are extracted from OSM and used as input data for clustering techniques. Gaussian Mixture Models (GMM) are applied to these BSUs, allowing for the identification of coherent clusters that represent distinct urban typologies.

The study emphasizes the importance of adapting grid size to the spatial characteristics of each city, as demonstrated by the grid sizes of 1500 meters for Philadelphia and 450 meters for Lausanne. This adaptation facilitates an accurate comparison by accounting for differences in urban scale and density. Moreover, the research shows that even a simplified clustering, using degree centrality alone, can reveal meaningful structural patterns, suggesting that minimal feature sets still have descriptive power for urban analysis.

Results and Analysis

The clustering results reveal the presence of unique and shared urban typologies within and across cities. In Lausanne, clusters delineate areas such as the hypercenter, characterized by high densities of pedestrian-friendly streets and cultural amenities, and areas around major transportation hubs, which show significant concentrations of natural elements and commercial activity. Similarly, Philadelphia clusters highlight the urban core's high connectivity and multimodal integration, distinctive from the residential and peripheral areas.

Through comparative analysis, shared urban forms emerge between the two cities, despite their geographical and cultural differences. This suggests the existence of potential universal principles in urban spatial organization, supporting the hypothesis that certain urban typologies can be transferable across different contexts. Additionally, qualitative validations via satellite images confirm the clustering’s ability to capture spatial homologous regions by matching structural and land-use characteristics despite their geographic separation.

Implications and Future Directions

The findings underscore the importance of considering spatial scale when conducting cross-city analysis, as appropriate resolution is crucial for capturing relevant urban morphologies. The research also highlights the potential for leveraging open-source geospatial data and clustering techniques in urban planning practice to identify areas requiring targeted interventions.

Future research should focus on integrating additional data sources to cover unrepresented qualitative factors, such as safety and sidewalk quality, which significantly affect walkability. Expanding the framework to include more cities globally, particularly in rapidly urbanizing regions, could further validate the generalizability of the findings. Additionally, human-centered validation approaches, such as surveys and participatory mapping, would enhance the understanding of how clustering outcomes correlate with residents' perceptions of walkability and urban quality.

In summary, the paper presents a robust framework for analyzing urban forms that can aid in the development of globally sustainable urban design principles. By demonstrating the systematic classification of urban typologies across continents, it offers valuable insights for enhancing urban environments aligned with the needs of diverse communities.