Post by TWIN2EXPAND

New Publication! A persistent assumption in spatial network analysis is that hybrid street models (where manual refinement complements automated extraction) constitute the analytical benchmark. Our recent Q1 paper in Environment and Planning B revisits this assumption by examining the consequences of two fundamentally different modelling logics: “Automated versus hybrid street network modelling for centrality and accessibility analysis”. Rather than treating network construction as a neutral preprocessing step, the paper frames it as an epistemological choice. Specifically, we contrast: – the geometry-preserving segmentation logic of the automated workflow (road/street centreline manipulation) – the continuity-merging rules applied in the hybrid model (Space Syntax) Across multiple urban contexts, including Nicosia and London, the analysis asks: To what extent does the configurational logic of cities persist under these competing representations? The results point to a conditional stability: – At the level of global centrality, the geometry-preserving segmentation reproduces configurational structure with high fidelity – At finer scales, divergence becomes systematic, emerging precisely where the continuity-merging rules suppress or reinterpret marginal spatial elements – Accessibility becomes inseparable from the underlying modelling logic, rather than a property of the urban system alone This suggests that what is often treated as a technical distinction is, in fact, constitutive of the analytical outcome. If segmentation and merging logics generate different yet internally consistent spatial structures, then the question is no longer which model is “correct,” but under what conditions each becomes analytically meaningful. The broader implication is a shift from seeking a single “best” representation toward recognising multiple valid modelling regimes, each encoding a different reading of urban space. It would be interesting to hear how others are negotiating this tension between scalability, reproducibility, and theoretical alignment in large-scale spatial analysis. https://lnkd.in/d_CFM_SB #UrbanAnalytics #SpaceSyntax #GIS #ComputationalUrbanism #UrbanMorphology #EvidenceBasedDesign #SmartCities #OpenScience Walid S Abdeldayem Ilaria Geddes Ahmed Hazem Eldesoky Ioanna Stavroulaki Gareth Simons Meta Berghauser Pont Nadia Charalambous