Conference Agenda

Quantifying urban dynamics through building change is an important component to plan for sustainable cities, in particular regarding issues related to densification and accessibility. While methods based on remote sensing images and machine learning on raster data have been recently thriving, methods based on change detection in vector data can offer a high resolution description of changes enriched with several attributes, but yet remain underexplored in the literature. Indeed, change detection in geospatial vector data using matching algorithms still faces technical difficulties. This contribution investigates how to improve the quality of such approaches.

More precisely, we tackle different complementary issues: (i) the lack of ground-truth datasets, for which we build a collaboratory annotation web platform, on which users annotate simultaneously observed changes in aerial imagery and expected matching links in corresponding vector datasets; (ii) the various performance of existing algorithms depending on the urban context, which we handle by introducing a novel multi-modelling algorithm combining geometric matching of areas (performing better on links between single features) with multi-criteria matching (performing better on multiple links); (iii) the dependence of algorithm performance to hyperparameters, which we optimise on ground truth data obtained in the first step (F-score indicator) using the OpenMOLE platform and a NSGA2 genetic optimisation algorithm.

We illustrate the application of the approach in a comparative manner on two different European urban areas: Strasburg, France, and Dortmund, Germany. First results show a significantly better performance of the optimised multi-modelling algorithm, compared to optimised versions of single algorithms, and to versions with default parametrisations. Country differences in optimised parameters do not seem to be significant, suggesting a potential genericity of the approach. Future work will include the scaling of the approach to larger areas, and its test on more various types of urban form.

There is an increasing need to make cities more sustainable. One strategy is to improve the pedestrian and cycling accessibility to essential daily amenities. An urban planning concept used to measure the accessibility within cities is the concept of the “x-minute city”. Here, ‘x-minute’ refers to the time needed to access all essential amenities. This concept has been applied to a variety of cities to compare accessibility due to their particular spatial distribution of traffic infrastructure and amenities. However, when this analysis is conducted at the level of whole cities, it tends to over-generalise the results to the entire urban area. Yet, cities are not uniformly composed: they consist of a combination of multiple types of urban structure that have evolved over time, each with their own distinctive characteristics. The urban morphology, the connectivity of the street network and the spatial distribution of amenities vary greatly depending on these types of urban structures and on where they are located within cities. To date, the impact of urban structure types on pedestrian accessibility, using the measure of x-minuteness, has only been studied for a few specifically selected and very distinctive urban structure types. A comprehensive study researching accessibility at the city level, differentiated by types of urban structures, has yet to be conducted. The city of Berlin offers a dataset at the block level with information on the occurring urban structure types for the whole city. Using network analysis, we calculated the pedestrian accessibility to 16 amenities of daily essential needs (e.g. pharmacy or supermarket) across seven different residential urban structure types. Our results demonstrate that accessibility is highly dependent on the underlying urban structure type. For example, while the average x-minuteness for the city of Berlin as a whole is of 30.5 minutes, we can show that the mean x-minuteness for areas consisting of Wilhelminian-period block-edge development (historical type of urban structure dating from the end of the 19^th century) is of only 13.5 minutes. This differs significantly from the urban structure type Single-family homes with private greening which showcases an average of 36.5 minutes. Moreover, we find not only that the average accessibility varies between different urban structural types, but also that the spatial distribution of the amenities is clearly distinctive for each type of urban structure. This kind of analysis can specifically identify areas with insufficient accessibility by highlighting which amenities lack good access in each urban structure type. These insights can support urban planners and city councils by pinpointing where pedestrian accessibility can be improved to generate more sustainable, liveable and pedestrian-friendly cities.

Suburban territories continue to challenge sustainable urbanism. Their car-oriented layouts and fragmented public spaces contrast with contemporary demands for a more human-scaled, coherent, and dense urban fabric, as well as for models that minimize land artificialization. These conditions call for renewed suburban paradigms capable of generating suburban urbanities. When examined through the lens of street networks, however, these landscapes reveal latent structures and potential centralities embedded within their existing morphologies. This contribution develops the concept of a street logic of suburbs—an analytical and design framework linking multiscalar centralities to the morphological characteristics of suburban streets. Drawing on the DUT-funded Evolutive Meshed Compact City (EMC²) project, the research explores how suburban street systems can support more compact, connected, and walkable patterns of development. The study focuses on five European metropolitan areas—Vienna, Gothenburg, Lille–Roubaix–Tourcoing, the French Riviera, and Versilia—representing a range of suburban morphologies. A combination of multiscalar configurational analysis and street-based morphometrics is applied to identify and characterize main street networks across scales. Findings reveal that many suburban axes display strong configurational centrality but lack the morphological attributes that foster public life. Understanding this misalignment offers a foundation for rethinking suburban transformation, placing the street at the core of strategies for future, more urban, and sustainable suburban structures.

As European cities seek pathways toward sustainable urbanisation, the circular economy has emerged as a guiding principle for reducing resource consumption and environmental impacts. The transition toward a circular economy in cities demands spatially grounded approaches that can capture and respond to the complexities of the built environment. Among the various scales of analysis, the urban block sitting between the granular scale of buildings and the systemic view of entire urban regions, provides a spatially coherent unit that integrates land use, built form, and infrastructural systems. Using the urban block as the analytical entry point allows for targeted interventions and context-sensitive strategies that align with circular economy principles such as reuse, adaptability, and density optimisation. Anchoring circularity at this meso-scale facilitates integrated assessments of material stocks, socio-economic conditions, and environmental risks.

While increasing attention is being paid to circularity in the built environment, the literature remains fragmented across scales. Most existing analyses focus either on the building level where life cycle data is available or on macro-level indicators at city or regional scales. This leaves a conceptual and operational gap at the block level, where spatial, material, and social dimensions intersect. Existing archetype classifications are typically city-specific and focus on energy or microclimatic modelling, often neglecting key circularity factors such as reuse potential, embedded materials, and socio-cultural values. A need exists for transferable, data-driven typologies that can support comparative analysis and decision-making across diverse urban contexts.

To address this, the present study introduces a geospatial methodology for creating urban block archetypes aimed at supporting circular regeneration. The classification approach uses harmonized spatial datasets including cadastral data, building attributes, land use, and construction age to group blocks based on their morphology, functional mix, and circularity potential. Key indicators include density, land use diversity, compactness, accessibility, and exposure to environmental risks. Clustering techniques are applied across several European urban agglomerations to identify recurring spatial patterns and characterize representative block types.

These archetypes provide a scalable framework for further analysis, including scenario modelling, policy testing, and life cycle assessment. By capturing both commonalities and differences across cities, the typology allows for a more nuanced understanding of how local conditions shape the feasibility and impact of circular strategies. The archetypes also serve as building blocks for developing simplified digital models of urban form, supporting more efficient and participatory planning tools.

The study contributes to ongoing efforts to integrate circular economy principles into spatial planning by offering a methodological foundation for analysing and comparing urban blocks as drivers of transformation. It highlights the value of using open spatial data and interoperable tools to support urban sustainability and decarbonization goals. Ultimately, the framework strengthens the link between urban morphology and circularity, providing operational guidance for planning, design, and policy in support of resilient, inclusive, and resource-efficient urban futures.