Urban planners in Belo Horizonte, Brazil, have tested a computational framework that treats the master plan less like a fixed legal document and more like a living analytical model. In a study published in the journal City, Territory and Architecture, researcher Marina Borges presents Adaptive Master Plans, or AMPs, as a way to connect parametric modelling, geospatial analysis and urban performance indicators so that alternative development scenarios can be evaluated before they are written into law. The work responds to a persistent problem in statutory planning: diagnoses are collected at one moment, legislative procedures can take years, and by the time a master plan is approved the data behind it may already be outdated.
The study is explicitly exploratory rather than a validation of a finished planning system. Three sequential case studies in Belo Horizonte illustrate distinct analytical dimensions of the approach. The first simulates the microclimatic consequences of densification in areas designated for growth. The second examines walkability and access to essential services in peripheral neighbourhoods, where steep terrain and fragmented street networks distort conventional benchmarks. The third evaluates regenerative strategies, from rain gardens to ecological corridors, through simulations of ecosystem services and environmental performance. Together, the cases show how computational planning environments can support dynamic territorial diagnostics and predictive scenario evaluation.
The methodological workflow proceeds in three stages. Geospatial datasets covering land use and land cover, built form and building height, infrastructure networks, public facilities and socio-demographic indicators are organized into a structured spatial database using GIS tools. Parametric urban models are then built in Rhinoceros 3D and Grasshopper, together with the Urbano plugin, allowing planners to manipulate variables such as density, building height, land-use distribution and open-space configuration. Finally, environmental simulations run through Ladybug Tools, an open-source suite, measuring indicators including air temperature, solar radiation, relative humidity and the Sky View Factor. The same workflow can also incorporate accessibility modelling, walkability assessments and ecosystem service potentials.
The first case study targets the Regional Centralities designated in the Belo Horizonte Master Plan, Law No. 11.181/2019, where density and mixed-use development may be intensified to promote compact growth and reduce sprawl. By manipulating building height, floor area ratio and land-use distribution, the research generated alternative densification scenarios and simulated their microclimatic effects. The simulations indicated that increased building density may produce measurable microclimatic impacts, including localized increases in air temperature and reductions in sky visibility and natural ventilation potential. The analysis suggests that mitigating such effects may require reconsidering not only floor area ratios but also permeable surface requirements, building setbacks and open-space configurations.
The second case study shifts from environmental performance to spatial equity. Using routing algorithms and geospatial datasets, the research simulated access distances to schools, parks and health centers across different areas of the city. The analysis revealed significant territorial disparities: residents in peripheral neighbourhoods, many characterized by lower average incomes and complex topographical conditions, often face substantially longer travel distances to essential facilities. In areas with steep slopes and fragmented street networks, pedestrian routes become longer and more physically demanding. The results quantitatively demonstrate that conventional walkability benchmarks, including the widely cited 15-minute city concept, are often based on relatively flat urban morphologies and highly connected street networks, and may not adequately capture the spatial realities of peripheral territories with irregular topography and discontinuous infrastructure.
The same case also tested planning responses. Small-scale connectivity interventions, including pedestrian passages, extensions of sidewalk networks and alternative spatial distributions of public facilities, were simulated across different scenarios. In some scenarios, these interventions indicated the possibility of reducing average travel times by up to 20 percent. The finding illustrates how computational analysis can link urban morphology, accessibility patterns and spatial equity, enabling planners to identify context-sensitive solutions rather than applying universal benchmarks. The study also cautions that the ways spatial data are produced and interpreted influence which territorial inequalities become visible and actionable within planning processes, reinforcing the importance of informational justice in data-driven frameworks.
The third case study explored the integration of nature-based solutions into regenerative strategies for consolidated urban areas. An initial set of simulations tested short- and medium-term mitigation measures within existing urban fabrics: rain gardens in public spaces, green roofs adapted onto existing buildings and increased permeable surfaces within developed lots. The results indicated measurable improvements in environmental performance, particularly in stormwater infiltration capacity and reductions in peak surface runoff. Incorporating rainwater harvesting systems and distributed green infrastructure also demonstrated the potential to reduce potable water demand by approximately 15 percent in some scenarios. Yet the simulations showed that such measures, when implemented only at the scale of individual lots or small interventions, can attenuate environmental impacts without substantially transforming environmental conditions.
A second set of simulations therefore explored more structural scenarios at a broader territorial scale, including the strategic acquisition and conversion of selected urban parcels into larger green infrastructure areas functioning as ecological corridors, stormwater retention areas and multifunctional green spaces. These interventions produced significantly greater environmental improvements, with higher infiltration capacity, reduced runoff peaks and enhanced ecosystem service provision. The comparison allowed the study to quantify the trade-offs between incremental mitigation and structural regeneration, illustrating one of the framework’s central claims: computational planning environments can identify trade-offs between spatial configuration, environmental performance and accessibility that remain largely invisible within conventional planning instruments.
Beyond the technical results, the study situates Adaptive Master Plans within broader debates on digital governance. Digital urban infrastructures frequently operate through algorithmic logics that privilege institutional expertise and specialized technical knowledge, potentially marginalizing local and situated knowledge. The research argues that Adaptive Master Plans are not neutral tools but infrastructures that shape how urban inequalities become visible and actionable, and it points to collaborative practices such as participatory mapping and citizen-generated data as complements to institutional datasets. It also proposes pathways for integrating computational analytical environments into statutory systems through performance-based planning criteria, conditional zoning mechanisms and adaptive regulatory thresholds linked to environmental indicators, an approach the author describes as regulated adaptability.
Several limitations frame the findings. The analytical environments presented are structured applications of the proposed framework rather than fully operational planning systems, the participatory dimension was not empirically implemented, and the simulations relied on available datasets and simplified modelling assumptions that may not capture the full complexity of urban dynamics. The case studies should be read as illustrative demonstrations of analytical potential rather than empirical validation of a complete planning model. Future research, the study concludes, should investigate how such environments could be integrated with statutory planning frameworks and institutional decision-making while ensuring transparency, accountability and equitable access to urban data infrastructures.
Subject of Research: Computational urban analysis and adaptive master planning for urban performance and regenerative planning in Belo Horizonte
Article Title: Adaptive master plans: computational urban analysis for urban performance and regenerative planning
Article References: Borges, M. (2026). Adaptive master plans: computational urban analysis for urban performance and regenerative planning. City, Territory and Architecture, 13(1), Article 21. https://doi.org/10.1186/s40410-026-00314-x
Image Credits: AI Generated
DOI: 10.1186/s40410-026-00314-x
Keywords: Adaptive Master Plans, computational urbanism, parametric modelling, geospatial analysis, urban densification, microclimate, walkability, spatial equity, ecosystem services, nature-based solutions, Belo Horizonte, regenerative planning
News Source: Denise Maddox. (October 30, 2026). Computational Urban Analysis Reshapes Master Planning in Belo Horizonte. Scienmag.
