Before any design decision, an architect must read the place: understanding the site's physical constraints, urban context, regulatory framework, and latent opportunities. This territorial intelligence — traditionally gathered over days of research — can now be automated. AI systems query public geospatial APIs, process terrain and urban data, and generate interpretive site analysis reports in minutes.
What AI-Driven Territory Analysis Covers
A complete AI-driven site analysis pipeline integrates several data streams: terrain and topography from SRTM or regional elevation models; urban fabric from OpenStreetMap, Overture Maps, or national cadastral APIs; sun exposure and solar radiation from climate data services; regulatory constraints from building permit databases and zoning APIs; and surrounding building context from 3D city models (CityGML, 3D BAG, or equivalent). The AI layer processes all of these, identifies patterns and constraints, and generates a narrative interpretive report — what the place affords, what it prohibits, and what opportunities the analysis reveals.
From Data to 3D Context Model
The pipeline produces not just a report but geometry: terrain surfaces, building volumes, and urban context in a format directly importable into the architect's design environment. Rhino, Grasshopper, and Blender all have Python APIs that accept geometry from external pipelines. The 3D context model becomes the base on which design begins — already calibrated to the actual place, without manual reconstruction from aerial photographs and cadastral maps.
For urban projects, this step alone eliminates 2–3 days of preparatory work per site. For residential projects, the terrain model and solar analysis that previously required surveying are generated automatically.
Regulatory Intelligence
Planning regulations — floor area ratios, height limits, setback requirements, use restrictions — are typically read manually from PDF documents and cross-referenced with site measurements. AI makes this systematic: a language model reads the applicable regulations, extracts the constraints as structured parameters, and validates the design geometry against them automatically. This regulatory AI layer catches compliance issues before permit submission, not after.
The same approach applies to environmental regulations, heritage protection zones, and infrastructure easements — constraints that architects must know but that are distributed across dozens of public documents.
MIAW Module F1 — Territory with AI
AI-driven site analysis is the core subject of MIAW Module F1. Students build an automated territorial analysis pipeline applied to their own real project site, producing an AI-generated interpretive report and a 3D context model integrated in their design environment. The pipeline becomes a reusable asset — applicable to any future site with minimal reconfiguration.