Research · Computational urban science
Complex Urban Systems
An umbrella research program on urban inference when no single dataset gives a complete account of a city.
TypeResearch program
StatusOngoing
DomainUrban science
The problem
Urban decisions often depend on behavior that no agency observes directly. This program studies how mobility, demographic, transport, and infrastructure records can be combined while preserving the gaps and disagreements between them.
Lines of work
- Probabilistic inference for urban processes that cannot be observed directly.
- Integration of mobility, demographic, transport, and infrastructure data.
- Large-scale models that preserve interpretable mechanisms and uncertainty.
- Machine-learning methods whose outputs remain tied to their evidence boundaries.
Modeling rule
Every dataset is treated as a partial view. A model must state what each source can support, where the sources disagree, and which conclusions remain unresolved.