Urban Simulation Model
Also known as: Land-Use Microsimulation, Urban Growth Simulation, Agent-Based Urban Model, Integrated Land-Use Transport Simulation
Urban simulation models reproduce the dynamics of urban growth and land-use change by simulating, over time, the decisions of agents — households, firms, developers — or the transitions of cells on a grid. They span agent-based models, cellular automata such as SLEUTH, and microsimulation platforms such as Paul Waddell's UrbanSim, which represents individual households and jobs choosing locations through discrete-choice models linked to a transport network. Rather than predicting a single equilibrium, these models let many local rules and choices interact and feed back through prices and accessibility, generating emergent patterns of sprawl, densification, and redevelopment under alternative policies.
Key highlights
- Represents heterogeneous agents and local interactions, producing emergent patterns top-down models cannot.
- Couples land use and transport through explicit feedback so accessibility and development co-evolve.
- Supports scenario and policy testing — zoning, pricing, infrastructure — by re-running under new rules.
- Disaggregate outputs allow analysis of distributional and equity effects, not just totals.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use an urban simulation model when you need to explore how a city might evolve under alternative policies, infrastructure investments, or external pressures, and when bottom-up interaction, feedback, and heterogeneity matter more than a closed-form forecast. Microsimulation and agent-based models suit disaggregate questions about who locates where and equity impacts; cellular automata suit questions about the spatial pattern and rate of physical urban expansion. They are less appropriate when data are too sparse to estimate behaviour or calibrate transitions, when a quick aggregate forecast suffices, or when the model's complexity would outstrip the analyst's ability to validate it.
Strengths & limitations
- Represents heterogeneous agents and local interactions, producing emergent patterns top-down models cannot.
- Couples land use and transport through explicit feedback so accessibility and development co-evolve.
- Supports scenario and policy testing — zoning, pricing, infrastructure — by re-running under new rules.
- Disaggregate outputs allow analysis of distributional and equity effects, not just totals.
- Data-hungry and labour-intensive to build, calibrate, and validate, especially microsimulations.
- Many parameters and rules create risks of overfitting and of equifinality (different rules, same fit).
- Sensitive to initial conditions and stochastic seeds, requiring many runs to characterize uncertainty.
- Validation against the future is impossible; calibration to the past does not guarantee predictive skill.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
What is the difference between cellular-automata and agent-based urban models?
In a cellular-automaton model the unit of analysis is a cell of land that changes state (e.g. non-urban to urban) according to rules based on its neighbours and suitability, so the focus is on spatial pattern and the agents are implicit. In an agent-based or microsimulation model the units are explicit decision-makers — households, firms, developers — that choose locations and actions, so the focus is on behaviour and heterogeneity. Cellular automata excel at modelling physical expansion; agent-based models excel at who does what and why.
How is UrbanSim different from a traditional land-use–transport model?
Traditional models such as the Lowry framework allocate aggregate population and employment to zones at equilibrium. UrbanSim instead simulates disaggregate households, jobs, and real-estate development making location and development choices through estimated discrete-choice models, with land prices and transport accessibility updated dynamically each year. It is behavioural, disaggregate, and dynamic rather than aggregate and equilibrium-based, which makes it better suited to policy and equity analysis but far more data-intensive.
Can these models predict the future of a city?
Not in a literal, deterministic sense. Urban simulation models are best understood as scenario engines: calibrated to past behaviour and patterns, they show how the city might plausibly evolve under explicit assumptions and policies. Because they are stochastic, data-limited, and sensitive to parameters, their value lies in comparing scenarios and revealing mechanisms and uncertainty, not in producing a single precise forecast. Treating their outputs as conditional 'what-ifs' rather than predictions is essential.
Sources
- 1.Waddell, P. (2002). UrbanSim: Modeling urban development for land use, transportation, and environmental planning. Journal of the American Planning Association, 68(3), 297–314.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Urban Simulation Model. ScholarGate. https://scholargate.app/urban-studies/urban-simulation-model