Radiation Model of Mobility and Migration
Also known as: Radiation Law of Human Mobility, Parameter-free Mobility Model, Simini Radiation Model, Radyasyon Modeli
The Radiation Model, introduced by Simini et al. in 2012, is a parameter-free model for predicting human mobility and migration flows between geographic locations. Drawing an analogy from radiation physics, it predicts trip volumes based solely on population sizes at origin and destination, and the intervening population within the circle connecting them. It has been widely applied to commuting flows, migration, and epidemic spreading.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use the Radiation Model when predicting commuting, migration, or mobility flows between spatial units and calibration data for gravity-model parameters are unavailable or unreliable. It assumes population size is a valid proxy for opportunity density, and that individuals select the nearest acceptable opportunity. It performs best at aggregate spatial scales with reliable census population data. When detailed impedance functions or mode-specific costs matter, gravity or intervening opportunity models may be more appropriate.
Strengths & limitations
- Requires no free parameters — flows are predicted directly from population data without calibration
- Grounded in a mechanistic analogy (radiation physics) that produces interpretable, testable predictions
- Shown to outperform the gravity model for commuting prediction in large-scale empirical tests
- Applicable across spatial scales from intra-urban commuting to international migration
- Population size is assumed to be a direct proxy for opportunity, which may not hold in economically heterogeneous regions
- Requires reliable census or population data at fine spatial resolution; quality degrades with coarse or outdated inputs
- Does not incorporate travel cost, transportation network structure, or mode-specific friction
- Performance can be weaker for long-distance or international flows where cultural, legal, and economic barriers dominate
Frequently asked
Why is the Radiation Model called parameter-free?
Unlike the gravity model, which requires empirically fitted exponents for population and distance, the Radiation Model derives flow predictions from population counts alone using a fixed analytic formula. There are no coefficients to estimate from observed flows, making it immediately applicable wherever census data exist and avoiding overfitting to historical patterns.
How does the Radiation Model differ from the gravity model?
The gravity model predicts flows as proportional to origin and destination populations divided by a power of distance, with parameters calibrated to data. The Radiation Model replaces distance with the intervening population, requires no calibration, and uses a mechanistic formula grounded in opportunity competition rather than an empirical analogy to Newtonian gravity.
Can the Radiation Model be applied to international migration?
It can be applied, but with caution. International migration flows are shaped by visa regimes, language barriers, income differentials, and colonial ties that population size does not capture. Empirical studies show the model performs well for domestic commuting but less reliably for cross-border flows, where augmented versions incorporating economic or policy variables are often preferred.
Sources
- Simini, F., González, M. C., Maritan, A., & Barabási, A.-L. (2012). A universal model for mobility and migration patterns. Nature, 484, 96–100. DOI: 10.1038/nature10856 ↗
How to cite this page
ScholarGate. (2026, June 2). Radiation Model of Mobility and Migration. ScholarGate. https://scholargate.app/en/spatial-analysis/radiation-model
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Huff ModelSpatial analysis↔ compare
- Migration ModelsDemography↔ compare
- Spatial Interaction ModelSpatial analysis↔ compare