Population Synthesis and Spatial Microsimulation Techniques

Population synthesis entails the generation of complete virtual populations whose demographic, socioeconomic and spatial attributes mirror those of real‐world populations. Spatial microsimulation extends this idea by allocating synthetic individuals or households to fine‐grained geographic units, ensuring that aggregate characteristics conform to observed small‐area totals. Together, these techniques underpin agent-based models, small-area estimation, transport demand forecasting and policy impact assessment across disciplines from urban planning to public health. Core methods include constraint‐based approaches such as iterative proportional fitting, combinatorial optimisation and Monte Carlo sampling, each balancing accuracy, computational efficiency and data availability. Recent methodological advances have addressed integerisation of non‐integer weights, uncertainty quantification through credible intervals and multi‐resolution population synthesis. Open-source pipelines now enable reproducible workflows that integrate diverse data sources—census microdata, travel surveys and mobile phone records—while retaining transparency in constraint selection and weighting. The global significance of these tools lies in their capacity to simulate “what-if” scenarios for future population change, to inform infrastructure investment, to model disease spread or to evaluate the social impacts of policy interventions in regions lacking detailed local data.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Researchers have released an open-source synthetic population dataset for Canada at fine geographic granularity, incorporating detailed socioeconomic attributes and projecting years up to 2042. This dataset has been rigorously validated against census counts and offers a flexible platform for “what-if” scenario analysis and extension with local survey inputs. In another contribution, a reproducible framework for Paris and Île-de-France demonstrates the construction of a synthetic travel demand database from open data, linking households, individuals and daily activity chains for transport simulations; this work highlights the effects of implicit correlation structures on downstream modelling. Methodological innovation has also been achieved through an integerisation procedure known as ‘truncate, replicate, sample’, which converts non‐integer weights from iterative proportional fitting into discrete individuals, enhancing accuracy and computational speed for spatial microsimulation applications.

Population Synthesis and Spatial Microsimulation Techniques publication trend

The graph below shows the total number of articles in population synthesis and spatial microsimulation techniques across all publications each year (not limited to Nature Index journals).

Technical terms

Synthetic population: A statistically generated dataset of individuals or households whose attributes match the marginal distributions of real populations.

Spatial microsimulation: A method that combines synthetic populations with geographic constraints to produce detailed small-area estimates.

Iterative proportional fitting (IPF): An algorithm that adjusts joint distributions of microdata to align with known marginal totals.

Integerisation: The conversion of fractional weights from fitting procedures into whole counts of unique individuals for simulation models.

References

1. A synthetic population for agent-based modelling in Canada. Scientific Data (2023).
2. Synthetic population and travel demand for Paris and Île-de-France based on open and publicly available data. Transportation Research Part C Emerging Technologies (2021).
3. Population Synthesis Handling Three Geographical Resolutions. ISPRS International Journal of Geo-Information (2018).
4. ‘Truncate, replicate, sample’: A method for creating integer weights for spatial microsimulation. Computers Environment and Urban Systems (2013).
5. Estimating uncertainty in spatial microsimulation approaches to small area estimation: A new approach to solving an old problem. Computers Environment and Urban Systems (2017).
6. Introducing the eqasim pipeline: From raw data to agent-based transport simulation. Procedia Computer Science (2021).
7. Getting the best of both worlds: a framework for combining disaggregate travel survey data and aggregate mobile phone data for trip generation modelling. Transportation (2020).