Please check the GitHub link for full reproducible code. The code has extensive notes about what each line of the ipfp function does: https://github.com/nipunkalraa/Spatial-Microsimulation-Food-and-Alcohol-in-the-UK-R

This project uses spatial microsimulation techniques to estimate the average daily consumption of various food and drink items, most notably alcohol, across Oxford's Lower Super Output Areas (LSOAS). It generates high-resolution, synthetic population data statistically aligned with official demographics, enabling localised public health insights.

📊 Datasets Used

• individuals.csv – Synthetic individuals with consumption attributes

• Constraint datasets (age_sex.csv, Ethnicity.csv, Health.csv, Work.csv) – UK 2011 Census-based marginal totals

• Geographic Focus: Final simulation and visualization are filtered for Oxford, UK (70+ LSOAs)

🔧 Methods & Workflow

Microsimulation Engine: An Iterative Proportional Fitting Procedure (IPFP) matches synthetic individuals to area-level demographic constraints.

Constraint Variables:

• Age-Sex combinations

• Ethnicity

• Health status

• Employment status

• Consumption Outputs:

• Estimated daily intake of 13 items, including meats (e.g., pork, beef), plant-based foods (fruits, vegetables), and alcohol.

🖼️ Map: Estimated Alcohol Consumption in Oxford (ml/day)