This project built a high-resolution population model for an industrial city in Saudi Arabia, to support infrastructure planning, service location and long-term expansion. It was developed for the Saudi government authority that plans and manages the city. Dr. Ghulam Mohey-ud-din led the spatial economic analysis.
The problem
City planners needed to know where people live and work at neighbourhood and block level, not just totals by district. Official statistics were too coarse to decide where schools, clinics, mosques, retail and emergency services should go, or to test how different industrial growth paths would change demand.
Method
- Building footprints were extracted and classified from high-resolution satellite imagery using machine-learning-assisted image recognition in Google Earth Engine.
- Population density surfaces were calibrated against census blocks and administrative records, using LandScan ambient population data (1 km resolution, Oak Ridge National Laboratory) as a starting layer.
- A 100 m × 100 m population grid was produced for all residential, commercial and mixed-use zones.
- Outputs were validated against ground-truth survey data.
Use
The model lets planners identify population concentrations, optimise the siting of public facilities and model growth scenarios under different industrial expansion paths. It feeds directly into the city’s master plan update and its GIS systems.
Related service: Spatial Economic Diagnostics.