• Spatial disaggregation of explicit spatial attributes during population synthesis. • Open-source tool development for spatializing aggregated data to coordinate points. • Spatialization on zone units, along roads, and within buildings. • Application of random distribution to sample coordinate points. Modeling and simulation are crucial for decision-making in transportation, often relying on synthetic populations when real data are unavailable due to privacy concerns or insufficient resolution. While considerable literature has focused on generating synthetic individuals and their sociodemographic attributes, the allocation of these individuals to real geographic locations remains underexplored. This paper addresses this gap by introducing SpatialzOSM, an open-source tool designed to spatialize synthetic populations by converting aggregated spatial data into real coordinates using three sampling techniques: across areas, along roads, and within building footprints. We apply SpatialzOSM to synthetic populations from Townsville and Frankfurt am Main, demonstrating its adaptability, scalability, and transferability to different urban contexts and zoning levels. The results highlight trade-offs between realism, computational efficiency, and accuracy. SpatialzOSM enhances the realism and accuracy of explicit-location-based models, offering a flexible solution for synthesizing geographically explicit locations in urban studies. Our approach provides researchers and practitioners with a reproducible pathway to create foundational data that captures the crucial link between where people live, where they conduct activities, and how they travel.
Toaza et al. (Wed,) studied this question.