Geospatial Retail Market Analysis (Project)
Developed a geospatial data pipeline to collect and prepare retail location data for analysis using OSM sources and drive-time clustering. Implemented spatial clustering to group raw points into commercial centers and compute a feasibility score for market viability decisions. The work focuses on structuring location datasets rather than training an ML labeler model. • Harvested 1,087 retail nodes via Overpass API. • Used QGIS and DBSCAN to cluster points into centers within a 20-minute ORS drive-time isochrone. • Engineered a feasibility scoring algorithm to evaluate anchor-store synergy. • Filtered to 80 high-potential commercial centers for downstream placement targets.