School cluster discovery using spatial pattern analysis: a study for economic management of education

India’s National Education Policy 2020 recommends the clustering of schools to improve access to educational resources, with each cluster anchored by a secondary school. This study evaluates clustering for 3451 schools in Varanasi district, India. Spatial point pattern analysis was conducted using Ripley’s K function and Besag’s L function to determine whether such clustering is feasible. All school categories, except tribal welfare department schools, display significant clustering within a 5–10 km range, supporting the formation of proximity-based complexes. K-means, Voronoi polygon-based and Fuzzy C-means (FCM) clustering were applied to analyse the geolocations of schools. FCM produced the most coherent configuration (BRISQUE 56.92), supporting evidence-based complex planning.