<p>Urban green spaces (UGS) are essential components of sustainable cities, providing ecosystem services, health benefits, and opportunities for social interaction. Despite their importance, access to UGS remains highly uneven, particularly in dense and socioeconomically vulnerable urban areas. This study assesses pedestrian accessibility to leisure green spaces (LGS) in Sintra, Portugal, and develops a spatially explicit framework to prioritise locations for new green space provision. Pedestrian accessibility was modelled using network-based service area analysis with a 15-minute walking threshold, accounting for street connectivity and topographic constraints. Results indicate that approximately 21% of the municipal population (≈ 82,000 residents) lacks walkable access to any qualifying LGS, with the most acute deficits occurring in high-density eastern neighbourhoods. To address these inequities, a GIS-based multi-criteria decision analysis (MCDA) was applied, integrating demographic pressure, pedestrian accessibility, land-use suitability, proximity to schools, and environmental constraints. Criteria weights were derived using the Analytic Hierarchy Process (AHP), yielding a consistent weighting scheme (Consistency Ratio = 0.04) that prioritised population density (29%) and walkable accessibility (25%). The analysis identified 96 highly suitable candidate areas (&gt; 80% suitability), from which five priority sites (14–27&#xa0;ha) were selected. Network-based estimates suggest that these interventions could improve walkable access to LGS for approximately 68,200 residents. By moving beyond simple proximity measures, this study demonstrates how equity-oriented, network-based spatial analysis can support the translation of the 15-minute city and environmental justice goals into concrete planning decisions for urban green infrastructure.</p>