Integrating AI and spatial analysis for resilient and inclusive 15-minute cities: A case study of Gdańsk

This study explores the application of artificial intelligence (AI) and geospatial tools to design 15-min cities, focusing on proximity, public space quality, environmental resilience, and inclusion. Using the post-industrial Young City district in Gdańsk, Poland, as a case study, the research integrates semantic segmentation (DeepLabv3) and AI-driven object detection for evaluating public space quality, flood simulation modeling (InVEST UFRM) for climate resilience, and network analysis using adjusted walking speeds (2.95 km/h) to capture the needs of vulnerable populations. By identifying “hotspots” with compounded deficits in access to urban amenities, public space quality, and resilience, the study provides a replicable, data-informed tool for guiding equitable urban transformation. The findings reveal critical spatial and environmental disparities, particularly in former fence-line neighborhoods in Gdańsk, which remain underserved despite ongoing urban redevelopment. The proposed framework offers practical insights for mainstreaming inclusive adaptation strategies into planning policies and contributes to the evolving discourse on operationalizing the 15-min city (FMC) concept in complex contexts. • Introduces an enhanced 15-minute city (FMC) framework combining spatial quality, environmental resilience, and inclusivity. • Applies advanced geospatial tools, including DeepLabv3 (segmentation) and InVEST flood modeling, for scalable urban analysis. • Reveals hotspots where accessibility, livability, and climate resilience deficits overlap, mainly in former fence-line zones. • Demonstrates a replicable methodology that bridges proximity metrics with public space quality and climate adaptation. • Supports inclusive urban transformation by noting the need for equitable investment in historically marginalized areas.