TITLE:
Artificial Intelligence-Assisted Green Building Technologies for Urban Regeneration: A Design Support Framework for Low-Carbon, Climate-Adaptive Neighborhoods
AUTHORS:
Eric Nkurikiye
KEYWORDS:
Artificial Intelligence, Green Building Technologies, Urban Regeneration, Machine Learning, Climate Adaptation, Multi-Objective Optimization
JOURNAL NAME:
Current Urban Studies,
Vol.14 No.4,
November
10,
2026
ABSTRACT: Rapid urbanization and escalating climate change pressures have intensified the need for systematic, evidence-based approaches to urban regeneration, as existing built environments contribute approximately 40% of global energy consumption and 36% of CO2 emissions. Existing research on the application of artificial intelligence (AI) to the integrated selection of green building technology packages at the neighborhood scale remains limited, with conventional design workflows lacking the computational capacity to evaluate the vast combinatorial solution space of technology interactions, and with current frameworks rarely integrating thermal comfort, climate adaptation, and explainability within a unified multi-objective structure. This study aims to develop and validate a comprehensive AI-assisted design support framework enabling rapid, transparent, and evidence-based selection of optimal green building technology packages for low-carbon, climate-adaptive urban regeneration. To address this, a multi-scale framework was employed that couples city-level climate scenario modelling (RCP 4.5/8.5), neighbourhood-level microclimate simulation (CFD/ENVI-met), and building-level energy analysis (EnergyPlus/IES-VE) through an AI core engine comprising machine learning surrogate models (ANN, Random Forest, XGBoost), NSGA-II/III multi-objective evolutionary optimisation, urban morphology clustering, and SHAP-based explainable AI, trained and validated on a synthesis of 241 peer-reviewed studies (2022-2025). The results indicate that AI-optimized technology packages achieve 22% - 30% cooling energy reductions versus 6.5% - 20% for individual interventions, outdoor thermal comfort improvements of 2.0˚C - 3.5˚C (UTCI), carbon payback periods of 4 - 7 years, and a 19-percentage-point climate-resilience advantage over conventional retrofits under RCP 8.5 (2050), while reducing design iteration time from weeks to hours (surrogate R2 > 0.92). These findings suggest that AI-assisted, multi-scale, explainable optimization provides a scalable, transferable, and stakeholder-transparent pathway for accelerating low-carbon, climate-adaptive urban regeneration from individual buildings to district-scale projects of 500 - 2000 buildings.