AI‐Driven Multi‐City Optimization of Glazing and Shading Systems for Building Energy Use and Operational Carbon Reduction Across Global Climate Zones

Buildings account for a major share of global energy demand and operational carbon emissions, underscoring the importance of climate-responsive façade design in sustainable architecture. Façade parameters such as glazing ratio, shading systems, and glazing properties strongly influence thermal performance of buildings; however, most optimization studies remain confined to single climates and rarely provide practical design guidance. This study introduces an AI-assisted optimization framework for evaluating façade performance across four contrasting climate zones: Abu Dhabi (hot-arid), Singapore (hot-humid), Athens (Mediterranean), and Berlin (temperate). A standardized five-story residential prototype was modeled in DesignBuilder using the EnergyPlus simulation engine. Window-to-Wall Ratio, orientation, shading depth, and glazing type were optimized using the Non-Dominated Sorting Genetic Algorithm II to minimize Energy Use Intensity (EUI) and operational CO₂ emissions while maintaining thermal comfort through ASHRAE 55 discomfort-hour constraints. The optimized solutions reduced EUI by approximately 54% in Abu Dhabi, 56% in Singapore, 43% in Athens, and 40% in Berlin relative to baseline conditions. A post-optimization AI-assisted rule-based interpretation layer applied to identify recurring façade parameter patterns among Pareto-optimal solutions revealed a systematic transition from low-glazing, deep-shading solutions in hot climates toward higher glazing ratios with limited shading in temperate environments. These findings indicate that façade optimization is climate-dependent and that AI-assisted workflows can support interpretable, performance-driven architectural decision-making.