Agentic AI-Enabled Framework for Thermal Comfort and Building Energy Assessment in Tropical Urban Neighborhoods

Researchers from Singapore (Po-Yen Lai, Xinyu Yang, Derrick Low, Huizhe Liu, Jian Cheng Wong) have developed a novel agentic AI framework that integrates large language models (LLMs) with lightweight physics-based models to tackle urban heat island effects and building energy demands in tropical cities. The system, detailed in a paper accepted at IAQVEC 2026, uses prompt customization to let LLMs interpret urban design tasks, extract relevant policies, and autonomously activate appropriate physics-based models for evaluation. These streamlined models leverage core thermal and airflow principles, reducing computational time while accurately predicting microclimate variables like building surface temperature, ground radiant heat, and airflow conditions. This enables real-time estimation of thermal comfort indices, such as physiological equivalent temperature (PET), and building energy usage.

The framework allows users to explore a range of climate-resilient building surface strategies—like green façades and cool paint applications—that improve thermal comfort while reducing wall heat gain and energy demand. By combining the autonomous reasoning capacity of LLMs with rapid quantitative evaluation, the system demonstrates potential for cross-disciplinary applications in sustainable urban design, indoor-outdoor environmental integration, and climate adaptation planning. The source code and data are publicly available on GitHub, marking a significant step toward data-driven, AI-enabled urban climate solutions.