Parametric optimization of urban residential morphology for outdoor thermal comfort with integrated code regulatory and site development requirement

Urban heat island phenomena pose threats to outdoor thermal comfort in high-density residential areas in hot-humid climatic cities. This study presented an integrated parametric optimization framework that combined empirical data collection, environmental simulation and multi-objective genetic algorithms in order to improve the thermal comfort in residential neighborhoods. Based on the morphological structure of 972 residential areas located in Shenzhen's four central districts, morphological parameter ranges were defined for three basic layouts: row style, cluster style, and courtyard style. Using Rhino+Grasshopper platform, parametric models were created and optimized using coupled thermal comfort (UTCI), wind environment, and daylighting performance simulations. AIJ case and field validation was performed to ensure model accuracy. Multi-objective optimization demonstrated that thermal comfort can be improved by 0.2–0.6 °C UTCI with daylighting compliance imposed. The best values of mean UTCI for normal conditions in a summer week were found to be 31.66–33.05 °C for all configurations in comparison with 32.19–33.53 °C for worst-case scenarios. Regression analysis showed that the parameter sensitivities were layout dependent with cluster-style configurations showing strongest correlation (R² = 0.398–0.625) between morphological variables and thermal performance. The application of the Bao'an district case study in Shenzhen also confirmed practical applicability in good integration with local regulatory limitations. © 2026 Elsevier B.V.