Climate-Adaptive Thermal Comfort Optimization in Architectural Design
Climate change has intensified the challenges associated with maintaining indoor–outdoor thermal comfort, prompting the need for advanced, data-driven design approaches. This study introduces an integrated framework combining Multiobjective optimization (MOO) and explainable machine learning (ML) to analyze how spatial morphology affects indoor–outdoor thermal comfort (IOTC). By merging global optimization capability with transparent interpretability, the framework supports climate-responsive architectural decision-making and delivers insights that improve both design quality and environmental performance.
Multiobjective Optimization for Thermal Comfort
The framework employs a genetic algorithm (GA)–based MOO model to optimize nine morphological parameters related to building and courtyard forms. These parameters serve as decision variables for the simultaneous optimization of predicted mean vote (PMV) and the universal thermal climate index (UTCI) during contrasting seasonal conditions. By minimizing discomfort levels in summer while maximizing comfort in winter, the model provides a balanced and seasonally adaptive architectural strategy.
Machine Learning for Accurate Thermal Comfort Prediction
An ensemble machine learning model, enhanced through Bayesian hyperparameter optimization, was developed to predict thermal comfort with high accuracy. Among the tested algorithms, XGBoost demonstrated superior predictive performance, making it the preferred choice within the framework. Its ability to model complex, nonlinear relationships between morphological variables and thermal comfort outcomes ensures robust and dependable design evaluations.
Explainable AI for Morphological Parameter Interpretation
To overcome the opacity of traditional black-box models, SHapley Additive exPlanations (SHAP) analysis was applied to interpret the relative contributions of each morphological parameter. The results reveal that the building shape index (BSI), morphological coefficient (FSC), and building-to-courtyard ratio (BCR) collectively explain more than 76% of thermal comfort variability. Their strong interaction effects—particularly between BSI and BCR—highlight the crucial role of spatial configuration in climate-adaptive design.
Key Morphological Drivers of Thermal Comfort
The findings emphasize how specific geometric characteristics directly influence heat exchange, airflow, and environmental stability. Higher-impact parameters such as BSI and BCR shape the microclimate dynamics within courtyard-based architectural forms, while FSC supports optimized massing and orientation strategies. Understanding these drivers enables architects to modify form-related variables with precision, achieving measurable improvements in IOTC performance.
A Scalable Decision-Support Framework for Architects
By integrating optimization capability with explainability, the proposed framework serves as a powerful decision-support tool for architects and urban designers. It provides both optimal design solutions and the reasoning behind those solutions, making the process transparent and scalable for diverse climatic contexts. This interpretable and data-driven approach represents a major advancement toward climate-adaptive, high-performance architectural design.
Architecture Engineers Awards
🔗 Nominate now! 👉 https://architectureengineers.com/award-nomination/?ecategory=Awards&rcategory=Awardee
🌐 Visit: architectureengineers.com
📩 Contact: contact@architectureengineers.com
Get Connected Here:
*****************
Instagram : https://www.instagram.com/architecture_engineers_awards/
Facebook : https://www.facebook.com/profile.php?id=61576995475934
Tumblr : https://www.tumblr.com/blog/architectureengineers
Pinterest : https://in.pinterest.com/researcherawards123/
Blogger : https://architectureengineers.blogspot.com/
Twitter : https://twitter.com/
#ThermalComfort
#ClimateAdaptiveDesign
#ArchitecturalResearch
#MultiobjectiveOptimization
#MachineLearning
#ExplainableAI
#XGBoost
#SHAPAnalysis
#BuildingMorphology
#CourtyardDesign
#IOTC
#GeneticAlgorithm
#SustainableArchitecture
#EnvironmentalDesign
#Microclimate
#UrbanThermalComfort
#DataDrivenDesign
#HighPerformanceDesign
#ClimateResilience
#ArchitecturalScience
Comments
Post a Comment