ABSTRACT
In the case of sustainable construction, reducing energy consumption in buildings is essential. With global energy consumption on the rise, leveraging the Internet of Things (IoT), simulation tools and more advanced machine learning (ML) algorithms provides an innovative opportunity to optimize energy usage. To this end, this study combines both traditional model-based approaches and modern data-driven approaches (specifically with regard to ML models such as CatBoost and Random Forest) to integrate the Integrated Environmental Solutions Virtual Environment (IESVE) simulation tool while creating a comprehensive framework for energy assessment. The accuracy of the ML method that predicts heating and cooling demand is greatly enhanced through the real-time data acquisition made possible by IoT. The approach of this research, combining simulation-driven analysis with machine learning, enables energy conservation through practical solutions in the context of smart and sustainable infrastructure.
