AI-Driven Digital Twins for Energy-Efficient Building Operations: A Design Science Research Approach

In the global pursuit of carbon neutrality, improving the operational energy efficiency of buildings has become a central challenge for the architecture, engineering, and construction sectors. Advances in Artificial Intelligence (AI) and Digital Twin technologies offer new opportunities to optimize building performance through data-driven decision-making. This research investigates the role of AI-driven Digital Twins in building operations, positioning them as an effective approach for enhancing sustainability, reducing energy consumption, and supporting intelligent control strategies.

Design Science Research Framework

The study adopts a Design Science Research (DSR) methodology to systematically guide the development, implementation, and evaluation of a digital artifact for energy-efficient building operation. DSR enables the structured creation of a practical solution while ensuring theoretical rigor through iterative problem identification, artifact design, demonstration, and evaluation. This methodological approach ensures that the proposed Digital Twin framework addresses real-world operational challenges while contributing to knowledge in sustainable building systems.

Six-Layer Conceptual Architecture for Digital Twins

A six-layer conceptual architecture is proposed to support the integration of real-time building data, AI-based analytics, and intelligent control mechanisms. The architecture enables seamless data acquisition, processing, prediction, and decision-making across multiple system layers. By structuring the Digital Twin ecosystem in this way, the framework ensures scalability, interoperability, and adaptability, making it suitable for diverse building types and operational contexts.

Machine Learning-Based Energy Prediction

As a practical instantiation of the proposed architecture, the study presents an application case involving three buildings, where machine learning models are employed to predict energy consumption. Several regression algorithms are evaluated, with the Cat Boost Regressor achieving strong predictive performance, reaching R² values above 0.92. These results demonstrate the reliability of data-driven models in capturing complex energy-use patterns and supporting informed operational decisions.

Digital Shadow Implementation and Intelligent Control

The trained energy prediction model is deployed as a Digital Shadow using the open-source platform Node-RED, enabling real-time interaction between data streams and predictive analytics. Beyond energy forecasting, the architecture supports intelligent control strategies, as demonstrated through simulated application cases such as load shifting and anticipatory HVAC activation. These scenarios highlight the system’s ability to move from passive monitoring toward proactive and adaptive building operation.

Implications for Sustainable Building Performance

The results validate the effectiveness of the proposed AI-driven Digital Twin artifact and underscore its potential to enhance building performance and sustainability. By leveraging flexible, cost-effective, and open-source tools, the framework lowers barriers to adoption while enabling advanced operational intelligence. This research contributes to the growing field of smart and sustainable buildings, offering a practical pathway for integrating AI and Digital Twins into performance-oriented building management practices.

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/architectureengineersawards Pinterest :   https://in.pinterest.com/architectureengineersawards/ Blogger :   https://architectureengineers.blogspot.com/ Twitter :   https://x.com/Architectu54920 YouTube :  https://www.youtube.com/@Architechtureengineer LinkedIn :  https://www.linkedin.com/in/architecture-engineer-01a044361/