The Smart Building paradigm promises a future where buildings are intelligent, adaptive, and sustainable, offering real-time information retrieval that supports decision-making to enhance energy efficiency, occupant comfort, and security. However, achieving this paradigm is highly complex, one major reason being the seamless integration of (a) physical and functional representations of buildings (i.e., Building Information Modeling) and (b) real-time IoT (i.e., Internet of Things) data. To address this challenge, we propose a hybrid LLM–Knowledge Graph approach in which, on the one hand, a knowledge graph retains the buildings’ knowledge structures and a time-series database stores IoT data. On the other hand, a Large Language Model (LLM) serves as a mediator between a facility manager and the knowledge graph and IoT data, facilitating data-driven decision-making processes. Adopting the Design Science Research (DSR) methodology, we executed two distinct iterations within the implementation phase to investigate contrasting knowledge representation paradigms: Resource Description Framework (RDF) and Labeled Property Graph (LPG). Each iteration was evaluated to assess the effectiveness of the corresponding systems. Finally, we compared the RDFand LPG-based implementations of the proposed architecture and drew insights.
Laurenzi et al. (Thu,) studied this question.