Digital Twins (DTs) are becoming central to the architecture, engineering, construction and operations (AECO) domain, enabling real-time monitoring, simulation, and life cycle-encompassing optimization of building operations through unified virtual representations of physical assets. However, practical deployment remains challenging due to fragmented data sources, domain-specific engineering workflows, and the absence of standardized integration frameworks. To address these issues, we present a model-driven framework for scalable DT deployment, grounded in Building Information Modeling (BIM). Our approach automates the extraction of domain semantics and the generation of runtime interfaces, thereby reducing manual effort and enhancing maintainability. The framework integrates Eclipse Hono and Ditto for real-time data exchange and leverages Unreal Engine for rich, interactive visualization. It also features standardized extension points to enable modular, service-oriented DT platforms. We demonstrate the practical application of our approach in a Living Lab case study and evaluate the entanglement of the resulting DT using the Overall Digital Twin Entanglement (ODTE) metric, underscoring its interoperability and architectural soundness.