Digital Twin Framework for Personalized Building Management in Ambient Assisted Living

Ambient Assisted Living (AAL) systems, which integrate digital healthcare with Smart Home concepts, have become increasingly important for supporting independent living, particularly for aging populations and individuals with physical impairments. This paper proposes a sensor-based Digital Twin (DT) framework that incorporates user-centric perspectives to enhance AAL systems and improve energy management in smart homes. To this end, the framework relies on a comprehensive ontology describing the semantic behaviour and relationships of relevant buildings, their occupants, and connected objects inside them. Unlike traditional approaches, which focus solely on building performance, this framework integrates occupants' preferences and habits as critical elements contributing to both energy efficiency and personalized care. Through data collection mechanisms and machine learning techniques, the framework provides occupants with valuable insights into how their activities, particularly those involving appliance usage, impact overall energy consumption in their homes. This is highly beneficial for AAL applications, enabling occupants to make informed decisions on sustainable and efficient energy use, and ensuring a safer and more comfortable living environment. As a result, the paper outlines the proposed ontology and the initial implementation of the system prototype, which leverages the Microsoft Azure platform and its Cosmo DB No-SQL database. The prototype is complemented with a user-centric awareness dashboard, constructed using Flask and accessible through a web application interface.