Development of a Digital Twin for Smart Building over Edge-Cloud Continuum

A smart building relies on an integration of physical and computational components to produce a secure, comfortable, and energy-efficient environment for its occupants. Each smart building is distinct in terms of requirements and characteristics. Hence, developing smart buildings to achieve reliability and real-time environmental adaption is a challenging task. This paper describes the development of a proof-of-concept digital twin for management and maintenance of an academic building. The physical building's real-time data is gathered from a distributed set of IoT sensors and cameras together with the facility management system's APIs into its digital twin. Then, the data is processed across both edge devices and in the cloud environment using machine learning models to generate insights required to manage daily operation of its physical components holistically. This work can help inform future efforts in creating a universal framework for digital twin design for smart building applications.