Architecting Smart City Digital Twins: Combined Semantic Model and Machine Learning Approach

Abstract

This work was motivated by the premise that next-generation smart city systems will be enabled by widespread adoption of sensing and communication technologies deeply embedded within the physical urban domain. These technological advances (e.g., sensing, processing, and data transmission) are what makes smart city digital twins possible. This paper explores approaches and challenges in architecting and the operation of smart city digital twins. A smart city digital twin architecture is proposed that supports semantic knowledge representation and reasoning, working side by side with machine learning formalisms, to provide complementary and supportive roles in the collection and processing of data, identification of events, and automated decision-making. The semantic and machine learning sides of the proposed architecture are exercised on a problem involving simplified analysis of energy usage in buildings located in the Chicago Metropolitan Area.