Innovative AI strategies for enhancing smart building operations through digital twins: A survey

Abstract

The Digital Twins (DT) have emerged as a digital transformation automation process with ubiquitous applications that span various domains, including buildings, manufacturing, and healthcare. These virtual clones of physical systems provide relevant insights, enhance decision-making processes, and optimize operations, along with allowing the prediction of future operations. Artificial intelligence (AI) has been instrumental in enhancing the functionalities of DT. This survey paper explores recent developments in advanced AI algorithms tailored for DT in building settings. Moreover, a wide spectrum of AI techniques designed to address the challenges posed by DT in buildings are categorized and reviewed, including convolution neural networks (CNN), recurrent neural networks (RNNs), and generative adversarial networks (GANs), among other cutting edge transformative technologies. Furthermore, the integration of reinforcement learning (RL) and transfer learning (TL) into the DT ecosystem is discussed. This survey explores practical use cases, such as predictive scenarios, anomaly detection, and optimization of DT models. The incorporation of multimodal AI sensor data and edge computing in enhancing the accuracy and efficiency of DT is analyzed. Additionally, challenges and future directions in the field are explored, including data privacy concerns using Blockchain (BC), scalability issues, and the potential impact of quantum computing (QC) and large language models (LLMs) on DT technology. This comprehensive survey serves as a valuable resource for researchers, practitioners, and decision makers looking to utilize cutting-edge techniques to harness the full potential of DT technology in smart buildings (SB).