TY - JOUR
T1 - A Survey on Digital Twin for Industrial Internet of Things
T2 - Applications, Technologies and Tools
AU - Xu, Hansong
AU - Wu, Jun
AU - Pan, Qianqian
AU - Guan, Xinping
AU - Guizani, Mohsen
N1 - Publisher Copyright:
© ; 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Digital twin for the industrial Internet of Things (DT-IIoT) creates a high-fidelity, fine-grained, low-cost digital replica of the cyber-physical integrated Internet for industry. Powered by artificial intelligence (AI) and security technologies, DT-IIoT provides advanced features such as real-time monitoring, predictive maintenance, remote diagnostics, and rapid response for smart IIoT systems. A systematic review of key enabling technologies such as digital twin, AI, and blockchain is essential to develop DT-IIoT and reveal pitfalls. This paper reviews the preliminaries, real-world applications, architectures and models of digital twin-driven IIoT. In addition, advanced technologies for intelligent and secure DT-IIoT are investigated, including state-of-the-art AI solutions such as transfer learning and federated learning, as well as blockchain-based security solutions. Moreover, software tools for high-fidelity digital twin modeling are proposed. A case study on reinforcement learning-based integrated-control, communication, and computing (3C) design is developed to demonstrate the AI-driven intelligent DT-IIoT. Finally, this paper outlines the prospective applications, challenges, and integrations with ABCDE (i.e., AI, Blockchain, cloud computing, big data, edge computing) as the future directions.
AB - Digital twin for the industrial Internet of Things (DT-IIoT) creates a high-fidelity, fine-grained, low-cost digital replica of the cyber-physical integrated Internet for industry. Powered by artificial intelligence (AI) and security technologies, DT-IIoT provides advanced features such as real-time monitoring, predictive maintenance, remote diagnostics, and rapid response for smart IIoT systems. A systematic review of key enabling technologies such as digital twin, AI, and blockchain is essential to develop DT-IIoT and reveal pitfalls. This paper reviews the preliminaries, real-world applications, architectures and models of digital twin-driven IIoT. In addition, advanced technologies for intelligent and secure DT-IIoT are investigated, including state-of-the-art AI solutions such as transfer learning and federated learning, as well as blockchain-based security solutions. Moreover, software tools for high-fidelity digital twin modeling are proposed. A case study on reinforcement learning-based integrated-control, communication, and computing (3C) design is developed to demonstrate the AI-driven intelligent DT-IIoT. Finally, this paper outlines the prospective applications, challenges, and integrations with ABCDE (i.e., AI, Blockchain, cloud computing, big data, edge computing) as the future directions.
KW - Digital twin
KW - artificial intelligence
KW - blockchain
KW - industrial Internet of Things
KW - integrated design
UR - http://www.scopus.com/inward/record.url?scp=85178095641&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85178095641&partnerID=8YFLogxK
U2 - 10.1109/COMST.2023.3297395
DO - 10.1109/COMST.2023.3297395
M3 - Article
AN - SCOPUS:85178095641
SN - 1553-877X
VL - 25
SP - 2569
EP - 2598
JO - IEEE Communications Surveys and Tutorials
JF - IEEE Communications Surveys and Tutorials
IS - 4
ER -