Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

AIoT-informed digital twin communication for bridge maintenance

Gao, Yan ORCID:, Li, Haijiang ORCID:, Xiong, Guanyu and Song, Honghong 2023. AIoT-informed digital twin communication for bridge maintenance. Automation in Construction 150 , 104835. 10.1016/j.autcon.2023.104835

[thumbnail of 1-s2.0-S092658052300095X-main.pdf] PDF - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (13MB)


Digital twin (DT) has been moving progressively from concept to practice for bridge operation and maintenance (O&M), but its issues of data synchronization and fault tolerance remain problematic. This paper investigates the time delay of bridge DT services according to communication and computation complexity, revealing the distinct impact of their sequence, and proposes an AIoT-informed DT communication framework to solve the above issues. The information hierarchy and two-way communication can be leveraged to minimize communication complexity in the framework. Meanwhile, the data flow and resilience of the proposed framework are demonstrated using a Petri net. Moreover, the framework is developed into a prototypical DT through cross-platform integration and validated with different cases. The results demonstrate that compared with other existing bridge DTs, the proposed framework has high efficiency, low-latency, and excellent fault tolerance, which can contribute to the efficiency and safety of bridge O&M, especially under communication-constraint circumstances. The framework is also promising for federated learning to protect the AI-model privacy of different stakeholders and has the potential to support agent-based intelligent bridge management in the future with little human intervention.

Item Type: Article
Date Type: Published Online
Schools: Engineering
Publisher: Elsevier
ISSN: 0926-5805
Date of First Compliant Deposit: 22 March 2023
Date of Acceptance: 11 March 2023
Last Modified: 19 May 2023 01:00

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics