With the rapid development of infrastructure construction, the sector faces prominent challenges such as high accident rates, low productivity, and skilled labor shortages. Artificial intelligence (AI) has emerged as a key solution to address these issues, yet a comprehensive analysis of its current application landscape and future directions in infrastructure construction remains necessary. Therefore, Ke CHEN, Xiaojie ZHOU, Zhikang BAO, Mirosław Jan SKIBNIEWSKI, and Weili FANG from institutions including Huazhong University of Science and Technology, Heriot-Watt University, and the University of Maryland have jointly conducted a research entitled “Artificial Intelligence in Infrastructure Construction: A Critical Review”.
This study adopts both quantitative and qualitative analysis methods to systematically explore the application of AI in infrastructure construction. For the quantitative analysis, 594 relevant papers from the Web of Science database (2013-2023) were selected, and tools like VOSviewer were used to analyze publication distributions, co-authorship networks, and keyword co-occurrence networks. The results show that the number of AI-related research papers in infrastructure construction has grown rapidly since 2018, with China, the United States, and Australia leading in research; the top journals, represented by “Automation in Construction”, have made significant contributions; and research focus is mainly on safety monitoring and control, followed by process management. For the qualitative analysis, 91 high-citation papers were selected, which were categorized into four areas: safety monitoring and control (covering target detection and tracking, posture recognition, compliance checking, safety risk assessment and prediction), process management (including progress tracking and monitoring, schedule prediction and optimization), cost estimation and prediction, and quality assessment. The analysis details the application of key AI technologies such as machine learning, computer vision, and natural language processing in each area, demonstrating AI’s advantages in improving construction efficiency and safety.
The study also points out several directions for future research, including expanding the scope of AI applications to cover more aspects like environmental performance of construction, exploring the potential of underutilized AI technologies such as robotics, and enhancing AI applications through standardized datasets, integration of domain knowledge, and generative AI models. These insights provide important guidance for the further development of AI in infrastructure construction, with significant engineering practical value.
The paper “Artificial Intelligence in Infrastructure Construction: A Critical Review” authored by Ke CHEN, Xiaojie ZHOU, Zhikang BAO, Mirosław Jan SKIBNIEWSKI, Weili FANG. Full text of the open access paper: https://doi.org/10.1007/s42524-024-3128-5.