Artificial intelligence in infrastructure construction: A critical review

Ke Chen, Xiaojie Zhou, Zhikang Bao, Mirosław Jan Skibniewski, Weili Fang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)
199 Downloads (Pure)

Abstract

Artificial intelligence (AI) has emerged as a promising technological solution for addressing critical infrastructure construction challenges, such as elevated accident rates, suboptimal productivity, and persistent labor shortages. This review aims to thoroughly analyze the contemporary landscape of AI applications in the infrastructure construction sector. We conducted both quantitative and qualitative analyses based on 594 and 91 selected papers, respectively. The results reveal that the primary focus of current AI research in this field centers on safety monitoring and control, as well as process management. Key technologies such as machine learning, computer vision, and natural language processing are prominent, with significant attention given to the development of smart construction sites. Our review also highlights several areas for future research, including broadening the scope of AI applications, exploring the potential of diverse AI technologies, and improving AI applications through standardized data sets and generative AI models. These directions are promising for further advancements in infrastructure construction, offering potential solutions to its significant challenges.
Original languageEnglish
JournalFrontiers of Engineering Management
Early online date5 Jul 2024
DOIs
Publication statusE-pub ahead of print - 5 Jul 2024

Keywords

  • artificial intelligence
  • infrastructure construction
  • literature review
  • qualitative analysis
  • quantitative analysis

ASJC Scopus subject areas

  • Decision Sciences (miscellaneous)
  • Management Science and Operations Research

Fingerprint

Dive into the research topics of 'Artificial intelligence in infrastructure construction: A critical review'. Together they form a unique fingerprint.

Cite this