Machine Learning to Assist Structural Engineers: A Bibliometric Analysis 2020-2025

Resumo

The role of Machine Learning (ML) in modern society is increasingly critical, driving advancementsacross numerous fields, including structural engineering. This paper presents a concise review of ML applicationsin structural engineering, based on English-language publications indexed in the Scopus database between 2020and 2025. A bibliometric analysis was conducted following the PRISMA methodology to address the followingresearch questions: What are the main applications of ML in structural engineering? Who are the leadingcontributors? Which publications are most frequently cited? Where are the papers being published? A total of737 articles were identified and analyzed. The results reveal key areas of application, including structural design,the development of constitutive models, reliability analysis, and structural health monitoring. The data show anexponential growth in publications over the past five years, with Chinese institutions leading in research output.Notable contributions have also been made by Brazil and Colombia within the South American context. Amongthe most cited papers, key topics include structural health monitoring and reliability analysis, with a particularlyinfluential study advancing the understanding of explainability and interpretability in ML. The findings highlightthe significant potential of ML to enhance the structural engineering field.