Title:
Data-driven Concrete Science: Applications, Challenges, and Best Practices
Author(s):
Li
Publication:
Web Session
Volume:
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
4/2/2023
Abstract:
With the ability to tackle complex tasks autonomously, data-driven techniques, such as machine learning, have opened up new research avenues in concrete science. Here, we provide an extensive bibliographic analysis of peer-reviewed publications in concrete science and describe the status and most compelling applications of data-driven models in this field. We discuss recurring methodological challenges found in the literature regarding data preparation, model validation, and model interpretation. We conclude by providing best practice recommendations and outlining future directions for data-driven concrete research.