Using machine learning to predict concrete’s strength: Learning from small datasets

Using machine learning to predict concrete’s strength: Learning from small datasets

OUYANG, Boya (University of California, USA)
LI, Yuhai (University of California, USA)
SONG, Yu (University of California, USA)
WU, Feishu (University of California, USA)
YU, Huizi (University of California, USA)
WANG, Yongzhe (University of California, USA)
BAUCHY, Mathieu (University of California, USA)
SANT, Gaurav (University of California, USA)

2021-04-29

Despite previous efforts to relate concrete proportioning and strength, a robust knowledgebased model for accurate concrete strength predictions is still lacking. As an alternative to physical or chemical-based models, machine learning (ML) methods offer a new solution to this problem. Although ML can handle the complex, non-linear, non-additive relationship between concrete mixture proportions and strength, it requires large datasets. This is a concern as reliable strength data is rather limited, especially for industrial concretes. Here, based on a large dataset (>10,000 observations) of measured compressive strengths from industrially-produced concretes, we compare the ability of select ML algorithms to “learn” how to reliably predict concrete strength as a function of the size of the dataset. Based on these results, we discuss the competition between how accurate a given model can eventually be (when trained on a large dataset) and how much data is actually required to train this model.

modeling
concrete
strength prediction
machine learning

http://resolver.tudelft.nl/uuid:cb7d725a-53bb-44be-a4ce-3f8aa5a18e9d

Conference proceedings

conference paper

(c) 2021 the authors