Document Type

Article

Publication Date

10-16-2026

Journal / Book Title

iScience

Abstract

Recycled aggregate concrete (RAC) supports more sustainable construction, but recycled concrete aggregate (RCA)'s heterogeneity complicates compressive-strength prediction. This study evaluates the Tabular Prior-data Fitted Network (TabPFN) against nine conventional machine-learning models using a standardized literature-derived database of 833 RAC mixtures. Cube strengths were converted to equivalent cylinder strengths, with conversion sensitivity also assessed. Among conventional models, CatBoost achieved the highest accuracy (R2 = 0.843), whereas TabPFN achieved R2 = 0.870, mean absolute error (MAE) = 3.924 MPa, and root mean squared error (RMSE) = 5.551 MPa. TabPFN also produced a lower RMSE than CatBoost across all four RCA replacement intervals without extensive task-specific hyperparameter tuning. Feature importance and Shapley additive explanation (SHAP) analyses identified effective water-to-cement ratio, aggregate-to-cement ratio, RCA density, and RCA absorption as key predictors, while RCA replacement level alone had limited influence. The moderate improvement over CatBoost indicates that further gains depend on improved RCA characterization.

DOI

10.1016/j.isci.2026.117559

Rights

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Published Citation

Fan, Jin, et al. “Assessing Machine Learning Prediction of Recycled Aggregate Concrete Strength with a Tabular Foundation Model.” iScience, vol. 29, no. 10, Oct. 2026, p. 117559. https://doi.org/10.1016/j.isci.2026.117559.

Download
Link to Publisher

Share

COinS