What question did this study set out to answer?

The aim is to forecast the flexural strength of waste aggregate concrete beams reinforced with fiber-reinforced polymers using machine learning methods.

April 15, 2026Buildings0 citationsOpen Access

Predicting Flexural Strength of FRP-Strengthened Waste Aggregate Concrete Beams with Machine Learning: A Step Towards Sustainability

ASArissaman SangthongtongNavamindradhiraj University BCBurachat ChatveeraThammasat University GSGritsada Sua-iam

Key Points

The aim is to forecast the flexural strength of waste aggregate concrete beams reinforced with fiber-reinforced polymers using machine learning methods.
Developed and assessed four machine learning algorithms: Random Forest, Decision Tree, Neural Network, XGBoost.
Utilized 92 experimental datasets for model training and validation.
Employed regression plots and statistical measures for performance evaluation.
Used SHAP and LIME for model interpretability.
Random Forest outperformed Neural Network in predictive accuracy.
XGBoost showed performance comparable to Random Forest.
Key predictors included beam length, fiber length, steel tensile strength, fiber width, and concrete compressive strength.
Partial dependence plots identified optimal parameter ranges for flexural strength prediction.

Abstract

Using waste materials in the manufacture of concrete has many environmental advantages. However, it can be difficult to estimate structural performance, especially when beams are reinforced with fiber-reinforced polymers (FRP). In order to provide a data-driven approach to sustainable structural design, this work explores the use of machine learning (ML) approaches to forecast the flexural strength of FRP-strengthened waste aggregate concrete beams. A total number of 92 experimental datasets were used to develop and assess four ML algorithms: Random Forest (RF), Decision Tree (DT), Neural Network (NN), and Extreme Gradient Boosting (XGBoost). Regression plots, Taylor diagrams, statistical measures (R2R²R2, RMSE, MAE, MSE), and explainable AI (XAI) tools, including SHAP, LIME, and partial dependence plots (PDPs), were used to evaluate the model’s performance. RF outperformed NN in terms of predictive accuracy, while XGBoost exhibited similar performance to RF. The most significant predictors, according to a SHAP analysis, were beam length and fiber length, with the lower followed by steel tensile strength, fiber width, and concrete compressive strength. LIME offered local interpretability for individual predictions, but PDPs demonstrated optimal parameter ranges and a nonlinear feature strength relationship. The findings provide engineers with a strong decision-support tool for designing green infrastructure, since they show that ensemble-based models can accurately represent the intricate, nonlinear dynamics controlling flexural behavior in sustainable FRP-strengthened waste aggregate concrete beams.

Perguntar à IA

Bookmark

View Full Paper