An explainable machine learning model for encompassing the mechanical strength of polymer-modified concrete

Aug 5, 2026

Upload by Sage

Polymer-modified concrete (PMC) offers improved durability, tensile strength, adhesion, and reduced chemical degradation. Machine learning has emerged as a practical approach for predicting compressive strength, supporting optimized mix design. This study trains eight ML models—decision tree, SVM, KNN, bagging regression, XG-Boost, Ada-Boost, linear regression, and gradient boosting—using 382 literature-based experimental points. SHAP and SHAP interaction analyses, plus partial dependence plots, identify the dominant inputs and explain model behavior.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages24

Reading Time1 h 0 min

AccessFree to download

Introduction

Concrete properties and motivation for improved prediction

Polymer-modified concrete and compressive strength design needs

Machine learning approaches in concrete strength prediction

Asian Journal of Civil Engineering [https://doi.org/10.1007/s42107-024-01230-6](https://doi.org/10.1007/s42107-024-01230-6) An explainable machine learning model for encompassing the mechanical strength of polymer‑modified concrete Md. Habibur Rahman Sobuz1 · Mita Khatun1 · Md. Kawsarul Islam Kabbo1 · Norsuzailina Mohamed Sutan2 Received: 10 November 2024 / Accepted: 18 November 2024 © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024 Abstract Polymer-modified concrete (PMC) is an advanced building material with more excellent durability, tensile strength, adhesion, and lesser susceptibility to chemical degradation. Recent developments in machine learning (ML) have shown that prediction of compressive strength (CS) of PMC key input factors needed to obtain an optimized mix design are among the areas of applicability of ML. This study used eight machine learning models, which are Decision Tree, Support Vector Machine, K-Nearest Neighbors, Bagging Regression, XG-Boost, Ada-Boost, Linear Regression, Gradient Boosting to predict compressive strength and perform SHAP (Shapley additive explanation) analysis. These hybrid predictive PMC models were developed using a wide-ranging dataset of 382 experimental data points compiled from the literature. A SHAP interaction plot was also used to show how each feature affected predictions on the model outputs. As highlighted in the results, hybrid models had significantly higher performance than conventional models, and the XG-Boost and decision tree model had the highest accuracy. In particular, the XG-Boost and decision tree model reached R2 scores of 0.987 for training and 0.577 for testing, proving its remarkable prediction ability for PMC compressive strength. The SHAP analysis confirmed that coarse aggregate, cement, and SCMs had the most significant influence on CS, with all other variables contributing lower values. The Partial Dependence Plots (PDP) analysis allowed a relatively simple interpretation of the contribution of individual inputs to the CS predictions. These results are useful for construction purposes and provide engineers and builders with first-hand knowledge and insight into the importance of individual components on PMC development and performance. Keywords Compressive strength prediction · Machine learning · Polymer modified concrete · SHAP · PDP analysis Introduction Concrete is one of the most commonly used construction materials that shows better results in terms of performance and durability. It is brittle and has low tensile strength, with crack generation under flexural loading (Ahmed & Sobuz, 2011 ; Habibur Rahman Sobuz et al., 2023a, 2023b; Hasan et al. , 2015) . Additionally, traditional concrete has low deformability and compressive toughness, hindering its ability to sustain dynamic loads (Kaveh, 2022) . Polymer is found to enhance concrete properties, such as durability, * Md. Habibur Rahman Sobuz [habib@becm.kuet.ac.bd](habib@becm.kuet.ac.bd) 1 Department of Building Engineering and Construction Management, Khulna University of Engineering and Technology, Khulna 9203, Bangladesh 2 Department of Civil Engineering, University Malaysia Sarawak, 94300 Kota Samarahan, Sarawak, Malaysia impact resistance, and water absorption capacity, which are essential for engineering structures and the maintenance of concrete structures (Hasan et al. 2023a, 2023b, 2023c, 2023d, 2023e; Saha et al. , 2020 ; Saha, Tonmoy, et al. , 2024a, 2024b; Sobuz et al. , 2012 ; Uddin et al. , 2012) . So, it can be used as an alternative to improve concrete quality. Prediction of compressive strength is essential for concrete design and use in construction. Most conventional concrete prediction models are built on empirical relationships and regression techniques, which commonly do not capture possible nonlinear interactions between concrete inputs/ outputs in complex compositions like polymer-modified concrete (PMC). Polymer-modifie

FAQs

Which machine learning models are used to predict polymer-modified concrete compressive strength?

The study evaluates eight models: decision tree, SVM, K-nearest neighbors, bagging regression, XG-Boost, Ada-Boost, linear regression, and gradient boosting.

How is explainability handled in the proposed approach?

SHAP analysis and SHAP interaction plots explain how features influence predictions, and partial dependence plots provide an additional interpretation of input contributions.

What factors are found to most strongly affect compressive strength in SHAP results?

Coarse aggregate, cement, and SCMs show the most significant influence on compressive strength, while other variables contribute lower values.

Aug 5, 2026

Upload by Sage