PSI - Issue 68

ScienceDirect Structural Integrity Procedia 00 (2025) 000–000 Structural Integrity Procedia 00 (2025) 000–000 Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect Available online at www.sciencedirect.com ScienceDirect

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Procedia Structural Integrity 68 (2025) 245–251

European Conference on Fracture 2024 Prediction of the compressive strength of normal weight concrete with variable fly ash percentages using machine learning Hussam Safieh a , Rami A. Hawileh a, *, Sayan Kumar Shaw a , Jamal A. Abdalla a a American University of Sharjah, Department of Civil Engineering, P.O. Box 26666, Sharjah, United Arab Emirates Abstract This study explores how the strength of Ordinary Portland Cement (OPC) can be predicted, specifically looking at how different levels of fly ash impact it. The goal was to create models that can accurately predict OPC’s strength using machine learning techniques such as Linear Regression, Support Vector Machines (SVM), and Random Forest. Among these methods, Random Forest stood out as the most effective, significantly outperforming the other models. The dataset consisted of concrete samples with varying fly ash percentages and curing periods, with compressive strength as the target variable. Random Forest achieved the highest R² value of 0.7893 and the lowest Root Mean Squared Error (RMSE) of 6.5392, indicating its superior performance in predicting compressive strength compared to simpler models. The model’s ability to capture non-linear relationships made it well suited to understanding the complex interactions between fly ash content, curing time, and strength. This research highlights the potential of using machine learning to optimize concrete mix designs for enhanced performance. © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers Keywords: Ordinary Portland Cement (OPC); Fly Ash; Machine Learning; Compressive Strength Prediction; Random Forest; Support Vector Machines (SVM). European Conference on Fracture 2024 Prediction of the compressive strength of normal weight concrete with variable fly ash percentages using machine learning Hussam Safieh a , Rami A. Hawileh a, *, Sayan Kumar Shaw a , Jamal A. Abdalla a a American University of Sharjah, Department of Civil Engineering, P.O. Box 26666, Sharjah, United Arab Emirates Abstract This study explores how the strength of Ordinary Portland Cement (OPC) can be predicted, specifically looking at how different levels of fly ash impact it. The goal was to create models that can accurately predict OPC’s strength using machine learning techniques such as Linear Regression, Support Vector Machines (SVM), and Random Forest. Among these methods, Random Forest stood out as the most effective, significantly outperforming the other models. The dataset consisted of concrete samples with varying fly ash percentages and curing periods, with compressive strength as the target variable. Random Forest achieved the highest R² value of 0.7893 and the lowest Root Mean Squared Error (RMSE) of 6.5392, indicating its superior performance in predicting compressive strength compared to simpler models. The model’s ability to capture non-linear relationships made it well suited to understanding the complex interactions between fly ash content, curing time, and strength. This research highlights the potential of using machine learning to optimize concrete mix designs for enhanced performance. © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers Keywords: Ordinary Portland Cement (OPC); Fly Ash; Machine Learning; Compressive Strength Prediction; Random Forest; Support Vector Machines (SVM). © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers

* Corresponding author. Tel.: +971 6 515 2496; fax: +971 6 515 2979. E-mail address: rhaweeleh@aus.edu 1. Introduction * Corresponding author. Tel.: +971 6 515 2496; fax: +971 6 515 2979. E-mail address: rhaweeleh@aus.edu 1. Introduction

Concrete is the second-favoured construction medium after water due to its global availability of raw materials, simplicity of processing and handling, and capacity to transition from a fluid condition, capable of solidifying in a mould and withstanding structural loads (Wangler et al., (2019)). Concrete is primarily made up of cement, sand, coarse aggregate, water, and other components. The production process for Portland cement involves the calcination of calcium oxide emits CO 2 and has a significant environmental impact (Barcelo et al., (2014); Kajaste et al., (2016)). In recent years, research interests have switched towards applying various ANN models to predictive-based challenges involving building materials such as composites, steel, and concrete (Abdalla & Hawileh, (2011), (2013); Concrete is the second-favoured construction medium after water due to its global availability of raw materials, simplicity of processing and handling, and capacity to transition from a fluid condition, capable of solidifying in a mould and withstanding structural loads (Wangler et al., (2019)). Concrete is primarily made up of cement, sand, coarse aggregate, water, and other components. The production process for Portland cement involves the calcination of calcium oxide emits CO 2 and has a significant environmental impact (Barcelo et al., (2014); Kajaste et al., (2016)). In recent years, research interests have switched towards applying various ANN models to predictive-based challenges involving building materials such as composites, steel, and concrete (Abdalla & Hawileh, (2011), (2013);

2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers 2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers

2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of ECF24 organizers 10.1016/j.prostr.2025.06.049