PSI - Issue 70

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 70 (2025) 461–468

Structural Integrity and Interactions of Materials in Civil Engineering Structures (SIIMCES-2025) Machine Learning-Based Prediction of Concrete Compressive Strength Incorporating GGBS Ch. Vihas a , P. Rama Rao a, *, DLM Santosh Kumar a , V. Vishwak Sena Reddy a , N. Abhilash a a Department of Civil Engineering, VNR Vignana Jyothi Institute of Engineering & Technology, Hyderabad, India, 500090. Abstract This study examines the application of machine learning in predicting the compressive strength of concrete incorporating GGBS. A dataset comprising 560 experimental values, which include 13 features, was collected from existing literature. To understand the patterns and trends in data through exploratory data analysis (EDA), which facilitates data cleaning, statistical analysis, and visualization. The Machine Learning models used in this study are Artificial Neural Networks (ANN), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Machines (SVM). The hyperparameters of each model were tuned to increase the performance of the ML model. The predictive models were validated through the experimental casting of concrete samples based on machine learning predictions. The XGBoost model demonstrated superior performance. It is observed that compressive strength exhibits a direct correlation with the age of concrete and GGBS content in the concrete, suggesting that an increased curing period and higher GGBS proportions contribute to enhanced strength development. The study reveals the capability of machine learning (ML) techniques in optimizing concrete mix design by providing accurate strength predictions, reducing experimental costs, and enhancing material efficiency. Future research could enhance model robustness by integrating larger datasets and employing deep learning architectures. The present work advances data-driven approaches in sustainable concrete design, particularly in utilizing industrial by-products, such as GGBS, for improved mechanical 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 the responsibility of International Conference on Structural Integrity Organizers

Keywords: Random Forest; XGBoost; ANN; SVM; Hyperparameter tuning.

* Corresponding author. Tel.: +91-9542240112 E-mail address: ramaraopanugalla@gmail.com

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 the responsibility of International Conference on Structural Integrity Organizers 10.1016/j.prostr.2025.07.078