PSI - Issue 70

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 70 (2025) 432–439

review under the responsibility of International Conference on Structural Integrity Organizers Keywords: High Strength Concrete; Predictive Model; Machine Learning; Compressive Strength Abstract High Strength Concrete (HSC) is required for the construction of Tall buildings, bridges and dams which are subjected to static and dynamic loading. Many Researches are carried out globally for developing high strength concrete using sustainable cementitious materials. These researches contribute lot of data for the upcoming future development of HSC. The vast availability of these data can be efficiently utilized on Machine learning to predict various properties of HSC with different sustainable materials. The objective of this paper is to develop a predictive modelling of HSC properties using machine learning techniques like Random forest, XGB Regression, Decision Tree Regression and Linear Regression. The paper focuses on the high strength ranging from 60MPa to 100MPa and aims to predict the 28 th day strength of HSC specifically for different combination of supplementary cementitious materials. On a dataset of 355 data, 20% were tested and 80% were trained by using ten input parameters. The XGB Regressor shows high R2 Score of 0.85 less Mean Squared Error of 0.23 and Mean Absolute Error of 2.66, thus makes the predicted value 95% accurate than the other algorithms. The ML model is validated by casting and testing of HSC with 8 different compositions for M70 grade concrete. By using this ML model optimization of HSC with different industrial waste becomes easy which reduces the experimentation time and cost and also reduces material wastage. © 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 Structural Integrity and Interactions of Materials in Civil Engineering Structures (SIIMCES-2025) Predictive modelling for the material optimization of high strength concrete using ML Abdul Musavvir a , Dharun Vikash a , Kabilan a , Sathish Kumar a , S. Lavanya Prabha a, * a Civil Engineering, Easwari Engineering College, Chennai, Tamilnadu, India

* Corresponding author. Tel.: +91-9444064940 E-mail address: lavanyaprabha.s@eec.srmrmp.edu.in

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.2025.07.074