Application of Machine Learning Techniques for Predicting the Compressive Strength of GGBS-Based Geo-polymer Mortar

DOI

10.2991/978-94-6239-697-5_13How to use a DOI?

Keywords

GGBS; Compressive Strength; Geo-polymer; Multiple Linear Regression; XG-Boost

Abstract

The process of determining the ideal mixture of alkaline activators enables scientists to achieve higher compressive strength results for Geo-Polymer mortar with the used of Ground Granulated Ballast Furnace Slag (GGBS) that undergoes ambient temperature curing. The research used a combined methodology which included laboratory tests together with machine learning (ML) modeling to investigate how different sodium hydroxide (NaOH) concentrations from 6M to 14M sodium silicate(SS) (Na₂SiO₃) Solution to The Sodium hydroxide (SH) (NaOH) solution ratio affected strength development. In this Study standard mortar production methods which required a consistent binder: sand (1:3) and an activator: binder (0.70). In experimental studies compressive strength at two distinct curing periods which were 7 and 28 days. In Experimental Result produced that compressive strength developed a distinct nonlinear pattern which depended on the sodium hydroxide solution concentration. The 7-day strength at lower molarities which ranged from 6M to 8M showed strength values between 22 MPa and 30 MPa. The 28-day compressive strength reached its highest points of 60 to 62 MPa when specimens received activation through 10 to 14 M NaOH solution, resulting in a 25 to 30 percent increase over the results from lower molarity solution mixes. In this study created predictive models by using experimental data which they obtained through Multiple Linear Regression (MLR) Model, a Support Vector Machine (SVM) Model, Random Forest Modeling (RF) and Extreme Gradient Boosting (XG-Boost) methods. The XG-Boost model achieved the best predictive performance which produced determination coefficients of approximately 0.97 at 7 days and 0.98 at 28 days together with minimal error measurements, which included RMSE below 1.7 MPa and MAE below 1.3 MPa. The research findings proved that machine learning methods based on ensemble techniques can precisely predict compressive strength, which enables reliable and sustainable methods to optimize concrete mixes and design GGBS-based geo-polymer mortars for performance.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ram Bahadur
AU  - Arun Kumar
AU  - Shreekanth Birgonda
AU  - Ajay Kumar
PY  - 2026
DA  - 2026/06/04
TI  - Application of Machine Learning Techniques for Predicting the Compressive Strength of GGBS-Based Geo-polymer Mortar
BT  - Proceedings of the Conference on Bridging Engineering Disciplines with AI and Machine Learning (BEDAIML 2026)
PB  - Atlantis Press
SP  - 137
EP  - 149
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-697-5_13
DO  - 10.2991/978-94-6239-697-5_13
ID  - Bahadur2026
ER  -