Utilizing Machine Learning Approaches for Assessing the Unconfined Compressive Strength of a Crushed Stone-Cement-Pozzolan Mixture

Abstract

This research examines the use of artificial pozzolans, industrial waste products, as a sustainable alternative to traditional cement in soil improvement works aiming to reduce environmen-tal damage caused by industrial production. Various regression and machine learning techniques, including Support Vector Re-gression, Random Forest, Extreme Gradient Boosting, Ridge, LASSO, and Elastic Net Regression, are applied to predict uncon-fined compressive strength (UCS) with various compositions of cementitious crushed stone soil samples with cylindrical shapes. Cylindrical cementitious crushed stone soil samples have different contents including different amounts of water, bind-er, and pozzo-lan additives like fly ash, silica fume, and granulated blast furnace slag, type of pozzolan, curing time, and number of layers of cylin-drical samples and blow for each compacted layer. Among the algorithms utilized, XGBoost performs better in estimating UCS with an RMSE of 1.5920, an MSE of 2.5346, and an MAE of 1.2858. The study underscores the environmental benefits of com-bining experimental approaches with advanced Artificial Intelli-gence methods, offering a sustainable solution that recycles waste materials while enhancing performance.

More by Dr. Zarina Oflaz

• Optimal adaptive control of methane production in anaerobic digestion via reinforcement learningJournal of Process Control · 2026
• Development of machine learning models for intelligent prediction of workability, mechanical, and durability characteristics of cement mortar made of recycled sandConstruction and Building Materials · 2026 · Citations: 1
• Leveraging eco-friendly pozzolans and machine learning for reduced cement use in soil improvement of road base layersEuro-Mediterranean Journal for Environmental Integration · 2026
• Comparative Analysis of Prediction Accuracy in Clean Energy Forecasting: Implications for Pricing and Profitability in Wind and Biogas Energy SystemsSpringer proceedings in earth and environmental sciences · 2026