Rocznik Ochrona Środowiska 2026, vol. 28, pp. 310-328
Meshel Q. Alkahtani 
 |
King Khalid University, Saudi Arabia |
 |
|
 |
https://doi.org/10.54740/ros.2026.022 |
   |
|
 |
The traditional soil stabilization techniques normally utilize cement and lime, which are known to lead to high carbon emissions and environmental degradation. This has led to increased interest in ecologically friendly alternatives, such as bio-based additives. Nevertheless, the shear strength behavior of bio-stabilized soils cannot be easily predicted from differences in soil properties and treatment conditions. ML models were developed using experimental data: Random Forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoost). Input variables: additive content (0-12%), dry density, moisture content (OMC±2%), and curing period (0-56 days). Output: peak shear strength (kPa). XGBoost achieved the highest prediction accuracy (R² = 0.96, RMSE = 3.2 kPa on test data). Model predictions validated with direct comparison to experimental values. This paper explores the shear strength performance of soils stabilized with bio-based additives using a combination of experimental and machine learning (ML) methodologies. Direct shear tests on soil samples with varying percentages of bio-based additives were conducted under controlled laboratory conditions of moisture and curing. ML models were then created using the experimental data, including Random Forests and Artificial Neural Networks, to predict shear strength as a function of important input variables, such as additive content, dry density, and moisture content. Findings show that bio-based stabilization greatly reduces shear strength, and the best additive ranges yield the greatest benefits. The ML models demonstrated high predictive performance, with a strong correlation between measured and predicted values. The results outline opportunities to combine experimental testing with ML tools to optimize sustainable soil stabilization. This design will help achieve sustainable geotechnical construction by reducing the use of conventional chemical stabilizers and encouraging the use of renewable resources.
shear strength, bio-based additives, soil stabilization, machine learning, sustainable geotechnics
 |
 |
AMA Style
Alkahtani M.. Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications. Rocznik Ochrona Środowiska. 2026; 28. https://doi.org/10.54740/ros.2026.022
ACM Style
Alkahtani M.. 2026. Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications. Rocznik Ochrona Środowiska. 28. DOI:https://doi.org/10.54740/ros.2026.022
ACS Style
Alkahtani M., Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications Rocznik Ochrona Środowiska 2026, 28, 310-328. https://doi.org/10.54740/ros.2026.022
APA Style
Alkahtani M. (2026). Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications. Rocznik Ochrona Środowiska, 28, 310-328. https://doi.org/10.54740/ros.2026.022
ABNT Style
ALKAHTANI M.. Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications. Rocznik Ochrona Środowiska, v. 28, p. 310-328, 2026. https://doi.org/10.54740/ros.2026.022
Chicago Style
Meshel Q. Alkahtani. 2026. "Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications". Rocznik Ochrona Środowiska 28, 310-328. https://doi.org/10.54740/ros.2026.022
Harvard Style
Alkahtani M. (2026) "Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications", Rocznik Ochrona Środowiska, 28, pp. 310-328. doi:https://doi.org/10.54740/ros.2026.022
IEEE Style
Alkahtani M., "Experimental and Machine Learning Investigation of Shear Strength of Soils Stabilized with Bio-Based Additives for Environmentally Friendly Applications", RoczOchrSrod, vol 28, pp. 310-328. https://doi.org/10.54740/ros.2026.022