Strength Prediction of Modified Clayey Soil with Municipal Solid Waste and Nano-MgO Using AI-Driven Models

This study explores a sustainable approach to clay soil stabilization by incorporating Municipal Solid Waste (MSW) and nano-magnesium oxide (nano-MgO), combined with Machine Learning (ML) for strength prediction. A dataset of 243 laboratory tests was analyzed using Multiple Linear Regression (MLR), Support Vector Regression (SVR), and Artificial Neural Networks (ANN) to estimate Unconfined Compressive Strength (UCS). Comparative analysis revealed that ANN achieved the highest predictive accuracy (R² = 0.90 for training and 0.94 for testing) with the lowest error metrics, outperforming SVR and MLR. Sensitivity analysis indicated that nano-MgO content was the most influential factor, while curing time had minimal impact. Parametric analysis confirmed that increasing nano-MgO significantly improved UCS, whereas higher MSW content reduced strength. These findings demonstrate that ML-based models can reliably predict UCS, reducing reliance on costly and time-consuming laboratory tests. The integration of MSW and nano-MgO offers an environmentally friendly alternative to traditional stabilizers, supporting circular economy principles and reducing CO₂ emissions in geotechnical applications.