Estimation of Soil Water Characteristic Curve Using Machine-Learning Algorithms and Its Application in Embankment Response

Abstract

The parameters of the soil water characteristic curve (SWCC) play a pivotal role in the examination of unsaturated soil behavior. This study employs three machine learning models—random forest (RF), extreme gradient boosting (XGBoost), and multiexpression programming (MEP)—to predict the SWCC using key soil properties. Among them, the RF model demonstrated the most robust performance in SWCC prediction. The Shapley Additive Explanation (SHAP) analysis further reveals that suction is the most influential factor affecting SWCC predictions, with other input parameters also contributing significantly. Additionally, the MEP model offers a straightforward expression for SWCC estimation and, thus, proved practical for predicting embankment responses and exhibited superior accuracy over traditional methods, such as the Arya and Paris model (ACAP). For a precise assessment of the hydromechanical response of the embankment subjected to infiltration, an increase in pore pressure is observed when employing the MEP model compared to the ACAP model for fine-grained soils. The findings emphasize the potential of RF and MEP in enhancing SWCC prediction and their practical implications for soil engineering applications. This research has significant practical applications in civil engineering and geotechnical fields. By employing machine learning models such as RF, XGBoost, and MEP, this study provides a cost-effective and efficient method for predicting the SWCC, a critical factor in assessing unsaturated soil behavior. These predictions enhance embankment design and safety by enabling better stability assessments under varying moisture conditions, thus mitigating failure risks. The MEP model, in particular, offers a simple yet effective tool for estimating SWCC in typical embankment scenarios, improving predictions of hydromechanical responses to infiltration and enabling precise water management. Furthermore, this study’s use of SHAP deepens the understanding of the relationship between SWCC and soil parameters, fostering innovation in predictive modeling. This research aids in sustainable construction practices by optimizing moisture control and reducing extensive soil treatments. Additionally, the methodologies and findings serve as valuable educational resources and can drive further advancements in machine learning applications within geotechnical engineering, ultimately contributing to safer, more efficient, and environmentally friendly engineering practices.