Artificial Neural Network Approach to Evaluate Geotechnical Properties of Lime-Treated Mountain Soil: A Slope Stabilization Perspective

Measuring geotechnical parameters like unconfined compressive strength, elastic modulus and shear strength of lime-treated mountain soil (in the present study, termed as artificially structured soil) in the laboratory is a resource-intensive, time-consuming and challenging task. This paper addresses these challenges by developing predictive models to estimate these geotechnical parameters of lime-treated mountain soil using artificial neural network. A practical connection between the strength parameters and the stability analysis of hill slopes by deploying the limit equilibrium method has also been established. The result shows that ANN-based models exhibit superior predictive capabilities compared to traditional statistical method such as the multiple linear regression analysis. Additionally, the impact of incorporating the engineered properties of soil, such as those achieved through additives, i.e. hydrated lime, on slope stability models has been explored. The findings of the study reveal that such modifications lead to an increase in the slope’s factor of safety. Consequently, this research underscores the potential of conventional soil treatments, such as hydrated lime, to positively influence slope stability.