Slope Stability Predictions using Machine Learning Techniques

Abstract

The geotechnical analysis is considered to be an important aspect in providing a safe mine working environment. It not only covers the active monitoring of open pit walls but also effectively predicts the slope deformations and failures. The approach or method used in slope failure is considered to be legitimate when it predicts the time of failure of slope prior to its actual failure. This research plays a paramount role in mitigating the risk associated with slope failure. The aim of this study is to demonstrate the application of the Machine Learning technique to effectively predict the occurrence of slope failure. Specifically, random forest, support vector classifier, and logistic regression algorithms are employed to assess the stability of the slopes. The dataset included in the study uses cohesion, angle of friction, and unit weight of the designed slopes. The performance of the implemented machine learning models for the factor of safety (FOS) prediction is analysed and compared.