Scholarlink

This paper presents a comprehensive stability analysis of slopes commonly encountered in the hilly terrains of North-East (NE) region of India. The study employs Limit Equilibrium Method (LEM) using GeoStudio v2018 to assess the stability considering both static and earthquake-induced loading scenarios. Pseudo-static analyses are performed by incorporating earthquake acceleration coefficients for Zone V as per IS 1893 Part 1. Various geotechnical parameters and slope geometries pertinent to the NE region are investigated to evaluate slope stability, with the Factor of Safety (FoS) serving as the primary criterion. A systematic parametric study is conducted to explore the influence of different factors on slope stability such as slope geometry, shear strength parameters of soil and location of water table. Furthermore, Regression and Machine Learning (ML) techniques are applied over the generated dataset to establish correlations between FoS and the influencing factors, thereby aiding in the formulation of predictive relationships for assessing stability of slopes. The relationships achieved through this study exhibit a high level of accuracy, with regression and ML-based algorithms yielding correlation coefficients of 96.43% and 98.03% for static case, 95.81% and 98.15% for earthquake-induced loading case, respectively. The established correlations can provide an easy tool for practicing engineers to judge the slope stability with least effort, thereby facilitating quick first-hand predictions of risk assessments. This integrated approach combining traditional engineering methods with advanced data-driven techniques paves the way to offer a robust framework for analyzing landslide hazards and enhancing disaster mitigation efforts in the region.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of Prediction Equations for Static and Pseudo-Static Stability of Slopes Using Regression and Machine Learning Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rubi Chakraborty,
  • Furquan Ahmad,
  • Arindam Dey,
  • Pijush Samui

摘要

This paper presents a comprehensive stability analysis of slopes commonly encountered in the hilly terrains of North-East (NE) region of India. The study employs Limit Equilibrium Method (LEM) using GeoStudio v2018 to assess the stability considering both static and earthquake-induced loading scenarios. Pseudo-static analyses are performed by incorporating earthquake acceleration coefficients for Zone V as per IS 1893 Part 1. Various geotechnical parameters and slope geometries pertinent to the NE region are investigated to evaluate slope stability, with the Factor of Safety (FoS) serving as the primary criterion. A systematic parametric study is conducted to explore the influence of different factors on slope stability such as slope geometry, shear strength parameters of soil and location of water table. Furthermore, Regression and Machine Learning (ML) techniques are applied over the generated dataset to establish correlations between FoS and the influencing factors, thereby aiding in the formulation of predictive relationships for assessing stability of slopes. The relationships achieved through this study exhibit a high level of accuracy, with regression and ML-based algorithms yielding correlation coefficients of 96.43% and 98.03% for static case, 95.81% and 98.15% for earthquake-induced loading case, respectively. The established correlations can provide an easy tool for practicing engineers to judge the slope stability with least effort, thereby facilitating quick first-hand predictions of risk assessments. This integrated approach combining traditional engineering methods with advanced data-driven techniques paves the way to offer a robust framework for analyzing landslide hazards and enhancing disaster mitigation efforts in the region.