Analysis of Slope Stability Based on Four Machine Learning Models - An Example of 188 Slopes

Aug 3, 2026

Upload by Skyler

Rapid, accurate slope-stability prediction supports timely decisions in geotechnical engineering. The study presents an intelligent assessment approach based on machine learning to improve the precision of Factor of Safety (FOS) evaluations. Using 188 slope cases from domestic and international sources, six key feature variables were identified for FOS assessment. Data were trained and tested via 5-fold cross-validation. GBM, SVM, XGB, and RF were compared with MAE, MSE, RMSE, and R², then validated against a GeoStudio finite element model in engineering examples.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Abstract

1 Introduction

[https://doi.org/10.3311/PPci.37630](https://doi.org/10.3311/PPci.37630) Creative Commons Attribution b | 1 Periodica Polytechnica Civil Engineering Analysis of Slope Stability Based on Four Machine Learning Models An Example of 188 Slopes Menghan Zhang1, Jin Wei1* 1 School of Highway, Chang'an University, Xi'an 710064, China * Corresponding author, [e-mail: weijin@chd.edu.cn](e-mail: weijin@chd.edu.cn) Received: 08 June 2024, Accepted: 13 January 2025, Published online: 06 February 2025 Abstract To achieve rapid and precise prediction of slope stability, we propose an intelligent assessment method utilizing machine learning techniques. This approach aims to enhance the precision of slope stability evaluations, facilitating more effective and timely decisionmaking in geotechnical engineering. By analyzing 188 slope cases from domestic and international sources, we have identified six key feature variables to evaluate the Factor of Safety (FOS) for slope stability assessment. The dataset was established for evaluating slope stability, and to ensure robustness, it was divided into training and testing set using a 5-fold cross-validation approach. Four slope stability prediction models-GBM, SVM, XGB, and RF-were developed using machine learning algorithms. The accuracy of the models in predicting FOS for slopes was assessed using metrics such as MAE, MSE, RMSE, and R2. The best-performing machine learning model, along with the finite element model developed using GeoStudio, was applied to engineering examples to compare their feasibility and efficiency. The research findings demonstrates that the GBM model has a minimal error between the predicted and actual slope FOS, highlighting its high accuracy. The model shows a strong correlation between predicted and actual FOS, indicating its superior performance relative to other models. GBM model and the finite element model align well with the actual field conditions. However, the GBM model stands out due to its higher accuracy and faster computational efficiency. Therefore, the GBM model offers a high degree of fit between the predicted FOS and the actual values, making it well-suited for evaluating slope stability. Keywords factor of safety, machine learning, slope stability, finite element model 1 Introduction Natural disasters occur frequently worldwide, and China is no exception. Due to its vast and geologically complex geology, China is particularly susceptible to such calamities. In 2022 alone, there were a staggering total of 5659 recorded geological disasters in the country, with 3919 instances of slope instability disasters. Consequently, several researchers [1, 2] have taken up studies on these cases of instability. The consequences of slope instability accidents can be devastating, causing significant harm to lives, properties and critical infrastructure. Notable examples include the Cher Tara Open Coal incident, the Chana Landslide [3], the Aniangzhai Landslide [4], the slope in Fa'er Town [5], the Baiyun Slide Complex [6] as well as various slopes in Fengjie County. Given this context, the evaluation of slope stability emerges as an immensely crucial research endeavor. In recent decades, FOS is a parameter used to evaluate slope stability. Usually, the factors affecting slope safety are taken as the evaluation factors of slope safety factor. In recent decades, FOS has been widely used as a parameter to evaluate slope stability. Geometric parameters such as H and β, as well as strength parameters including internal ϕ and c, play crucial roles in slope stability [7, 8]. Scholars have extensively researched and analyzed slope stability [9–12]. Various methods for determining the FOS include theoretical analysis, simulation tests, and numerical simulations [13]. The traditional limit equilibrium method, a type of theoretical analysis, assumes a predetermined critical slip surface and calculates resistance based on equilibrium equations [14]. How

FAQs

What is the main goal of the proposed method for slope stability assessment?

To achieve rapid and precise prediction of slope stability by improving the accuracy of Factor of Safety (FOS) evaluations using machine learning.

How many slope cases are used, and what role do feature variables play?

188 slope cases are used, and six key feature variables are identified to evaluate the Factor of Safety for slope stability assessment.

Which machine learning models are compared and how is performance evaluated?

GBM, SVM, XGB, and RF are developed and evaluated using MAE, MSE, RMSE, and R², with additional comparison against a GeoStudio finite element model in engineering examples.

Aug 3, 2026

Upload by Skyler