Machine Learning in the Stochastic Analysis of Slope Stability - A State-of-the-Art Review

Aug 3, 2026

Upload by Asher

Traditional slope stability analysis often treats soil and rock properties as uniform or conservative averages across the entire region, which can increase cost and still miss real spatial variability. Stochastic analysis incorporates uncertainty and variability to support more economical, reliability-focused assessment. Over the past decades, machine learning has expanded rapidly and is widely used as surrogate modeling to improve computational efficiency in stochastic slope stability studies. This review synthesizes 159 supervised-learning studies from the last 20 years, summarizing advances in safety factor prediction and slope stability classification, and outlines four research challenges and directions for future work.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages28

Reading Time1 h 11 min

AccessFree to download

Introduction

Deterministic and empirical slope stability analysis

Sources of uncertainty and spatial variability

Stochastic slope stability and probabilistic approaches

Random field theory and reliability metrics

Monte Carlo simulation and probabilistic back analysis

Machine learning in stochastic slope stability

Surrogate modeling for computational efficiency

Safety factor prediction and classification

Review Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review Haoding Xu , Xuzhen He *, Feng Shan, Gang Niu and Daichao Sheng Citation: Xu, H.; He, X.; Shan, F.; Niu, G.; Sheng, D. Machine Learning in the Stochastic Analysis of Slope Stability: A State-of-the-Art Review. Modelling 2023, 4, 426–453 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)modelling4040025 Academic Editor: Luca Lenti and Salvatore Martino Received: 14 August 2023 Revised: 26 September 2023 Accepted: 29 September 2023 Published: 1 October 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . School of Civil and Environmental Engineering, University of Technology Sydney, Ultimo, NSW 2007, Australia; [haoding.xu@student.uts.edu.au](haoding.xu@student.uts.edu.au) (H.X.); [feng.shan@student.uts.edu.au](feng.shan@student.uts.edu.au) (F.S.); [gang.niu@student.uts.edu.au](gang.niu@student.uts.edu.au) (G.N.); [daichao.sheng@uts.edu.au](daichao.sheng@uts.edu.au) (D.S.) * Correspondence: [xuzhen.he@uts.edu.au](xuzhen.he@uts.edu.au) Abstract: In traditional slope stability analysis, it is assumed that some “average” or appropriately“conservative” properties operate over the entire region of interest. This kind of deterministic conservative analysis often results in higher costs, and thus, a stochastic analysis considering uncertainty and spatial variability was developed to reduce costs. In the past few decades, machine learning has been greatly developed and extensively used in stochastic slope stability analysis, particularly used as surrogate models to improve computational efﬁciency. To better summarize the current application of machine learning and future research, this paper reviews 159 studies of supervised learning published in the past 20 years. The achievements of machine learning methods are summarized from two aspects—safety factor prediction and slope stability classiﬁcation. Four potential research challengesand suggestions are also given. Keywords: slope stability; factor of safety; slope stability classiﬁcation; machine learning; geotechnical engineering; uncertainty; reliability analysis 1. Introduction Landslides are sudden and serious disasters that can cause signiﬁcant damage to nearby facilities, resulting in economic and casualty losses. Therefore, the evaluation of slope stability is a critical prerequisite for disaster prevention and mitigation. In numerical simulations, slope stability is presented as a factor of safety (FOS), which is obtained using deterministic analyses such as limit analysis [1], ﬁnite element limit analysis [2], displacement-based ﬁnite element analysis combined with the strength reduction method [3,4], or ﬁnite element analysis with the gravity increasing method [5,6] . In site investigation and analysis, slope stability is typically classiﬁed using empirical formulas or expert judgments. Slope stability analysis concerns earth materials such as soil, rock, and other materials. Soil is one of the engineering materials that has the most complex physical, mechanical, and chemical behaviors and is made of three phases. Due to geological action and stress history, soils exhibit complex spatial variability and anisotropy, which makes studying soilsand predicting their behavior difﬁcult. In geotechnical engineering, two main approaches have traditionally been used to study the mechanical behavior of geotechnical materials: (1) empirical methods, such as laboratory tests and site investigations, and (2) numerical and analytical methods. The cost of experiments and ﬁeld tests tends to rise with the number of tests con

FAQs

Why is deterministic slope stability analysis often limited?

Deterministic approaches typically assume average or conservative properties over the full region, which can lead to higher cost and may overlook the inherent spatial variability of real soils and rock.

How does stochastic analysis improve slope stability evaluation?

Stochastic analysis explicitly accounts for uncertainty and spatial variability, enabling probabilistic evaluation such as failure probability or reliability indices to quantify safety margins.

What roles does machine learning play in stochastic slope stability studies?

Machine learning is largely used as surrogate models to improve computational efficiency, with review coverage focusing on two main outcomes: safety factor prediction and slope stability classiﬁcation.

Aug 3, 2026

Upload by Asher