Risk assessment of open pit slope in cold regions based on Bayesian network

Many large open pits are operated in cold regions, where the effect of freeze-thaw action may increase landslide risk. To improve the accuracy of risk assessment for the open pit slope in cold regions, this paper proposes a risk assessment method of rock slope based on Bayesian network (BN). By comb...

Full description

Saved in:
Bibliographic Details
Main Authors: Hui Li, Xiaofei Han, Wancheng Zhu, Xige Liu, Leilei Niu
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Geomatics, Natural Hazards & Risk
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19475705.2025.2529949
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many large open pits are operated in cold regions, where the effect of freeze-thaw action may increase landslide risk. To improve the accuracy of risk assessment for the open pit slope in cold regions, this paper proposes a risk assessment method of rock slope based on Bayesian network (BN). By combining the mechanism knowledge and monitoring information, the risk assessment framework on rock slope based on BN is established. Firstly, the key variables of rock slope stability are determined. Furthermore, the relationships among the key variables are quantified to determine the structure and parameters of risk assessment model. Then, the rock slope risk assessment result can be obtained by BN reasoning to provide real-time risk evaluation. Finally, taking a copper mine in Tibet, China as an example, and the effectiveness of risk assessment model is validated against the assessment results for different scenarios. Sensitivity analysis and key risk factors analysis are implemented to determine the roles of different influence factors. As the addition of multi-source monitoring information, the proposed method can obtain more objective rock slope risk assessment results and avoid the possible bias compared with the single monitoring information source.
ISSN:1947-5705
1947-5713