South Gobi Underground Mass Mining InstituteRisk Management – Major Cave Hazards and the Application of Machine Learning to Id Hazards
SGUMMI COURSECourse Summary
The course provides an overview of essential concepts in geotechnical risk management in cave mining as well as a practical framework for quantitative analysis of risk that can be applied on site. It starts with a review of cave mine risk events and their causes, risk controls and management, before exploring statistical analysis of geotechnical risk. The pros and cons of knowledge-driven and data-driven hazard estimation will be explored. Methods of statistical inference and hazard probability analysis will be presented, culminating in a practical workflow for site application of machine learning (ML) methods.
COURSE OVERVIEWCourse Objectives
01
Review risk events, causes, and controls in cave mining.
02
Provide an understanding of the basic concepts of statistical inference and machine learning (ML) in a geotechnical context.
03
Through case studies, build an understanding of the machine learning workflow for estimation of geotechnical hazard parameters through back-analysis.
04
Provide a framework for operational site applications.
TARGET AUDIENCEThis course is ideal for
Early to mid-career geotechnical or geological engineers in underground mining
Mine planning and mine design professionals seeking stronger risk management and hazard identification skills.
Technical teams supporting mining operations.
Professionals working with geo data, geo models and their application in mine design.
OUTCOMESKey Learning Outcomes
By the end of the course, participants will be able to:
- After completing the course, attendees will understand statistical, probabilistic, and machine learning approaches to geotechnical hazard analysis. They will feel comfortable in discussions of statistical and machine learning methods in a geotechnical context and their relationship to standard knowledge-driven methods. They will be able to evaluate the possibility of usefully deploying data-driven methods in operational applications and be able to undertake simple machine learning for hazard analysis projects on their own.
COURSECourse Topics Covered
01
Principles and practices of risk management in cave mining
02
Knowledge and data-driven analysis
03
Statistical inference and machine learning for geotechnical hazard
04
Analysis workflow through a cave mining case study
05
Site application
FORMATDelivery Format
Duration:
5 days
Location:
Oyu Tolgoi Mine, Mongolia
Format:
In-person, expert-led instruction
Structure:
Approximately 70% theoretical content (including case studies), 30 % hands-on practical activities
Language:
English
COURSE INSTRUCTORDr. John McGaughey
Dr John McGaughey is the founder and president of Mira Geoscience. He has extensive mining industry experience, focusing on quantitative, multi-disciplinary 3D and 4D analysis for mineral exploration and geotechnical decision support. Prior to founding Mira Geoscience, John spent 10 years in the Noranda Technology Centre as a senior scientist in their rock mechanics group. He obtained an MSc in geological engineering and a PhD in geophysics from Queen’s University in Canada.
COURSE FAQFrequently Asked Questions
01/
Who should attend this course?
This course is intended for geotechnical engineers, mine planners, operations managers, and technical staff involved in underground mass mining. It is especially relevant for those responsible for hazard identification, monitoring, and implementing risk management frameworks in caving environments.
02/
What practical skills will I gain?
Participants will learn how to identify and assess major cave hazards, apply monitoring data to risk frameworks, and implement Trigger Action Response Plans (TARPs). The course also introduces the use of machine learning for geohazard detection and prediction, giving participants tools to strengthen operational safety and decision-making.
03/
How is the course delivered?
The course is delivered face-to-face at the Oyu Tolgoi mine in Mongolia. It combines classroom-based theory with case studies, practical exercises, and field-based risk assessment activities to reinforce real-world application.
04/
Do I need prior experience with machine learning or advanced risk tools?
No prior knowledge of machine learning is required. The course is structured to introduce the concepts in a mining context, while focusing on practical application for geotechnical and operational risk management. A basic understanding of geotechnical or mining engineering is recommended.
TESTIMONIALSWhat Our Industry Partners Are Saying
I’ve always been passionate about mining, and SGUMMI offers a unique opportunity to learn directly in the field. The program uses the Oyu Tolgoi mine itself as a training hub, which makes the learning incredibly practical and hands-on. It’s an ideal environment to build real-world skills.
Zaya Baitag
I truly enjoyed the SGUMMI course, which provided valuable insights into geotechnical principles and caving knowledge. The lessons were very practical and went beyond the math to explore the reason behind the concept. We are gaining experience-based knowledge from world-leading experts and professors. I’m very much looking forward to the remaining part of the course.
Batpurev Ganbat
The SGUMMI training is helping me gain a more systematic understanding of underground mass mining. It bridges the gap between theoretical learning and practical applications, and I’m already applying what I learn directly to my daily work. This is supporting my development toward the next step in my career.
Bayar-Erdene Ganbold
SGUMMI has been an amazing chance to learn from top mining experts. The sessions are well-connected and practical, helping me understand cave mining and geomechanics better. Now I feel more confident to grow professionally in this field.
Suvdaa Erdenebayar
This training plays a vital role in strengthening the education sector at a time when near-surface resources are being depleted and the demand for deep mining is increasing. As a young university lecturer, it is a valuable opportunity for me to share with mining company engineers and students what I’ve learned from professors. I am especially grateful for the chance to learn not only from internationally renowned professors but also from world-class practitioners in the mining industry.
Orkhon Erdenebaatar
Course Delivery Months
Planned Delivery:
August/September 2026
Exact dates will be confirmed closer to delivery.