Discussion on “Refined Approaches for Open Stope Stability Analysis in Mining Environments: Hybrid SVM Model with Multi optimization Strategies and GP Technique” Rock Mech Rock Eng (2024)

Research output: Contribution to journalComment/debatepeer-review

Abstract

In open stope mining, it is crucial to design the stope size and dimensions appropriately to optimize ore extraction. Inadequate design can negatively impact the stability of the stope walls, thereby jeopardizing the safety and profitability of the operation, particularly in conditions of increasing mining depth. The Mathew’s stability graph, an empirical design method, has been widely used for open stope design due to its simplicity and practicality at mine-sites. Many mines around the world have adopted it as a conventional design approach. However, this method is limited by its approximate and empirical nature. Over the past decade, the importance of the topic has motivated the employment of machine learning (ML) technologies to address these limitations and enhance the empirical design methods. However, it is crucial to remember that, despite their immense usefulness, machine learning technologies necessitate the application of sound engineering judgment in data analysis to appropriately choose the algorithms’ training data, thereby preventing misleading outcomes. In this paper, a discussion on a recently published paper titled “Refined Approaches for Open Stope Stability Analysis in Mining Environments: Hybrid SVM Model with Multi-Optimization Strategies and GP Technique,” is presented, with the purpose of drawing the authors and readers attention to a few aspects that might be worth further analysis or dialogue in connection with open stope design. Most importantly, it appeared that some of the stope performance data used to train and test the proposed models did not pertain to open mine stope operations, potentially invalidating the design principle of the stability graph method and necessitating further clarifications. The different aspects of the discussions presented in the paper could lead to further research and an open dialog about important practical mine design issues.

Original languageEnglish
JournalRock Mechanics and Rock Engineering
DOIs
Publication statusAccepted/In press - 2025

Keywords

  • Machine learning
  • Open stope design
  • Stability graph

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Geotechnical Engineering and Engineering Geology
  • Geology

Fingerprint

Dive into the research topics of 'Discussion on “Refined Approaches for Open Stope Stability Analysis in Mining Environments: Hybrid SVM Model with Multi optimization Strategies and GP Technique” Rock Mech Rock Eng (2024)'. Together they form a unique fingerprint.

Cite this