A Sustainable and Practical Machine Learning Approach Using Scikit-Learn for Predicting Stope Instability: Identification of Critical Geotechnical Factors.
Saved in:
| Title: | A Sustainable and Practical Machine Learning Approach Using Scikit-Learn for Predicting Stope Instability: Identification of Critical Geotechnical Factors. |
|---|---|
| Alternate Title: | Održiv i praktičan pristup strojnom učenju pomoću Scikit-Learn baze za procjenu nestabilnosti čela radilišta: identifikacija ključnih geotehničkih čimbenika. |
| Authors: | Bemo, Amos1, Shonuga, Deji Olatunji2, Zvarivadza, Tawanda3, Onifade, Moshood4, Khandelwal, Manoj4 m.khandelwal@federation.edu.au |
| Source: | Rudarsko-Geološko-Naftni Zbornik. 2025, Vol. 40 Issue 5, p179-198. 20p. |
| Subject Terms: | *Geotechnical engineering, *Rock mechanics, *Machine learning, *Prediction models, *Mines & mineral resources |
| Abstract (English): | Stope instability remains a persistent and hazardous challenge in underground mining, impacting safety, efficiency, and sustainability. Traditional stability assessment methods, while valuable, are often limited by site-specific calibration, simplifications, and adaptability issues in dynamic underground conditions. While machine learning shows potential for improved accuracy, a critical gap persists in understanding how geotechnical factors interact in practice. This study introduces a novel, practical machine learning framework (Scikit-Learn) to predict stope instability, and crucially, to quantify the nuanced, non-linear influence and interaction of critical geotechnical factors in a shallow gold mine. Comprehensive geotechnical investigation (observations, lab tests, rock mass classifications, blast damage assessments) and advanced data analysis (Random Forest feature importance, RFE, decision boundary analysis) identified water ingress, blast-induced damage, and rock mass quality (RMR) as the most significant instability factors. Water ingress profoundly impacted stability, with moderate blast damage exacerbating instability under high water ingress. Rock strength showed comparatively lower significance. The developed model achieved robust predictive performance (accuracy: 0.83, precision: 0.88, recall: 0.83, F1-score: 0.83). Based on these insights, tailored support patterns (e.g. 22mm/16mm cone bolts, timber props) are proposed to mitigate specific risks. This research significantly advances targeted rock mechanics solutions by providing a deeper, quantifiable understanding of complex instability mechanisms, enhancing mine safety and operational efficiency in shallow gold mining. [ABSTRACT FROM AUTHOR] |
| Abstract (Croatian): | Nestabilnost čela radilišta i dalje je stalan i opasan izazov u podzemnoj eksploataciji, s negativnim utjecajem na sigurnost, učinkovitost i održivost. Tradicionalne metode procjene stabilnosti, iako korisne, često su ograničene potrebom za baždarenjem na specifičnome lokalitetu, pojednostavnjivanjem i slabom prilagodljivošću u dinamičnim podzemnim uvjetima. Iako strojno učenje pokazuje potencijal za poboljšanu točnost, i dalje postoji velik nedostatak razumijevanja kako geotehnički čimbenici međusobno djeluju u praksi. Ovo istraživanje predstavlja novu, praktičnu okosnicu strojnoga učenja (Scikit-Learn) za procjenjivanje nestabilnosti čela radilišta te ključnoga kvantificiranja suptilnoga nelinearnog utjecaja i međudjelovanja ključnih geotehničkih čimbenika u plitkome rudniku zlata. Sveobuhvatno geotehničko ispitivanje (opažanja, laboratorijska ispitivanja, klasifikacije stijenskih masa, procjene oštećenja uslijed miniranja) i napredna analiza podataka (Random Forest, Recursive Feature Elimination, Decision boundary analysis) identificirali su najvažnije čimbenike nestabilnosti: prodor vode, oštećenja uzrokovana miniranjem i kvalitetu stijenske mase (RMR). Prodor vode imao je velik utjecaj na stabilnost, pri čemu su umjerena oštećenja od miniranja dodatno pogoršavala nestabilnost u uvjetima visokoga prodora vode. Čvrstoća stijenske mase pokazala se relativno manje važnom. Razvijeni model postigao je snažnu prediktivnu učinkovitost (točnost: 0,83, preciznost: 0,88, odziv: 0,83, F1 mjera: 0,83). Na temelju tih znanja predloženi su prilagođeni oblici podgrade (npr. konusna sidra 22 mm / 16 mm, drvena podgrada) kako bi se ublažili specifični rizici. Ovo istraživanje znatno doprinosi ciljanom rješavanju problema u mehanici stijena pružajući dublje, kvantificirano razumijevanje složenih mehanizama nestabilnosti, čime se poboljšava sigurnost i operativna učinkovitost kod plitke eksploatacije zlata. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
| FullText | Links: – Type: pdflink Text: Availability: 0 |
|---|---|
| Header | DbId: enr DbLabel: Energy & Power Source An: 190565091 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: A Sustainable and Practical Machine Learning Approach Using Scikit-Learn for Predicting Stope Instability: Identification of Critical Geotechnical Factors. – Name: TitleAlt Label: Alternate Title Group: TiAlt Data: Održiv i praktičan pristup strojnom učenju pomoću Scikit-Learn baze za procjenu nestabilnosti čela radilišta: identifikacija ključnih geotehničkih čimbenika. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Bemo%2C+Amos%22">Bemo, Amos</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Shonuga%2C+Deji+Olatunji%22">Shonuga, Deji Olatunji</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Zvarivadza%2C+Tawanda%22">Zvarivadza, Tawanda</searchLink><relatesTo>3</relatesTo><br /><searchLink fieldCode="AR" term="%22Onifade%2C+Moshood%22">Onifade, Moshood</searchLink><relatesTo>4</relatesTo><br /><searchLink fieldCode="AR" term="%22Khandelwal%2C+Manoj%22">Khandelwal, Manoj</searchLink><relatesTo>4</relatesTo><i> m.khandelwal@federation.edu.au</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Rudarsko-Geološko-Naftni+Zbornik%22">Rudarsko-Geološko-Naftni Zbornik</searchLink>. 2025, Vol. 40 Issue 5, p179-198. 20p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Geotechnical+engineering%22">Geotechnical engineering</searchLink><br />*<searchLink fieldCode="DE" term="%22Rock+mechanics%22">Rock mechanics</searchLink><br />*<searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br />*<searchLink fieldCode="DE" term="%22Prediction+models%22">Prediction models</searchLink><br />*<searchLink fieldCode="DE" term="%22Mines+%26+mineral+resources%22">Mines & mineral resources</searchLink> – Name: Abstract Label: Abstract (English) Group: Ab Data: Stope instability remains a persistent and hazardous challenge in underground mining, impacting safety, efficiency, and sustainability. Traditional stability assessment methods, while valuable, are often limited by site-specific calibration, simplifications, and adaptability issues in dynamic underground conditions. While machine learning shows potential for improved accuracy, a critical gap persists in understanding how geotechnical factors interact in practice. This study introduces a novel, practical machine learning framework (Scikit-Learn) to predict stope instability, and crucially, to quantify the nuanced, non-linear influence and interaction of critical geotechnical factors in a shallow gold mine. Comprehensive geotechnical investigation (observations, lab tests, rock mass classifications, blast damage assessments) and advanced data analysis (Random Forest feature importance, RFE, decision boundary analysis) identified water ingress, blast-induced damage, and rock mass quality (RMR) as the most significant instability factors. Water ingress profoundly impacted stability, with moderate blast damage exacerbating instability under high water ingress. Rock strength showed comparatively lower significance. The developed model achieved robust predictive performance (accuracy: 0.83, precision: 0.88, recall: 0.83, F1-score: 0.83). Based on these insights, tailored support patterns (e.g. 22mm/16mm cone bolts, timber props) are proposed to mitigate specific risks. This research significantly advances targeted rock mechanics solutions by providing a deeper, quantifiable understanding of complex instability mechanisms, enhancing mine safety and operational efficiency in shallow gold mining. [ABSTRACT FROM AUTHOR] – Name: Abstract Label: Abstract (Croatian) Group: Ab Data: Nestabilnost čela radilišta i dalje je stalan i opasan izazov u podzemnoj eksploataciji, s negativnim utjecajem na sigurnost, učinkovitost i održivost. Tradicionalne metode procjene stabilnosti, iako korisne, često su ograničene potrebom za baždarenjem na specifičnome lokalitetu, pojednostavnjivanjem i slabom prilagodljivošću u dinamičnim podzemnim uvjetima. Iako strojno učenje pokazuje potencijal za poboljšanu točnost, i dalje postoji velik nedostatak razumijevanja kako geotehnički čimbenici međusobno djeluju u praksi. Ovo istraživanje predstavlja novu, praktičnu okosnicu strojnoga učenja (Scikit-Learn) za procjenjivanje nestabilnosti čela radilišta te ključnoga kvantificiranja suptilnoga nelinearnog utjecaja i međudjelovanja ključnih geotehničkih čimbenika u plitkome rudniku zlata. Sveobuhvatno geotehničko ispitivanje (opažanja, laboratorijska ispitivanja, klasifikacije stijenskih masa, procjene oštećenja uslijed miniranja) i napredna analiza podataka (Random Forest, Recursive Feature Elimination, Decision boundary analysis) identificirali su najvažnije čimbenike nestabilnosti: prodor vode, oštećenja uzrokovana miniranjem i kvalitetu stijenske mase (RMR). Prodor vode imao je velik utjecaj na stabilnost, pri čemu su umjerena oštećenja od miniranja dodatno pogoršavala nestabilnost u uvjetima visokoga prodora vode. Čvrstoća stijenske mase pokazala se relativno manje važnom. Razvijeni model postigao je snažnu prediktivnu učinkovitost (točnost: 0,83, preciznost: 0,88, odziv: 0,83, F1 mjera: 0,83). Na temelju tih znanja predloženi su prilagođeni oblici podgrade (npr. konusna sidra 22 mm / 16 mm, drvena podgrada) kako bi se ublažili specifični rizici. Ovo istraživanje znatno doprinosi ciljanom rješavanju problema u mehanici stijena pružajući dublje, kvantificirano razumijevanje složenih mehanizama nestabilnosti, čime se poboljšava sigurnost i operativna učinkovitost kod plitke eksploatacije zlata. [ABSTRACT FROM AUTHOR] |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=190565091 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.17794/rgn.2025.5.14 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 20 StartPage: 179 Subjects: – SubjectFull: Geotechnical engineering Type: general – SubjectFull: Rock mechanics Type: general – SubjectFull: Machine learning Type: general – SubjectFull: Prediction models Type: general – SubjectFull: Mines & mineral resources Type: general Titles: – TitleFull: A Sustainable and Practical Machine Learning Approach Using Scikit-Learn for Predicting Stope Instability: Identification of Critical Geotechnical Factors. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Bemo, Amos – PersonEntity: Name: NameFull: Shonuga, Deji Olatunji – PersonEntity: Name: NameFull: Zvarivadza, Tawanda – PersonEntity: Name: NameFull: Onifade, Moshood – PersonEntity: Name: NameFull: Khandelwal, Manoj IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 11 Text: 2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 03534529 Numbering: – Type: volume Value: 40 – Type: issue Value: 5 Titles: – TitleFull: Rudarsko-Geološko-Naftni Zbornik Type: main |
| ResultId | 1 |
