The following is an article by Robyn Teet & Jack Mowen, Resolve Mining Solutions, Australia which appeared in the Deep Mining 2024 Supplement.
Utilising ‘smart data’ to combat long-term geotechnical risk exposure and support planning for success
By Robyn Teet & Jack Mowen, Resolve Mining Solutions, Australia
As mining progresses deeper and typically favours mass mining methods with sublevel caves, block caves, and recently mega caves, the geotechnical risk to development and major excavations increases. The reliance on long-term infrastructure due to the nature of the mining method also introduces additional time-dependant hazards that are typically a minor concern in stoping operations and historically smaller-scale mass mining methods.
As the length of time that underground workings remain open increases, the geotechnical implications on ground support lifespan as a function of corrosivity and capacity reduction through deformation, stress loading, and cyclic stress loading increases. This means that the importance of collecting and understanding ground support data becomes more critical for the long-term stability of the excavations involved as well as the operational budget. With more workings open over a long period of time, and more workings required to reach the depths and rates required for modern mass mining methods, the volume of data that can be collected is substantial and a step change from previous approaches. This not only results in data collection, storage and logistics issues, but also issues with managing the resultant ‘big data’ interpretation efficiently.
This article aims to highlight how existing software and classifications can support the long-term success of modern caving operations and how the data collection can be introduced into standard site operational practices. To demonstrate this, three aspects of geotechnical risk prevalent within mass mining methods have been showcased in a model block cave within the mXrap software package and simplistic supporting spreadsheet model. These three aspects are:
- clay prediction across the life of the cave
- inrush event triggers
- corrosion and its impact on life of mine support capacity.
A brief overview of each of these aspects is provided in this article and whilst not an exhaustive discussion, the author encourages the reader to consider similarities, differences and juxtapose this work against their own operations and experiences as it is the process where value is added to an operation. The authors also acknowledge that personnel resourcing is often a key limiting factor to the collection, analysis and interpretation of data. Efforts are made to present approaches that do not specifically require geotechnical engineers for data collection, in an attempt to alleviate resourcing challenges that are prevalent across the industry at present.
Figure 1 (a) PGCA model for Ernest Henry sublevel cave mine (Power & Campbell 2016);
(b) Footprint sequencing in PCBC (Villa 2021)
Clay
The prediction of clay and fines migrating through a cave column is fundamental to understanding impacts on production and material handling systems, along with exposure to inrush and inundation risk either by wet or dry fines events. The use of tools such as PGCA and PCBC has allowed engineers to model particle flow and comminution using various mechanisms with a view to provide inputs into the planning and scheduling space and optimising the resource recovery.
Whilst the software packages do have graphical outputs (Figure 1), they typically require specific user knowledge and training for their generation. The original intent was to provide production forecasting capabilities, and different software packages and modules can be more or less reliable at predicting fines movement through the ore column. In addition, they are quite often restricted by block model considerations and smoothing across relatively large block sizes.
However, as the packages can export the data into a transferable format, there are other available programs that can be used to display the clay and fines prediction on a drawpoint-by-drawpoint basis. Not only can these predicted fines and clay forecast be displayed (as shown in mXrap in Figure 2), but real-world observations of fines and clay, either by visual observation, photogrammetric techniques or automated scanning techniques, can also be incorporated and comparison between the two made. This facilitates the calibration of models and feeds into inrush risk management discussed later in this article.
By using software that does not require specialised training to access basic functionality, integration of reporting of modelling and observations into operations becomes much more streamlined. In some instances, training is not needed at all, as the required data can be displayed on a window specifically set up for stakeholders, such as the mine control room or managers, with the ability to interact with the program limited to rotating data in 3D. This gives the individual users the functionality for viewing data with standard mouse controls, instead of needing specialist training on software packages, or being restricted to viewing 2D images in reports and emails.
Figure 2 Visualisation of predicted PCBC clay and fines from (a) month 1 and (b) month 12. Note the
changes in the scale on the right of each image
Inrush
Inrush and inundation are defined in Australian mining legislation as a principal mining hazard (or equivalent). Caving operations around the world in particular understand that inrush hazards require active management on a daily basis to manage the risk. Not only does it pose a significant risk to personnel through injuries and fatalities, but inrush also threatens production and assets through equipment damage, material handling difficulties, production delays and unplanned dilution (ICARN n.d.). At an operational level, having ready access to data required to work with trigger action response plans (TARPs) and additional controls to reduce the exposure and likelihood of a significant loss is key to managing inrush and inundation risk.
There are limited published examples from industry of integrated monitoring and risk management tools with PT Freeport Indonesia publishing their customised approach in Caving 2022 (Llewelyn et al. 2022) being the most recent example of integrated risk management at a caving operation. However, the development of a custom software solution is not necessarily accessible to all mine operators and more traditional approaches are often adopted. This can include manual data collection, mapping, photos, and fragmentation assessments among other data points. This information can either be collected on paper, directly into Excel or a digital database or via specialised collection tools (e.g. photogrammetry fragmentation tablets). Manual collection on paper results in data entry time, and even the specialised data collection tools typically require post processing, which is also time consuming.
Given the resourcing shortage that is present across the mining industry, combining these multiple data sources that inform inrush risk (fines percentage, moisture status, drawpoint status, existing control implementation, etc.) can be impractical involving multiple people over a significant amount of time. This makes real-time (or as near real-time as possible) access and application of triggers almost impossible impeding daily risk management and places operations and their personnel in potentially hazardous situations. The presentation of this data in an accessible format is also a significant inhibiting factor to clear risk communication with both management and the operational teams including definition of appropriate actions. In addition, the goals of these teams are often disparate with the technical teams as the generators of the data representation are driven by technical accuracy, whereas operational teams and management prefer brevity, clarity and actionable information. In all cases, however, clear and interpretable data needs to be available in as short a time frame as possible and produced with minimal intervention to be able to trigger immediate action or elevated monitoring, typically in TARP format, if necessary.
Utilising the existing mXrap software product suite, with some minor modifications, the aforemented data inputs for determining inrush and inundation likelihood can be automatically converted into a risk rating, generating a simple 2D plan of drawpoints, colour coded by risk and subsequent TARP level (Figure 2).
The example displayed in Figure 3 utilises the fragmentation data from the existing mXrap caving app, with an additional interface to import and code drawpoint observations – where possible, data that is electronically entered could be immediately ingested by the program, but otherwise this would require a data entry step of these observations. Once imported, mXrap can utilise the individual site TARP and then calculate and display the inherent risk for each individual drawpoint. By generating a simple and relatable graphic, that can either be printed, or displayed live on screens in shift change areas (or control rooms, crib rooms, etc. depending on the site set-up), clear communication can be achieved. Whilst this approach does not resolve inefficiencies in the data-collection methods, it does allow for simple, time-effective translation of data into information and support risk management on site.
Corrosion
Stepping back from the day-to-day management of operational risk on production levels, a typically medium-term to long-term hazard within underground operations, regardless of mining method, is ground support corrosion. The importance of corrosion as both geotechnical risk and financial planning of an operation has been well documented in literature including Hassel (2008), Hadjigeorgiou & Dorion (2013), Preston et al. (2019) and NIOSH (2020); however, the understanding of how corrosion is predicted to impact an operation and the subsequent cost implications continue to be poorly implemented in the mining environment.
Mines with an extended operating life that routinely monitor and collect data on corrosion would typically experience the same challenges with data collection (resourcing), data storage and management, and visualisation of data. This makes the integration of monitoring data into ‘smart data’ systems essential to operational success. This is especially relevant in mines prone to expedited corrosion rates due to groundwater conditions. Also, for mines operating in a marginal economic environment, an ineffective understanding of corrosion at the mine can result in a transition to sub-economic or unviable operations. This in turn can drive behaviours of limited materials and time available to complete rehabilitation and expose personnel to increased geotechnical risk.
Having a time-effective and cost-effective way of collecting, analysing and visualising key information pertaining to the corrosion status of operational areas underground not only facilitates risk management in the ground control space, but also provides realistic inputs into financial planning both in the short-term and medium-term. Moreover, data collected from one level or mine area may be able to be used as a predictive tool for future operations, especially where anticipated conditions are similar.
Utilising the corrosion classification system outlined by Dorion & Hadjigeorgiou (2013) and mapping data collected at Mine A on the main decline (Figure 4), a cost model and implementation schedule was constructed utilising 3D CAD software and a basic spreadsheet as a proof of concept, to determine likely rehabilitation requirements as a result of capacity loss (Table 1).
In the demonstrated example overleaf, the simple visualisation and material cost estimates facilitated discussions at the senior management level not only around the rehabilitation requirements but also the financial implications of mining deeper and requiring an extended decline life span and, therefore, increased rehabilitation costs. With repeated data collection, the time dependency of corrosion specific to the site can be determined and the Dorion & Hadjigeorgiou (2013) classification customised, further supporting risk management.
The vision with this approach is that it also becomes integrated into mXrap to provide additional functionality and visibility with accessible graphics, similar to the other tools presented earlier in this article. Once this is integrated, the mapping status can be viewed live as soon as the data is processed, and an automatically generated table can provide a running cost estimate for corrosion rehabilitation requirements. This project is currently in progress and aims to integrate site-specific ground support costs, corrosion mapping data and testing data to serve as an efficient data storage location, analysis tool and visualisation and communication tool, all in one, and within a cost-effective platform.
Conclusion
What does all this mean for operations? Ultimately, the solution to your data visualisation and risk management challenges could already be in use on site and budgeted for. Whilst there is a push to innovate, this often overlooks repurposing solutions that could be more feasible in terms of budget, time to implementation and ongoing upkeep and utilisation. By understanding where the bottlenecks are with existing procedures and systems in relation to managing risk, there may be opportunities to innovate with what you already have, for a significantly more cost-effective solution, and leverage data that is already available.
By generating outputs that are easily understandable across the organisation, the potential for miscommunication and misinterpretation of data is significantly reduced, and the overall risk management process related to geotechnical risk becomes more transparent and auditable.
References
Dorion, JF & Hadjigeorgiou, J 2013, ‘Corrosion considerations in the design and operation of rock support systems’, in Y Potvin & B Brady (eds), Ground Support 2013: Proceedings of the Seventh International Symposium on Ground Support in Mining and Underground Construction, Australian Centre for Geomechanics, Perth, pp. 497–509, https://doi.org/10.36487/ACG_rep/1304_34_Hadjigeorgiou
Hassel, R 2008, Corrosion of Rock Reinforcement in Underground Excavations, PhD thesis, Curtin University, Perth.
ICARN n.d., Inrush Hazard Management, International Caving Research Network, viewed 8 August 2024, https://icarn.ubc.ca/inrush-hazard-management/#:~:text=Inrushes%20pose%20a%20major%20risk,extreme%20cases%2C%20injuries%20and%20fatalities
Lett, J, Castro, R, Pereira, M, Osorio, A & Alvarez, P 2022, ‘BCRisk applications for rill swell hazard analysis in PC1: case study at Cadia East Operations’, in Y Potvin (ed.), Caving 2022: Proceedings of the Fifth International Conference on Block and Sublevel Caving, Australian Centre for Geomechanics, Perth, pp. 561–572, https://doi.org/10.36487/ACG_repo/2205_38
Llewelyn, K, Campbell, R, Yuniar, A, Sullivan, M & Di Ciolli, M 2022, ‘A risk-based approach to practical scope definition and management at PT Freeport Indonesia’, in Y Potvin (ed.), Caving 2022: Proceedings of the Fifth International Conference on Block and Sublevel Caving, Australian Centre for Geomechanics, Perth, pp. 67–78, https://doi.org/10.36487/ACG_repo/2205_02
NIOSH 2020, Mining Project: Managing Ground Support for Long-Term Stability in Underground Mines. U.S. Centres for Disease Control and Prevention, https://www.cdc.gov/niosh/mining/researchprogram/projects/project_managinggroundsupport.html
Power, GR & Campbell AD 2016, ‘Modelling of real time marker data to improve operational recovery in SLC mines, Proceedings of the Seventh International Conference and Exhibition on Mass Mining, Canadian Institute of Mining, Metallurgy and Petroleum, Sudbury, pp. 105–110.
Preston, RP, Roy, JM & Bewick, RP 2019, ‘Rusty bolts: planning for corrosion of ground support in underground mines’, in J Hadjigeorgiou & M Hudyma (eds), Ground Support 2019: Proceedings of the Ninth International Symposium on Ground Support in Mining and Underground Construction, Australian Centre for Geomechanics, Perth, pp. 423–436, https://doi.org/10.36487/ACG_rep/1925_29_Preston
Villa, D 2021, Geovia PCBC Mine Sequence Optimization forBlock Caving Using Concept of ‘Best and Worst Case’, from https://www.3ds.com/fileadmin/PRODUCTS-SERVICES/GEOVIA/PDF/New_Branding/GEOVIA-PCBC-WP-MineSeqOpt.pdf