Ore control based on value - experiences in design and execution (extract)

The guiding principles of this approach are:

• We mine rocks, not grades, and so all quality and geometallurgical characteristics that impact on the potential value to be extracted must be considered during ore definition.
• Processes must be developed that make the impact on value transparent across the mining value chain such that all technical disciplines understand the role they need to play to maximise value delivery.
• Ore control planning and execution decisions are holistically value
• Processes must be objective and produce repeatable. 

During the mining process it is important to focus on the value of all elements, impact of deleterious characteristics of the ore and the cross functional work involving different technical disciplines. All these activities help to optimise the quality of the feed to the plant and therefore contribute to the improvement of both productivity and profitability.

Often, mining operations predict annual revenue based on metal grade, a fixed price for a single metal/commodity and the tonnage of total ore blocks fed to the plant at a name-plate throughput rate and metal recovery, less mining costs. The operational reality is that, even under stable mining and plant conditions, ore throughput and commodity recovery fluctuate because of natural rock characteristics that are present within an orebody, such as variations in hardness and mineralogy, which affects recovery. As a result, not all ore blocks or even truckloads of a similar grade will provide the same contribution to value as shown in Figure 2.

Figure 2. Example of how rock type and characteristics can impact on value.

Most of the examples provided in this paper are sourced from the authors’ personal experience working with open pit mining operations in South Africa, Chile, Peru, Brazil and Australia. However, the ideas and principles are also considered applicable to any mining operation and extraction methodology.

What data is required to enable ore control based on value?

Ore control based on value is enabled by knowledge of:

• Revenue generating elements including both primary and by-products; for example, South African platinum mines can recover nickel and copper along with the platinum group
• Price penalty elements: for example, in iron ore mines in Australia and Brazil the potassium, phosphorous, silica and alumina content are important to steel makers and if specifications in shipments are not met then price penalties result in lower value being achieved.
• Cost driving elements including elements deleterious to mining (eg clays) and mineral processing (eg talc) as well as elements that need specific management processes to eliminate or mitigate safety, health or environmental risks (eg silica minerals in dust and acid generating materials in waste dumps).
• Geometallurgical elements/parameters that impact on plant performance such as processing throughput and product recovery.
• Mining and processing costs that are subject to production efficiency and variables such as fuel and electricity costs.
• Time value of information: Eg having data available well in advance of mining to facilitate planning and optimisation of blending activities to improve throughput and recovery in the processing plant.

Geoscientific data must be spatially representative for all mineral content, physical and geometallurgical properties to enable a comprehensive ore control model to be built for medium to short-term planning and mining. Proxies can be useful if strong correlations exist with primary data. For example, geochemical data can provide a good basis to model lithology if all the lithologies present are well understood and the appropriate geochemical data is available.

Fundamental to sampling is a scientific assessment of the sampling nomogram and selection of safe and appropriate processes for sample acquisition and handling. Any data that is gathered must be accompanied by appropriate quality assurance and quality control (QA/QC) procedures that allow real-time responses to sampling or analytical quality issues. Gathering data well ahead of mining allows for comprehensive QA/QC and proper data management systems as well as improved medium/short-term mine planning and blasting processes. Ore control data, which is used to deliver the value from the orebody, should be seen as just as important as exploration or Resource definition data and attract the same amount of effort in QAQC.

Examples of approaches that can be investigated and used to improve the collection of a full suite of data include:

• Reverse circulation (RC) drilling on statistically appropriate tight spacing, with automated sampling, well in advance of mining.
• X-ray fluorescence (XRF) and X-ray diffraction (XRD) scanning devices for geochemical and mineralogical analysis.
• Fourier Transform Infra-red (FTIR) analysis for mineralogy, geochemistry and
• Hyperspectral scanning of drill core and RC chips for texture and
• Measurements of drilling performance (for example the ROCMA System) to derive estimates of hardness and degrees of fracturing.
• Bench scale metallurgical test work to assess/update recovery
• Batching ore through processing plants and reconciling between ore control predictions and plant performance to validate geometallurgical assumptions.

Geology

Mapping, structure and modelling

Ore geometry is often controlled by lithological and structural geology features that increasingly are not being routinely mapped or modelled as part of the traditional grade control processes. A common excuse is because of safety considerations; however, a number of technologies are now available that ensure mine geologists do not need to approach blasted faces and can still safely capture appropriate mapping data quickly and in considerable detail. These include using smartphone or tablet applications and 3D scanning technology with photogrammetry via fixed cameras or drones. This means real-time updates to 3D models of ore controlling features are possible, leading to the limitation of ore loss and dilution that impacts directly on recovered value.

In addition to traditional sampling methods, photogrammetry techniques in underground mines can be used to map mining faces and correlate mined volumes to typical stratigraphic grade profiles, which is then used to estimate the production grade and ore characteristics of mining panels and reconcile back to the Resource model.

The ability to rapidly update 3D geological models using field observations as a direct input has a clear benefit to the accuracy of ore block designs and mining tactics. Implicit modelling software has now reached a level of commercial development that it is considered superior to manual, explicitly drawn interpretations, providing models that are objective and repeatable (Hodkiewicz, 2014). Initial set-up of an implicit modelling process can be time consuming; however, the benefits accumulate rapidly. This approach is even more powerful when combined with a data acquisition process that deliver input data well ahead of production such that questions about the model can be validated by re-visiting exposures in the mine and running multiple iterations of the model prior to mining.

Modern software and hardware have the capability to load and display large data sets, which can reveal structural and ore trends that might be missed if the focus is only on one blast at a time. Figure 3 shows a large blasthole sample data set that clearly shows a folded structure within grade data using the 'X-ray plunge projection' methodology described by Cowan (2014).

Figure 3. Folded structures revealed in large blasthole data set (after Cowan, 2014).

Design and planning - integrated planning across the value chain

At many operations mine planning focuses on equipment and material movements only. Including the activities involved in ore control data acquisition into the short-term mine plan and mining schedule is important to ensure that all activities in the mine are coordinated and safely executed to the right level of quality and detail at the right time (Ortiz and Magri, 2014). Active engagement by mine geologists in the operational routines is critical to embed and sustain ore control based on value. Stakeholder alignment and change management required to establish the correct routines are much more effective when a clear value proposition can be put forward.

RC drilling activities involve large mobile equipment operating in the mine in addition to blasthole drills, loading and hauling equipment and other mobile plant needed to operate the mine. Integrating the ore control drilling and sampling activities into the daily operational management processes, by including them in short-term mine planning, ensures safe and timely execution of the ore control data acquisition program, without interfering with production. Just as important to including these activities in the mine plan is that the results are used to ensure that the value delivered can be maximised. As an example, a key metric for many mines is the annual budget – it becomes a target (unfortunately in tonnes) to be achieve at almost any cost or methodology! By including the ore control data collection activities into the mines' 18-24-month planning cycle it is possible to schedule mining during the annual budget planning period on the actual ore control results, rather than on a Resource estimate. This value add is explained in more detail in the next section [available in the full paper].

Adding value by acquiring data well ahead of production

Significant value can be added by moving from 'just-in-time grade control' in favour of ore control data gathering well ahead of production. This can be achieved with RC drilling (Ortiz, Magri and Libano, 2012; Ortiz and Magri, 2014) 6-18 months ahead of the time when the ore will be mined. When collecting data ahead of production the value comes from:

• Gathering fit for purpose quality samples that provide a reliable and representative view of geological contact locations and rock properties, as opposed to making do with blasthole samples that cannot pinpoint rock type contacts which are very difficult to sample correctly and therefore have very poor quality (Pitard, 2008).
• Removing time and cost from blasthole drilling by removing the sampling activities from the process. It is common to discover that the dollar per metre manually sampling cost savings in blasthole drilling more than paid for the RC drilling ahead of production, which along with improved sample quality and early availability of data helped motivate for the RC drilling process and costs. Experiences detailing a 54 per cent reduction in sampling and assaying costs at the Mogalakwena platinum mine are documented in Kirk, Muzondo and Harney (2011).
• Having time for quality geometallurgical test work, which typically requires more complex laboratory tests and time compared to grade only assays.
• Acquiring spatially representative geometallurgical data that enables three-dimensional modelling of mineral/metal recovery which is a key input to estimating the value of each mining block, thereby determining the most value adding material destination.
• Understanding the spatial distribution of ore zones, lithologies and structures at the mining scale for areas larger than a single blast, such that the continuity of value driving parameters can be assessed with high confidence and used in planning mining tactics. For example:
  • Where to start mining and how to progress the development across a full bench
  • Design of blasting to control ore loss and dilution by separating ore and waste into different blasts and choosing the mining direction that leads to less mixing of materials (eg up-dip as opposed to along strike).
  • Identify where different rock properties present risks and opportunities to optimise blast designs and blending of material feed to the plant (eg hard versus soft areas, clay zones, wet zones etc).
  • Forewarn the processing plant that problematic ores will be delivered, or identify blending opportunities that may mitigate these issues for the plant and optimise recovery.
• Allowing for follow up field investigations and/or iterations of mine plans to identify the most valuable options.
• Providing the ability to re-assess strategies and tactics with high confidence in the event of changes in the mine such as equipment breakdown or geotechnical issues.

Want to keep reading? Access the full paper via the International Mining Geology 2024 Conference proceedings here.

References

Cowan, E J, 2014. 'X-ray plunge projection' - understanding structural geology from grade data, in Mineral Resource and Ore Reserve Estimation - The Aus/MM Guide to Good Practice, AuslMM Monograph 30, second edition, pp 207-220 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Hodkiewicz, P F, 2014. Rapid Resource Modelling - A Vision for Faster and Better Mining Decisions, in Proceedings of the Ninth International Mining Geology Conference 2014, pp 183-188 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Kirk, G, Muzondo, T and Harney, D, 2011. Improved Grade Control using Reverse Circulation Drilling at Mogalakwena Platinum Mine, South Africa, in Proceedings of the Eighth International Mining Geology Conference 2011, pp 329-340 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Ortiz, J Mand Magri, E J, 2014. Designing an advanced RC drilling grid for short-term planning in open pit mines: three case studies, The Journal of The Southern African Institute of Mining and Metallurgy, 114:631-637.

Ortiz, J M, Magri, E J and Libano, R, 2012. Improving financial returns from mining through geostatistical simulation and the optimized advance drilling grid at El Tesoro Copper Mine, The Journal of The Southern African Institute of Mining and Metallurgy Transactions, 112:15-22.

Pitard, F, 2008. Blasthole sampling for grade control - the many problems and solutions, in Proceedings of the Sampling Conference 2008, pp 15-21 (The Australasian Institute of Mining and Metallurgy: Melbourne).