If we could achieve so much in the 1990s, how much more can we achieve today when we have the same imperative, the same potential and a new generation of robust, reliable, high-technology tools?
The Mine to Mill methodology was developed during the commodity price downturn of the 1990s and was widely adopted at numerous sites around the world. Mine to Mill relies on the fact that comminution is usually the site processing bottleneck and that blasting is more efficient at breaking rock than grinding. Substantial savings can be achieved when operators:
- understand and characterise rock breakage from mining to the mill
- develop models and simulators for blast design, fragmentation, crusher and mill circuits
- develop tools to measure the particle size distribution of rocks on conveyors and run-of-mine stockpiles in real time
- ensure effective communication across the silos between geologists, blast design engineers, mining engineers and metallurgists.
This approach and the technology tools available in the 1990s were widely adopted around the world with significant benefits. Semi-autogenous (SAG) mill throughput increases of ten to 20 per cent were common. We thought that this approach to managing operations was here to stay since it appeared to be ‘free money’ for operators. However, while some operators advanced the approach during the subsequent boom and focus on ‘production at any cost’, many others reverted to old habits of optimising within organisational silos. Mine to Mill wasn’t crucial to their survival.
Now that the commodity cycle is back in downturn, the simple productivity gains offered by Mine to Mill look attractive again. Since the 1990s, many new or advanced technology tools have been developed to enhance the approach. Now is the time to adopt the new technologies and, this time, embed them into a more universal and lasting industry adoption.
Past practice and essential requirements
In its broadest sense, Mine to Mill integrates all aspects of geometallurgy and production steps with processing and marketing. In this paper, we confine our comments to the subset of Mine to Mill that focuses on the integration of blasting with comminution and separation. We refer to this as ‘advanced blasting for comminution’ – the ABC of Mine to Mill.
An excellent summary of the steps involved in any Mine to Mill project sorts them into three main areas (McKee, 2013):
- Good data needs to be collected about the ore before it is mined and as it passes through the production chain, the equipment and processes used in production and the process performance and cost.
- Good, robust analytical tools and models are needed to evaluate options and identify optimum operating points for a range of feed types and conditions. The best models also account for economic factors such as metal price and operating cost.
- Any Mine to Mill project requires tools for ongoing monitoring, assessment and evaluation. Findings need to be validated and optimisation kept on track to fully realise the benefits.
Non-technical factors are also critical, including sustained management support, availability of staff with specialist skills and an enabling organisational structure.
Early Mine to Mill projects included scheduling to smooth ore variations, building stockpiles to ‘campaign’ different ore types or redesigning underground activities to eliminate cash-negative ore while rescheduling surface operations to eliminate the supposedly fixed costs associated with them.
These examples show that Mine to Mill projects did whatever was required to improve overall mine site performance. They were specific to the orebody and situation. Many case studies reported significant gains, typically in the range of ten to 20 per cent productivity improvements across the mine site, with little or no capital expenditure. These were dramatic improvements by any measure compared with working in isolated silos (McKee, 2013).
Mine to Mill in 1996
By 1990, higher-speed computational power enabled innovative software for mathematical modelling and simulation of industrial processes, including mining and mineral processing. The AMIRA project ‘Optimisation of fragmentation for downstream processing’ (which ran from 1996 to 2002) was a collaborative research project to exploit this capability and develop operating strategies to enhance mining and downstream processing activities. The history, concepts and case study projects are presented in McKee (2013).
In 1996, the technology tools available to Mine to Mill were limited to elementary versions of tools such as (or similar to):
- KSimBlast, blast design software
- Kuz-Ram Fragmentation Model to predict the particle size distribution of the blast in the stockpile
- Split-Online or other image analysis of the particle size distribution at the truck tip to the primary crusher and on the conveyors from the primary crusher to the SAG mill feed
- JKSimMet, simulation software for the comminution circuit.
There was minimal information about mineralogy, and work-arounds had to be devised to deal with complex, multicomponent ore types and non-standard mining and processing scenarios. Early Mine to Mill projects relied heavily on the knowledge, experience and desire for cooperation of the project team. Refer to Figure 1 for a schematic representation of the work process.
Importantly, the projects and survival imperative of the times engendered dialogue, collaboration and cooperation between geologists, blast and mining engineers, metallurgists and general managers (Kanchibotla et al, 1998; Lam et al, 2001; Valery et al, 2001). In 2012, McCaffery suggested:
Mine to Mill (and geometallurgy) is just code for making the effort and putting the processes in place to record, in an accessible format, an understanding of the orebody, how changes in the orebody and operating practice drive productivity and production and understanding the operating parameters in the mine and mill that can be manipulated to improve productivity and operating cost. It is what people in mining and processing at sites should be doing as a normal part of their day-to-day business (McCaffery, 2012, personal communication).
New tools and the next generation of Mine to Mill projects
Since 2002, there have been marked developments in image analysis, GPS, simulation software, radio frequency tracking devices for ore, in-plant instrumentation to measure flows, online particle size monitors, mineral liberation analysis, geometallurgy and equipment monitoring instrumentation.
Though some early adopters lost their way, new adopters achieved outstanding improvements, such as Antamina’s 45 per cent increase in SAG mill throughput followed by a further ten per cent later (Rybinsky et al, 2011, Valery and Rybinski, 2012). Other case studies have been discussed by Bennett et al (2014), Hart et al (2011), Dance et al (2007), Diaz et al (2015), Renner et al (2006), McCaffery et al (2006) and Gomes et al (2010).
Now that the productivity imperative has returned, it is time for more operators to learn from the old and recent successes and adopt the ABC approach using the new technology tools, which include:
- More complex software for blast design, analysis and management.
- Advances in fragmentation modelling.
- Image analysis with comprehensive rock fragmentation software for automated particle sizing at truck dump and conveyor belt locations.
- End-to-end modelling packages such
as the Integrated Extraction Simulator developed by CRC ORE and provided by JKTech.
- Tools such as Split FX® that process
3D point clouds from LiDAR scans and photogrammetry to automatically characterise fracture attributes.
- GPS digital drilling systems to guide, monitor and save drill hole patterns.
- Measured While Drilling data that captures rock hardness measurements from blasthole drills to reliably categorise rock types.
- Blast movement monitors that accurately measure 3D blast movement
to minimise ore loss and dilution and significantly increase ore yield (La Rosa and Thornton, 2011).
- High-energy explosives to improve post-blast fragmentation (Hawke and Dominguez, 2015).
- Electronic detonators programmable in one millisecond steps to provide blast flexibility and precision.
- Rock tracking devices such as Metso SmartTags® (La Rosa et al, 2007) to track ore from blasthole through mining, handling, stockpiling and mill feed. Combined with mill performance and online size data, the effects of ore and blast changes can be correlated with their impact on processing (Figure 2).
- Image analysis of particle size distribution in the stockpile.
- MineWare’s Argus shovel monitor to improve shovel and operator performance, optimise truck loads and reduce costs.
- Prompt gamma neutron activation analysis like GeoScan or CB Omni for online measurement of elements in rock streams on belt conveyors for grade engineering or stockpile blending.
- Safe, accurate and quick measurements of the crusher closed side setting from six to 220 mm through the ‘C-Gap’ digital measurement tool.
- Accurate and reliable online particle size monitors and sampling stations to sample and measure the particle size distribution and per cent solids of the total grinding circuit classifier product.
- Acoustic monitoring software to monitor operating conditions in SAG mills and allow optimisation.
- Automated quantitative mineralogy to provide mineral liberation and association data (using area scan of particles) and elemental distributions of ore and waste minerals.
- Using geometallurgy, the integration of geological, mining, metallurgical, environmental and economic information, to maximise the net present value of an orebody while minimising technical and operational risk.
Twenty years later – what have we learned?
For many of the early adopters, the initial success of Mine to Mill fell victim to organisational systems and personal incentives that ‘sprang back’ to default when survival was no longer in doubt. Any reinvention of Mine to Mill needs to recognise why we failed to hold the gains the first time and then set about to fix those flaws.
Antamina and Cerro Corona demonstrated what can be achieved by applying some of the new tools in a Mine to Mill project. Models of drilling and blasting, crushing and grinding were combined with knowledge of different ore domains and plant surveys and SmartTag tracking of blasting changes. The result was a significant increase in mill throughput from 2750 to 4400 t/h and a 25 per cent reduction in specific energy (kWh/t; Valery et al and Rybinski, 2012). At Cerro Corona, throughput was increased by six per cent overall (and by as much as 15 per cent for harder ores) and the SAG-specific energy was reduced by over nine per cent (Diaz et al, 2015).
Published case studies from 1996-2002 (Table 1) and 2003-2016 (Table 2) have been provided. There are other operations that have realised benefits from Mine to Mill but have chosen not to publish the results.
The fatal flaw was that earlier projects did not lock the new operating methodology into organisational and management systems. When profits rose in the boom, attention was focused on expanding output and resources. Blasting engineers were rewarded for reducing their cost per tonne and miners were rewarded for increasing tonnage. Metallurgists were incentivised to increase tonnes and recovery and to reduce costs, but were rarely encouraged to increase product quality beyond ‘good enough to sell’. Smelters remove impurities at much higher cost than the concentrator, but that was the smelter’s problem.
If we are to truly succeed with Mine to Mill in the future, implementation must be supported, not undermined, by our organisational systems. We need to design key performance indicators (KPIs) that encourage integration and work across the silos to provide mine site targets.
Mining operations are complex, and we need as many good measures as we can find. While these measures are getting better, they are still imperfect. We need to distil the complexity to the simple fundamental basics of what makes a good integrated organisational team and then set the minimum few KPIs that let clever people get on with their own jobs while using their initiative for the group benefit. This will keep the shareholders on side.
We know Mine to Mill works. We made it work well with old technology. So now that we have more tools, more appropriate data and a business imperative, we can make it work much better again.
We have to recognise the complexity of sites and the differences between them. The principles of integration will work everywhere, but the crucial components will be site-specific. Therefore, off-the-shelf solutions provided by external management groups won’t work. The components of the solutions need to be specifically assembled by multidisciplinary teams of people who know the details and constraints of each site and understand the difference between minimising costs in silos and maximising mine site profits.
This time, we need to support Mine to Mill with an understanding of human and organisational behaviour and lock that into our incentive systems. We need to greatly simplify the overwhelming list of performance targets and distil them to the critical few that encourage the behaviour needed to integrate the operation. Then we can allow our clever people to get on with the job.
This paper was inspired by observation and comments from many people that Mine to Mill is often overlooked, yet the same problems that drove the original AMIRA project are now repeating themselves.
We thank the contributions made by the many suppliers of the technology tools that are now available for Mine to Mill projects. Valuable contributions were made by Walter Valery and Sarma Kanchibotla, both of whom have been associated with Mine to Mill projects almost since their inception.
Bennett D, Tordoir A, Walker P, La Rosa D, Valery W and Duffy K, 2014. Throughput forecasting and optimisation at the Phu Kham Copper-Gold Operation, in Proceedings 12th AusIMM Mill Operators’ Conference 2014, pp 381–390 (The Australasian Institute of Mining and Metallurgy: Melbourne).
Dance A, Valery W, Jankovic A, La Rosa D and Esen S, 2007. Maintaining the benefit – how to ensure Mine to Mill continues to work or you, in Proceedings Ninth Mill Operators’ Conference, pp 215–222 (The Australasian Institute of Mining and Metallurgy: Melbourne).
Diaz R, Mamani H, Valery W, Jankovic A, Valle R and Duffy K, 2015. Diagnosis of process health, its treatment and improvement to maximise plant throughput at Goldfields Cerro Corona, in Proceedings SAG 2015, 20 p.
Gomes M P, Tavarez Jnr L, Nunes E, Colacioppo J, Jankovic A and Valery W, 2010. Optimization of the SAG mill circuit at Kinross Paracatu, Brazil [online], Comminution10. Available from: www.metsominerals.com.ph/miningandconstruction/mct_service.nsf/WebWID/WTB-120106-22576-D2E74/$File/020.pdf
Hart S, Rees T, Tavani S, Valery W and Jankovic A, 2011. Process integration and optimisation of the Boddington HPGR circuit, in Proceedings SAG 2011, 16 p.
Hawke S J and Dominguez L A, 2015. A simple technique for using high energy in blasting, in Proceedings 11th International Symposium on Rock Fragmentation by Blasting, pp 321–326 (The Australasian Institute of Mining and Metallurgy: Melbourne).
Kanchibotla S S, Morrell S, Valery Jnr W and O’Laughlin P, 1998. Exploring the effect of blast design on SAG mill throughput at KCGM, in Proceedings Mine to Mill 1998, 17 p (The Australasian Institute of Mining and Metallurgy: Melbourne).
La Rosa D and Thornton D, 2011. Blast movement modelling and measurement, in Proceedings 35th APCOM Symposium 2011, pp 297–310 (The Australasian Institute of Mining and Metallurgy: Melbourne).
La Rosa D, Valery W, Wortley M and Ozkocak T, 2007. The use of radio frequency ID tags to track ore in mining operations, in Proceedings 31st International Symposium on Computer Applications in the Mineral Industries.
Lam M, Jankovic A, Valery W and Kanchibotla S, 2001. Increasing SAG mill circuit throughput at Porgera gold mine by optimising blast fragmentation, in Proceedings SAG 2001, pp 271–288.
McCaffery K, Mahon J, Arif J and Burger B, 2006. Batu Hijau – controlled mine blasting and blending to optimise process production at Batu Hijau, in Proceedings SAG 2006.
McKee D, 2013. Understanding Mine to Mill, 96 p (Cooperative Research Centre for Optimising Resource Extraction: Brisbane).
Renner D, La Rosa D, DeKlerk W, Valery W, Sampson P, Bonney Noi S and Jankovic A, 2006. Anglogold Ashanti Iduapriem mining and milling process integration and optimization, in Proceedings SAG 2006, pp 249–264.
Rybinski E, Ghersi J, Davila F, Linares J, Valery W, Jankovic A, Valle R and Dikmen S, 2011. Optimisation and continuous improvement of Antamina comminution circuit, in Proceedings SAG 2011.
Valery W and Rybinski E, 2012. Optimisation process at Antamina boosts production and energy efficiency: helping a large copper/zinc mine meet the economic challenge of processing harder ore types [online], Engineering and Mining Journal, September 2012. Available from: www.e-mj.com/features/2373-