Tag Archives: machine learning

Freeport to introduce Gecko robots to power AI inputs at operations

Automation, Base metals, Equipment maintenance, IoT, Mineral processing, Mining equipment, Mining services, Sustainable mining, , , , , ,

Freeport-McMoRan is to start using robots from US-based Gecko Robotics to generate the required data to feed machine-learning algorithms, attendees of the Financial Times’ Mining Summit 2024, in London, heard last month.

Freeport is looking to bolster copper production from its existing assets in expectation of the world requiring more of the red metal, and it sees AI and machine learning as one potential avenue to pursue.

Speaking on a panel at the event in late September, Bill Cobb, Vice President and Chief Sustainability Officer, acknowledged the copper major had just signed a contract with Gecko Robotics, led by CEO Jake Loosararian.

He explained: “We collect a lot of data, and it’s across the open pits; so all the haulage equipment, the drills, the shovels, etc in the processing facilities…And there is a lot of data, [so] we employ a lot of data science.”

The company is using this data to attempt to “reinvent what we do”, he added.

He explained: “And you heard it over the panels today. Whether it’s the loss of head grade that they’ve experienced over the last 10 years. You know, all of us in the copper sector are seeing decreases in head grades, right? That means we’re going to have to get more efficient, and so it’s really pushing the question on technology and what can technology do.”

Loosararian, for his part, said that the company’s robots were powering AI-based systems that customers use in the mining, energy, manufacturing, defence and infrastructure sectors.

“We gather this information and data sets about how to solve the problem a lot of these facilities have,” he explained. “[In many cases] they’re (the facilities) very old, they’ve reached their useful life and now we’re asking them to last way longer and also produce way more.”

Gecko’s technology is specifically based around how to extend the useful life of this infrastructure so it doesn’t result in catastrophic failures that end up increasing emissions and put the safety of people at risk, he explained.

“And, so, it’s really important to, first and foremost, reject the idea that we have to react to things breaking,” he said. “Technology exists to be able to ensure that, instead of react[ing] to it.”

Loosararian cited a study showing that if major industry eliminated forced outages and catastrophic failures in the US by 2030, there would be an 18% reduction in emissions.

“We need more data to understand the impacts of certain decisions,” he said. “But, I think, really, we need these materials, we need these minerals and we have to ensure that we can rely on – in a way that’s increasing the amount of production while reducing emissions – them to meet the needs for this transition.

“That’s the thing that gets me excited and, I think, that motivates the next generation of technology enthusiasts that need to be flooding through into the industry and places like Freeport.

“I see that beginning to happen because leadership is beginning to embrace technology.”

TOMRA taps into deep learning AI network for latest ore sorting advances with OBTAIN

Automation, Comminution of minerals, Environmental, Mineral processing, Mining services, Mining software, Sustainable mining, , , , , , , , ,

TOMRA is looking to leverage artificial intelligence as part of a plan to unlock new opportunities for mining operations using its sensor-based sorting technology.

The company explained: “The ability of computer systems to mimic human thought and decision making to perform tasks that traditionally required human intelligence has played an important role in TOMRA’s sensor-based sorting solutions for decades, automating the process and improving the accuracy and efficiency of the sorters, unlocking value for mining operations.

“Over the years, sensor-based sorting technology has developed, and TOMRA has been using machine learning in its X-ray Transmission (XRT) and Near-Infrared (NIR) sorters for the last 10 years.”

Now TOMRA Mining is opening a new era in sorting with its latest innovation, OBTAIN™, which leverages deep learning to bring single-particle precision to high throughput particle sorting, it says. This solution takes capacity, quality and recovery to a new level, and unlocks value through a wealth of extremely detailed and accurate data for better-informed decision making, it added.

This software uses a neuronal network to identify the properties of each particle accurately and independently of the sorter’s capacity, achieving new-found precision and reliability in detection and ejection. Based on its specific requirements, the mining operation has the flexibility to either enhance the throughput of the sorter while maintaining consistent sorting efficiency or improve sorting precision without compromising the existing throughput.

TOMRA says: “OBTAIN proves advantageous for a fully operational mine by enhancing recovery rates and elevating product quality within the existing throughput. Conversely, in mines with additional capacity, it facilitates increased throughput without compromising product quality. Furthermore, this innovative technology has the capability to unlock untapped value from low-grade ore, waste dumps, or materials previously deemed uneconomical for processing.”

OBTAIN will also add value to a mining operation with a wealth of extremely detailed and accurate data, such as precise online particle size distribution of the feed.

When used in combination with TOMRA Insight, it can provide the customer with detailed reporting on the performance of the sorter and its components to help them optimise the process, as well as enable them to plan for predictive maintenance, the company says.

The OBTAIN software has been developed for TOMRA’s XRT sorters. It will be available on new models, but there will also be an upgrade package available for existing machines, providing a significant opportunity for customers already operating TOMRA XRT sorters, to substantially enhance the sorting performance where it proves to be a suitable solution.

TOMRA has partnered with two customers to test the OBTAIN in real working conditions. The software has been operating for close to 18 months at the Wolfram Bergbau & Hütten tungsten mine in Mittersill, Austria, where it has delivered consistent and reliable performance. The vicinity of the mine to TOMRA’s development team, based in Germany, has made it a suitable testing ground for the first phase, as they have been able to monitor it closely. A second phase of testing to quantify the improvements has been carried out with a trusted long-standing customer in a magnesite application. The successful tests have shown that OBTAIN is ready to transform sensor-based XRT sorting in numerous applications, according to TOMRA.

Micromine introduces cloud-based AI capabilities to software suite

IoT, Mining services, Mining software, , , , , , , , , , , , , ,

Micromine has revealed its 2024 release, saying the update comes with new features and enhancements across the company’s entire product suite, further underscoring its mission to deliver state-of-the-art technology to reshape the industry.

Kiril Alampieski, Micromine’s Chief Strategy and Product Officer, said: “Addressing the ever-evolving needs and requirements of our clients is the driving force behind our relentless focus on innovation to ensure the industry can achieve more by integrating technology within the client operational workflows and reducing data errors and productivity bottlenecks.”

Powered by the company’s cloud-native data sharing and collaboration tool, Micromine Nexus, the 2024 release introduced two significant features: Micromine Origin Copilot and Micromine Geobank Panorama, enhancing the company’s exploration solutions.

Alampieski said: “The ground-breaking feature, Micromine Origin Copilot, is poised to revolutionise geological and resource modelling. The cloud-based AI companion can process data categorised or quantified and employs advanced machine-learning techniques to craft thorough and robust models autonomously.

“Micromine Origin Copilot plays the role of a skilled ally, offering a supplementary perspective to support and authenticate conventional resource estimation methods, thereby empowering geologists with greater confidence and peace of mind in their models. This AI journey begins with grade modelling and will be implemented to other features throughout Micromine Origin and the wider Micromine ecosystem.”

Micromine Geobank’s Panorama feature also benefits from AI and cloud-computing assistance, able to automate the labour-intensive task of creating a seamless down-hole image from drill core imagery, the company says.

The company’s three mine planning solutions received updates for the 2024 release.

Micromine Alastri, the company says, expands on its industry-leading battery-electric haulage modelling capabilities. The functionality allows mine planners to analyse, validate and implement robust decarbonisation strategies that describe what the mine of the future looks like in practice.

Micromine Spry evolved with a broader set of tools and industry-leading visualisation to tackle the demands of modern coal and soft-rock mine planning head-on. The updates are designed to provide a better understanding of mine data and more straightforward methods to communicate results.

Micromine Beyond improves strategic scheduling with new pit optimisation and materials management capabilities. The new functionality builds more confidence and certainty when developing life-of-mine plans.

Alampieski said: “Micromine’s mine planning and scheduling tools are engineered to meet the needs of mine planners at each planning horizon and primary mining method. The 2024 release is future-focused, delivering precise and dependable outcomes for today’s mine planners.”

Lastly, Micromine’s mine production, control and fleet management solution, Micromine Pitram, adds significant improvements, making it easier to track shift progress and gain valuable insights, as well as reducing time spent on data extraction and manipulation.

Leveraging the intelligent mine concept for improved profitability, sustainability

Automation, Equipment maintenance, IoT, Mining equipment, Mining maintenance, Mining services, , , , , , , ,

The case for digitalisation is clear, according to AspenTech’s Jeannette McGill*, with digitalisation being critical for the metals and mining industry to achieve sustainability and operational excellence in the years ahead.

This is why the more forward-looking decision makers in the mining industry are embedding digital capabilities in their operations so they remain agile, competitive and profitable over the long term, while realising immediate and measurable benefits, she says. Technology providers have responded to the industry’s needs with solutions designed for the mining sector that directly address the dual imperatives of greater business efficiency and enhanced sustainability.

The intelligent mine

In their digitalisation initiatives, today’s operators also know that managing data more efficiently and effectively will be crucial in helping them to meet the challenges they face. Multiple difficulties remain in the way that organisations across the sector manage their data.

Senior mining company executives frequently make tough decisions but, in doing so, they must aggregate isolated pockets of data to generate insights that are relevant and actionable. For data to be available is no longer sufficient. The top priority for effective decision making is for appropriate management of diverse and disparate data sets in a range of locations.

The key is to integrate data and conduct high-level analysis with an understanding of the domain work requirements. Mining companies achieving this will establish what has become known as the intelligent mine. This is a concept that focuses on centralising information from multiple locations and business processes to reveal useful insights. It supports senior level decision making with designed-for-purpose analytical platforms.

Data held in 50 separate systems will not in itself drive operational efficiencies or support sustainable operations. By addressing several issues simultaneously, an organisation is more likely to move towards the intelligent mine.

First, though, businesses must implement automated data gathering systems to capture relevant data from various parts of the mining process and facilities. Second, organisations must have tools that detect bad data because only good data enables good decisions. They must ensure all changes are consistent, correct and improve data processing. Third, the business should assist the mine personnel by providing the capability to integrate data with built-in relevant analytics, so they process and act on insights in a meaningful time frame.

The predictive dimension

To create an intelligent mine, good data is important but only one part of the equation. Organisations must also analyse data in different ways within a context of the problem to be solved with appropriate predictive outcomes.

Companies need to predict the degradation of equipment that, if unattended, will lead to equipment breakdowns and unplanned maintenance, thereby adversely affecting both operational efficiency, reliability, sustainability and safety. Mining is equipment- and infrastructure-intensive with expensive machinery. It demands operational continuity for profitability and sustainability.

Bringing in prescriptive maintenance

Traditional preventive maintenance methods generally fail on the benchmark of equipment availability and performance. Earlier preventive maintenance efforts were unable to deliver sufficient time-to-failure warnings to deliver a significant impact on profitability.

That is where modern prescriptive maintenance plays a vital role. The technology monitors data from sensors on and around the machine to develop intense multi-dimensional and temporal patterns of normal operation, abnormal operation and explicit degradation patterns that precede breakdown. This provides early warnings, using artificial intelligence (AI)/machine learning digital technology to spot patterns that humans will never pick up.

Also surpassing human capability, the technology can assess the health of numerous machines every few minutes. It also delivers early warnings to maintenance teams, often with prescriptive advice on resolution. Facing tough challenges and spread thinly over large sites, workers benefit from warnings. Much of the intense repetitive analytics and engineering help them prioritise what matters most. Maintenance teams with such prescriptive maintenance tools ensure an intelligent mine makes significant progress in eliminating unplanned breakdowns.

Finding a solution

An asset performance management (APM) approach – with integrated prescriptive maintenance capability – ensures mines improve reliability, availability and uptime, simultaneously reducing the considerable cost of redundant equipment.

Operations teams often work on the assumption of lower availability by, for example, installing three machines when they only need two, or purchasing 10 trucks to ensure they always have eight up and running. These practices are now deemed too wasteful and have become unsustainable.

By embracing the most effective technology, mines can achieve benchmark reliability without the need for more people, equipment, or expenditure. Companies can operate at the required production levels and either mothball or switch off redundant equipment. Being able to do this with full confidence it will actually enhance overall outcomes, makes a significant contribution to the bottom line. It reduces emissions and increases sustainability.

Yet to develop an efficient digitalisation strategy, certain components must be in place. All too often, mines try to invest minimally in digital solutions to save money. Without domain-centric AI/machine learning analytics, this limits the reach and value technology can deliver. Successful digital strategies deploy solutions that draw on data from sensors and other sources. Enterprise resource planning systems, manufacturing execution systems, laboratory information management systems and advanced process control systems are all part of the mix, as well as general mine planning and design systems. Machine learning and other data science techniques require timely delivery of available data, so historian technology plays a vital role.

Across the industry, a growing numbers of mines are pursuing an APM approach. Australia-based gold miner, Evolution Mining, for example, has deployed Aspen Mtell software at the company’s Mungari Gold Operations, in Western Australia, to help reduce unplanned downtime and provide information to support productivity improvements.

Greg Walker, previously Evolution Mining Mungari General Manager, said: “Evolution’s Data Enabled Business Improvement program has achieved excellent results in recent years. With this new technology, Mungari Gold Operations can achieve further productivity improvements via increased asset availability.”

Looking to the future

Today’s mining industry is now sufficiently mature that it should fully embrace digital optimisation technologies. Operators that fail to adapt and build a strategy to utilise such technology are destined to struggle against competitors that do. Prescriptive maintenance delivers quick results by improving the use of existing capital assets and eliminating the surprise of unplanned downtime, which directly affects productivity, safety and sustainability.

The industry should understand how scalable prescriptive maintenance solutions add value to assets. This works as well with a single asset, conveyer system, a processing plant, a large mill, as it does with equipment across a worldwide enterprise. The truly intelligent mine empowers mining companies across a vast array of contemporary challenges. From reducing unplanned downtime and decreasing safety risks, to greater operational efficiency, sustainability and increased profitability, this approach will be essential for mining companies to surmount all their challenges in short and longer terms.

*Jeannette McGill is VP and GM of Metals & Mining at AspenTech

Weir preparing to trial proprietary ore sorting tech by the end of 2022

Automation, Comminution of minerals, Explosives and blasting, IoT, Mineral processing, Mining equipment, Mining services, Mining software, Sustainable mining, Water management, , , , , , , , , , , , , ,

In the Weir Group Capital Markets Event presentation last week, Chris Carpenter revealed that the company was collaborating within its divisions on trials of ore sorting technology in an effort to move less rock at mine sites and optimise processing within the plant.

Carpenter, Vice President of Technology at Weir ESCO, said the company was combining Motion Metrics’ particle size distribution (PSD) capability with ore characterisation technology to explore “in-pit sorting” opportunities for its clients.

“Looking further out, we believe ore characterisation and in-pit ore sorting has the potential to transform mining by moving less rock, using less energy and creating less waste,” he said during his presentation. “Ore characterisation technology, which is underpinned by sophisticated sensing systems, captures critical data on properties and composition of rock, including rock hardness and mineral and moisture content.

“When coupled with Motion Metrics fragmentation analysis technology, it has the potential to be a game changer, giving miners a full picture of the size and characteristics of rocks.”

Motion Metrics, a developer of artificial intelligence (AI) and 3D rugged machine vision technology, was acquired by Weir almost a year ago, with the business incorporated into the Weir ESCO division. Its smart, rugged cameras monitor and provide data on equipment performance, faults, payloads and rock fragmentation. This data is then analysed using embedded and cloud-based AI to provide real-time feedback to the mining operation.

These technologies were initially developed for ground engaging tool applications but have recently been extended into a suite of products and solutions that can be applied from drill and blast through to primary processing.

Carpenter said the added PSD capability from Motion Metrics was expanding the company’s value presence across the mine to the processing plant, where Weir Minerals operates.

“Results from early adoption of Motion Metrics PSD solutions have been extremely encouraging,” he said. “Feedback from customers is positive; data sharing and collaboration have increased.

“Given this early progress, we are really excited about the opportunity and expect fragmentation analysis to be a key growth driver for Motion Metrics in the years to come.”

On the in-pit sorting potential, Carpenter said Weir ESCO had laboratory-validated equipment and field trials of its proprietary solution that were due to start at customer sites before the end of the year tied to these developments.

“If successful, this technology opens the door to in-pit sorting, where miners complete the first stage of crushing in the pit and analyse the outputs to make real-time decisions about which rocks have sufficient mineral content to be moved,” he said. “This is a step change from the current process, where energy is expended in transporting and processing all of the rocks, regardless of mineral content, and with significant waste generated from zero- and low-grade material.”

He concluded: “Our vision is to move less rock, moving only the rocks with sufficient mineral content and using the data that is captured on size and hardness to optimise processing. The natural evolution thereafter will be towards real-time automation control of processing equipment, ensuring the right rocks are processed in the most efficient way, using less energy and creating less waste.”

Gradiant concentrating its mining proposition

Automation, Base metals, Environmental, Mineral processing, Mining services, Mining software, Water management, , , , , , , , , , , ,

There are plenty of mining applications one can see Boston, Massachusetts-based Gradiant’s end-to-end water technology solutions serving.

A spinout of the Massachusetts Institute of Technology, the company calls itself the “experts” of industrial water, water reuse, minimum liquid discharge (MLD) and zero liquid discharge (ZLD), and resource recovery of metals and minerals.

That is a big remit, hence the reason why it caters to at least nine industries on a global basis in mission-critical water operations, with over 70% of its clients being Fortune 100 companies in the world’s essential industries.

Mining companies have historically been wary of suppliers that serve a variety of industries, believing their needs rarely cross over with the requirements of other industries. Gradiant believes it is different in that its solutions incorporate not only the hardware and software to fine-tune water technologies, but also the artificial intelligence (AI) to ensure the tools being used are effective regardless of the inputs.

This includes the RO Infinity™ (ROI™) platform of membrane-based solutions for complex water and wastewater challenges, which combine Gradiant’s patented counterflow reverse osmosis (CFRO) technology with reverse osmosis and low-pressure membrane processes. ROI solutions enable customers around the world to achieve sustainability goals to reduce their water and carbon footprint, the company says.

This platform is complemented with AI-backed SmartOps™, an integrated digital platform for asset performance management to optimise and predict plant operations using historical and real-time process data, resulting in performance and cost efficiencies.

Prakash Govindan, Co-Founder and COO of the company, says most water solutions on the market are built for consistent liquid/solid feeds and work effectively when the input is in accordance with these specifications. When the feed changes, they often become ineffective, needing to be updated or changed out, which costs money and impacts the various processes on either side of the water treatment section.

“The machine-learning algorithms we use – neural networks and time-series algorithms – ensure we consistently optimise the operation of our solutions,” Govindan told IM. “These tools make sure we always use the right performance metrics and don’t lose efficiency in the face of variability.”

The algorithms cannot change the hardware built into the water treatment plant, but it can, for instance, change the speed of the pumps or blowers. “We call it balancing, which is all part of our IP portfolio,” Govindan said.

SmartOps is an integrated digital platform for asset performance management to optimise and predict plant operations using historical and real-time process data

For mining companies looking to employ water treatment tools at their operations, this results in Gradiant’s technology being able to concentrate metals to a higher degree than any other solution on the market, according to Govindan.

“We can concentrate an aqueous solution to the point where you can produce a solid material that miners can then process,” he said.

Considering desalination applications represent a significant portion of the company’s work to this point – through its CFRO process – the mining sector has already provided some wins.

The CFRO process enables remote inland desalination and water reuse that was not previously possible due to a lack of viable brine management solutions, according to Gradiant, concentrating brines to saturation for disposal or crystallisation while producing a purified product water stream for beneficial reuse.

One significant nickel miner in Australia with a brine stream is using this solution to recover large amounts of concentrate it can feed through to its captive processing plant to produce an end-use product.

“Gradiant’s technologies enable clients to recover more than 50% of the nickel and cobalt from leached brine – this stream would have otherwise been wasted without our solutions,” Gradiant said. “Overall, this was a client benefit of about 20% increase in nickel and cobalt production across the entire operation.”

When considered together with the energy savings (75%), freshwater savings (25%) and environmental benefits, Gradiant continues to see high interest from miners around the world to adopt its solutions, it says.

That is before even factoring in the other complementary benefits that come with using SmartOps.

“All our products benefit from in-built sensors that not only allow us to update the operating parameters based on the detected materials, but also carry out scheduled maintenance on the hardware using these algorithms,” Govindan said. “This allows us to carry out 30-40% less service intervals than many conventional suppliers as we only take the solution out of operation based on what the data is telling us.

“Not only this, but we also have complete oversight of these parameters from remote locations, meaning you can monitor the systems from remote operating centres and not remain on site after installation.”

With mines getting more remote and hiring local employees getting even harder with the well-documented skills shortages, Gradiant feels its solutions will continue to win miners over.

SensOre expands AI-based geophysics capacity with Intrepid Geophysics acquistion

Mineral exploration, Mining mergers and acquisitions, Mining software, , , , , , , , , , , ,

SensOre Ltd says it has reached an agreement to acquire Intrepid Geophysics, a provider of geophysics software and services headquartered in Melbourne, Australia.

The deal, valued at around A$5 million ($3.4 million) will be primarily funded through the issue of new fully paid ordinary SensOre shares.

Intrepid Geophysics’ advanced automated geophysical software and geoscience expertise complement SensOre’s existing suite of machine learning and artificial intelligence-based mineral exploration software products and technology offerings, it said, adding that Intrepid Geophysics’ large client base and strong cashflows were integral to its strategic assessment of the transaction.

SensOre Chief Executive Officer, Richard Taylor, said: “Acquiring Intrepid Geophysics is a major opportunity for us. Intrepid Geophysics’ deep geoscience and machine-learning expertise in geophysics complements SensOre’s geochemistry and economic geology focus for targeting in mineral exploration. Demand for advanced geophysics software is strong and deployable globally.

“Intrepid Geophysics’ years of product leadership, data collation and collaboration with government geological surveys will benefit SensOre’s data platform development and client service offerings. We are looking forward to integrating Intrepid Geophysics’ exceptional talent with our team of innovators.”

Intrepid Geophysics Managing Director, Dr Desmond Fitzgerald, added: “The combination of SensOre and Intrepid Geophysics will unlock growth opportunities in a strong market for high-level exploration targeting. We look forward to being part of a growing and exciting geoscience group.”

SensOre and Intrepid Geophysics completed a successful pilot project in Victoria in the June quarter, confirming the technological synergies and product complementarity between the two companies, it said. The results of the pilot are expected to be deployed with clients in the September quarter, focused on the Stawell and Ballarat gold corridors. There is strong interest from prospective clients within these corridors, according to SensOre.

SensOre says its strategy has been to organise all of Australia’s geoscience data within its proprietary data cube technology, with the acquisition of Intrepid Geophysics significantly advancing that capability.

The deal also expands SensOre’s mineral exploration technology sector presence and exposure, with the deal seeing the company acquire a second team of dedicated geoscience professionals with cross-over expertise in the oil and gas, groundwater, geothermal and the emerging hydrogen storage sectors.

“In acquiring Intrepid Geophysics, SensOre gains access to multiple new targeting and decision-based proprietary technologies and strategic decision-based services using 2.5 dimension Airborne Electro-Magnetic inversion technology, tensor gradient technology, geology from geophysics feature extraction, and service automated workflows,” the company added.

These technologies, SensOre says, fill a gap in its product suite by incorporating geophysics products that take exploration targeting from the macro-focused Prospectivity Modelling and Discriminant Predictive Targeting® approach into drill target delineation in three dimensions.

Intrepid Geophysics’ software portfolio includes:

  • Intrepid 3D – an airborne and ground geophysical data processing and interpretation package;
  • Moksha-EM – an airborne electromagnetic full waveform inversion data processing and interpretation package;
  • Argus – a 3D geological modelling package with a tightly integrated geophysical forward and inverse modelling capability;
  • JetStream II – a web-based, spatially searchable data catalogue that enables geoscientists to quickly assess the coverage, type and vintage of georeferenced spatial data held over any given area; and
  • Sea-g Marine Gravity – a fully featured marine gravity processing application powered by Intrepid Geophysics technology for on-cruise and post-cruise use.

Maptek machine learning trial points to future of mineral deposit modelling

Base metals, Energy minerals, Mineral exploration, Mining services, Mining software, , , , , , , ,

A trial of Maptek DomainMCF at an underground metals mine has concluded that machine learning will most likely become the preferred modelling method for mineral deposits, according to the software company.

DomainMCF is a platform that Maptek says will ‘put the geology back into geologists’, applying deep learning and big data computing methods to generate domain boundaries directly from drill hole sample data. Such rapid generation of resource models is a game changer for operations, according to Maptek.

In the trial of DomainMCF, geologists at the IGO Limited Nova-Bollinger underground mine in Western Australia trialled the solution’s machine-learning tools for modelling its resource.

Nova-Bollinger is 700 km due east of Perth, with the operation mining and processing nickel-copper-cobalt sulphide ores.

Traditional resource modelling is based on a drill hole database containing 99 lithological and 11 sulphide mineralisation logging codes. From this database the mine geologists use implicit modelling to interpret 22 different domains – 21 sulphide domains and one all-encompassing waste halo domain.

The block modelling process is undertaken annually by a team of geologists on site and in the Perth corporate office, taking several months to complete. The team trialled DomainMCF in parallel to the standard workflow as part of the 2020 resource update.

The required inputs for DomainMCF are a csv file comprising drill hole database or composite data, and an optional upper and lower surface to define the spatial extents of the region to be modelled. A block model parameter file details the origin of the block model, the 3D spatial extents and the block/sub-block dimensions.

Grade estimates are done on a 6 m x 6 m x 2 m block size and sub-blocks are permitted down to a quarter of the parent block size, according to Maptek.

During the trial, three primary tests were run using different versions of the drill hole file to explore the capabilities of the application and see how they compared to the existing workflow.

Test 1 provided DomainMCF with the drill hole composite file for the 22 different domains. Six chemical elements (Ni, Cu, Co, Fe, Mg and S) were provided to assist with the training phase of the machine-learning algorithm.

For Test 2, the data used in the first test was augmented with lithology coded data from the drill hole information outside the estimation boundary limits. The chemical variables were again used to help train the algorithm.

The purpose of this test was to determine if a combined sulphide and lithological model could be produced, and to see if giving DomainMCF additional information would impact the prediction of sulphide domains.

A hands-off approach was used in Test 3 to see how DomainMCF modelled a file containing only mineralisation codes and the grouped lithology used for Test 2. None of the domain codes from Test 1 were used.

Test 3 examined if the DomainMCF model was comparable with a manually-coded domain model and whether it was useful in the mineral resource estimate process.

IGO Senior Mine Geologist, Fletcher Pym, presented the trial results in a paper to the AusIMM International Mining Geology Conference 2022 in March.

“We were able to run Test 3, which was a relatively complicated model, in 45 minutes,” Pym said.

For Pym, Test 3 also showed that machine learning can produce very comprehensive models without the strong influence of a geologist.

Because machine learning made resource modelling much faster, senior staff had more time to focus on training less experienced core loggers. Improving the processes resulted in better quality drill hole logging, according to Maptek.

Pym added: “Machine learning will become particularly attractive if the process can not only model geological domains, but also return reliable grade estimates for mine planning across the full range of mineralisation styles.

DomainMCF model section

“Providing a well-understood confidence measure can assist in risk quantification of both geology and grade.”

The study, Maptek says, highlighted several advantages of machine learning:

  • The inputs required for machine-learning processing can be readily prepared in most resource modelling software;
    Machine learning modelling times are relatively short;
  • The pay-by-use business model is more cost-effective than maintaining implicit modelling software systems;
  • The machine learning model returns an objective measure of uncertainty in the geological model, which is likely to be useful in mineral resource classification and mining reconciliation work; and
  • Multiple different geological models can be prepared in parallel, meeting the JORC requirement to investigate ‘the effect, if any, of alternative interpretations on mineral resource estimation’.

Maptek Technical Lead for DomainMCF, Steve Sullivan, says he is excited at the potential of machine learning for revolutionising resource modelling.

“I’m amazed at the response – we are already seeing companies subscribe to DomainMCF for use in domain modelling for their 2022 resource reports,” he said.

“Machine learning works best when all the available data is presented, as shown in Test 3. The more data the better.

“The industry is struggling to find experienced personnel during the current mining boom, so embedding years of experience into smart systems helps get the job done on time and under budget.”

Maptek continues to work on proposed enhancements following feedback from industry trials of DomainMCF, with grade trend prediction added in the March 2022 release.

This is an edited version of an article that appeared in Maptek’s Forge newsletter.

Unico to help further commercialise SensOre’s mining exploration technology platform

Mineral exploration, Mining services, , , , , , ,

SensOre Ltd has executed an agreement with Unico, now part of global IT and business consulting firm CGI, to collaborate on the commercialisation of SensOre’s mining exploration technology platform.

The project represents an opportunity to expand SensOre’s current client services to a cloud-based SaaS (software as a service) product, it said.

SensOre’s technology was created to improve exploration success rates and performance, leveraging artificial intelligence and machine-learning technologies, specifically its Discriminant Predictive Targeting® (DPT®) workflow. SensOre collects all available geological information in a terrane and places it in a multi-dimensional hypercube or Data Cube, with its big-data approach allowing DPT predictive analytics to accurately predict known endowment and generate targets for further discovery, it says.

The contract with Unico is an important step in SensOre’s technology development roadmap, it said. It is envisaged that development and deployment of the proprietary platform will open the door to scale the company’s products and expand SensOre’s capacity to service mineral exploration companies in Australia and overseas, while reducing the company’s unit cost per engagement.

The project will involve using SensOre’s AI-driven technology to create a digital twin of the Earth’s surface, enhancing the way exploration companies identify and analyse mineral exploration targets, SensOre says.

SensOre Chief Executive Officer, Richard Taylor, said: “Our background is creating and deploying technology and software that generates exploration targets using AI. While we have previously focused on Western Australia, a key objective of the Unico partnership is to enhance the pace of our data expansion across the globe.

“This project is a massive opportunity to use our technology to build a model that we believe will be in demand from mining exploration companies around the world. We are excited about the prospect of a global relationship with CGI to help expand our technology offerings into new markets.”

Unico’s Director of Innovation, Evan Harridge, said: “Imagine being able to analyse an MRI scan of the Earth. SensOre’s AI-driven analysis tools will be able to see what is underground in great detail, similar to how medical imaging technology can accurately see inside the human body. This technology would enable exploration to be more targeted and limit the overall environmental impact.”

METS Ignited-backed program to fund OreFox’s AI exploration ambitions

Base metals, IoT, Mineral exploration, Mining finance, Mining services, Mining software, , , , , , , , , , , ,

Artificial intelligence-based geological data analytics company, OreFox, has successfully secured funding via the Queensland METS Collaborative Projects Fund to further its geological mapping efforts at the Mount Chalmers mine site in Queensland, Australia.

The Queensland METS Collaborative Projects Fund is delivered by METS Ignited with funding support from the Queensland Government as part of its Queensland Mining Equipment, Technology and Services 10 Year Roadmap and Action Plan.

OreFox is working with QMines Ltd and Ironbark Marketing to further commercialise its technology that, it says, solves a pressing challenge facing the Queensland mining sector of how to accelerate critical mineral deposit discovery and mine more efficiently.

The consortium led by OreFox will use cutting-edge proprietary AI technology to gain further knowledge of the Mount Chalmers deposit, located near Rockhampton. As a historical producing mine, there is significant potential to increase the known mineral endowment and add new economy mineral opportunities.

QMines has commenced exploration activities across the Mount Chalmers project, including an aggressive drilling program and is planning an airborne EM survey. QMines has reported an initial JORC 2012 inferred resource equal to 73,000 t of contained copper. Historical drilling at the site shows the mineralisation is open in multiple directions.

The consortium will carry out a multi-element geochemical sampling program focused on critical minerals at the mine site and other notable prospective sites around the historic mine site.

Data collected from this program, including assays from the current drilling program, will be processed by OreFox, using its machine and deep learning systems to gain further insights. QMines believes the discovery of further economic minerals will enable the company to move to development faster, thus bringing economic benefits to the region and the state.

QMines has recently acquired Traprock Resources and Rocky Copper, which both held significant tenements in the Mount Chalmers region. QMines has extended their tenement and landholding within the area since these acquisitions.

OreFox Chief Executive Officer, Warwick Anderson, said: “This project has the potential to increase exploration activity in Queensland, particularly for new economy minerals and could be applied to numerous other historical mines and deposits.

“The partnership between OreFox, QMines and Ironbark Marketing is anticipated to bring more regional jobs to Queensland and aid Queensland exploration frontiers.

“If we can prove the value of this project, then that opens the door to a significant export market for the processes we are developing.”

The OreFox project is one of five recipients of the Queensland METS Collaborative Projects Fund receiving a share in A$1 million ($733,978) to accelerate the commercialisation of technology into industry.

METS Ignited CEO, Adrian Beer, said the growth centre is backing the collaborative projects to fast-track the commercialisation of innovative technologies and provide value to both the local and global resources sector.

“METS Ignited is driving collaborative projects to accelerate commercial outcomes for the Australian economy and promote collaboration opportunities as part of a long-term strategy for growth,” he said.

“We are backing projects using technologies such as sensors, data analytics, machine learning, optimised x-rays, and solar energy that result in improvements in productivity, efficiency, safety and sustainability.

“The OreFox project is a great example of how AI and data science technologies can be harnessed to improve exploration and unlock the economic benefits of a historical mine site.”