Tom Cawley has been appointed mining sector lead for the AI Accelerator Cooperative Research Centre, bringing MaxMine closer to Australia’s national efforts to build artificial intelligence tools.
The appointment comes as MaxMine rolls out its machine learning system for load and dump classification across Australian mining customers including Glencore, NRW Holdings and Macmahon. The system has been fully operational at the customer site for six months.
The Australian mining technology company said the software was built using more than 14 million hours of labeled operational data. It is designed to automate part of production tracking by categorizing load and dump events, a process that can impact how a mine’s production is recorded and analyzed.
According to MaxMine, the system has reduced the workload of field teams by reducing missed or erroneous loads and improving production tracking accuracy, especially in more complex operational scenarios. Our internal development processes also benefit from structured data and machine learning pipelines, allowing our models to cover a wide range of edge cases without custom work for each site.
AI in mining
Cawley’s new role puts MaxMine executives at the center of a broader push to expand domestic development of artificial intelligence in industry. The AI Accelerator Cooperative Research Center was established to help Australia build more of its own AI tools for sectors such as mining, rather than relying on overseas providers.
The move reflects broader challenges in the resources industry, where companies have invested in digital systems for years but often struggle to move projects from trials to day-to-day operations. Data quality, system integration, and site-to-site variability all limit widespread adoption.
MaxMine CEO Sean Mitchell linked the company’s latest development to the quality of the data used for training.
“The successful implementation of this new machine learning system confirms what we have been observing across industries: Organizations that are successful with AI are those with the highest quality datasets. As the adoption of AI accelerates in mining and other critical industry sectors, high-fidelity ground truth data will be essential to deliver accurate results, improve operational visibility, and enable faster, more informed decision-making,” said Mitchell.
Data quality remains at the heart of many industrial AI projects. In mining, operating conditions vary widely between fleets, pits, and sites, making it difficult to build models that continue to operate after moving from a controlled pilot to an actual production setting.
MaxMine says its latest model runs in a private environment and leverages high-resolution data throughout the loading and handling operations. This combination, the company says, allows the model to deliver stable results across a variety of assets and site types.
sector push
Coley said the industry still faces a gap between investment and real results at scale.
“Despite increased investment in AI, industries such as mining still struggle to move beyond pilot efforts to achieve operational outcomes at scale. Gartner estimates that 60% of AI projects rely on AI-enabled data. At MaxMine, we have demonstrated that Australia is capable of developing advanced AI tools that work effectively at scale in the mining industry. I hope that my role at Accelerator CRC will contribute to this, “driving further innovation across the sector and contributing to Australia’s competitiveness in key minerals markets,” Mr Coley said.
The reference to critical minerals highlights the strategic importance of digital systems in extractive industries. Australia has sought to strengthen its position not only as a supplier of mineral resources, but also as a developer of the technologies used to manage and process mineral resources.
For mining contractors and producers, classification systems serve a practical day-to-day function. Errors in load and dump records can impact reported production, shift analysis, and coordination of mine planning and actual activities. Automating that work can reduce manual intervention by field teams, but adoption depends on operators being able to trust the output under real-world conditions.
Professor Anton van den Hengel, chief scientist at the Australian Machine Learning Institute and interim chief executive officer of the AI Accelerator CRC, said the MaxMine dataset was distinctive.
“It’s rare to be able to deploy a model with such accuracy across such a wide variety of assets and sites. MaxMine’s ability to do this demonstrates a uniquely rich, accurate, and human error-free data set combined with a long-term, multi-site, multi-machine training data set,” said Van Den Hengel.
MaxMine’s customer base includes Australian operators and mining groups with international operations, including NRW Holdings, Macmahon, First Quantum Minerals and Kinross Gold.
