What you need to know:
- Artificial intelligence is helping the mining industry reach genuine autonomous operation.
- Sophisticated data collection and analysis is helping the sector make better decisions about how to run operations and process minerals.
- Advanced algorithms and data are improving exploration and creating better forecasts.
- The resource sector should be doing more to make the most of these opportunities.
The 21st century has been a time of great change in the resources sector. In 2008, Rio Tinto launched its Mine of the Future vision1, which centred on introducing autonomous systems into its Western Australian iron ore operations such as driverless trucks and trains and remote-controlled drills.2
Along with automation, smart mining now also encompasses digitisation, robotics, artificial intelligence and more innovative approaches to one of humanity’s oldest industries. Here are three of the biggest ideas that industry experts believe will shape smart mining in the years to come.
1. Artificial intelligence
“I actually don't think we have autonomous operation in a lot of larger operations,” says CEO of digital transformation consultancy idoba Sarah Coleman, who has 15 years’ experience in the resources sector.
“To an extent, we still have remote control operations because we haven't changed the way that we're actually working to embrace a fully autonomous operation.”
Artificial intelligence (AI) will be the path the mining industry takes to reach genuine autonomy, she believes. It will unite a smart mine’s digital services, driving decision making.
“Autonomy, connected with the digital smart mine and the optimisation layer across the top, will allow us to deal with the huge amount of complexity that’s coming,” Coleman says.
“I think we'll start to see people leveraging the data that they have to create more specific AI transformers that then look at how we solve complex problems in ways that we haven't been able to.”
And we’re only at the beginning of the transformation. Generative AI is still new technology, and Coleman believes that, in coming years, it will be applied with greater specificity to different aspects of mining — processing, for instance, or optimisation.
Chris Gibbs Stewart, CEO of Austmine, the leading industry association for the Australian mining equipment, technology and services sector, agrees that AI is increasingly applied to mining strategies but says not a lot of companies have AI strategies in place.
“A few do, and it helps them be more efficient in the processing plant,” she says.
“Because we have so much new technology — such as AI and machine learning — that we don't know a whole lot about, we really need to be aligned as an industry.”
2. Data analysis
Getting the best out of AI means using data effectively, and Coleman says the industry still has a way to go.
“I think we still don't have great data capture mechanisms,” she says. “Really rigorous AI needs quite specific tools and large data sets to train it.”
Stewart points to sensors as a technological advance currently transforming the way mines use data. These let workers collect data and analyse it in real time, allowing their operations to be more adaptable and flexible.
“It gives us so much more data about what's going on at the mine site and enables us to see the whole process end to end — from when the mineral is coming out of the ground to when it's going into the processing plant,” she says.
“Along the way, we can make better decisions about how we are going to run our operations and process the minerals.”
3. Prospecting and prediction
Having collected and analysed that data, the next opportunity lies in how companies use it. Coleman says low quality was the first hurdle preventing idoba from getting the most out of the vast amounts of data collected by it and its parent company Perenti.
The traditional approach would have been to physically collect more data: send engineers to a site and gather information.
“Then one of our quite amazing data scientists looked at it and said, if you think about it, the ore body behaves a certain way based on the geology of the Earth,” Coleman says. “Which means that the mine behaves a certain way, which means the equipment and the people in it behave a certain way,” Coleman explains.
With this insight in mind, the team looked for an algorithm that could improve the quality of its data set. The solution they found was a similar one to the method by which Facebook predicts the spending patterns of its users.
“That algorithm itself wasn't scalable, so we started using behavioural-based, deep learning neural network algorithms, and created some amazing algorithms that could take the data and start forecasting future performance,” Coleman says.
Stewart adds that exploration is also improving due to smart mining advances.
“As we know, demand is going to double and triple in the next 10 years, but we don't have enough mines — or even know where those tier-one deposits are,” she says.
“Using AI, people are finding those deposits more quickly, and I think that's only going to accelerate.”
A smarter mining industry
These advances matter because the future requires a mining sector that is smarter, more nimble and capable of meeting community expectations.
“Smart mining has enabled mining to become more efficient and more productive,” Stewart says.
“It has enabled us to go deeper, to extract those minerals that are harder to extract because they're no longer on the surface. It's allowed us to have deeper underground mines and has helped us to process minerals in a way that is much more efficient. It saves energy and gives us more control of the whole operation.”
References
- Rio Tinto, History
- Sustainable Minerals Institute, University of Queensland, Autonomous and remote operation technologies in Australian mining