AI’s big impact on automation & exploration in Australian mining

While the use of Artificial Intelligence (AI) and Machine Learning has been relatively slow in being taken up by some sectors of the Australian mining industry, a leading local advocate for the technology believes it is about to be taken up in a significant way.

WA Data Science Innovation Hub director Alex Jenkins says there is growing interest in the benefits AI can provide, particularly in providing automation and exploration solutions.

“I think that we will see increasing uses of Artificial Intelligence driving automation within the mining industry, as we will see in other places,” Mr Jenkins told the Australian Mining Review.

He said there are a number of economic drivers that will see mining companies continue to increase their adoption of AI into their automation operations.

“There is the potential to improve productivity and profitability and also safety,” Mr Jenkins said.

“For many of these processes, if we can remove the human element then we can make these operations much safer.

“I have already seen that trend increasing over the last decade, particularly in terms of remote operations and those trends continue.

“What is exciting to me is the use of Artificial Intelligence to drive more holistic operations, or optimisation of mine site and processing plant operations.

“I think what has happened in the past is that we have had very discrete parts of a mining operation that have been automated.

“That might be haulage, for example, and we have not moved very far in terms of automated haulage.

“Now, what is interesting to me is how this can be extended into the operation of a processing plant, so that we can optimise recovery and throughput, and create a global kind of optimisation of a mining operation.”

Mr Jenkins says an example of how AI can improve operations is in its use to assess the impact of a delay caused by unplanned downtime on the haulage fleet and in turn the productivity of the downstream operations.

“AI may be able to assess how other operations in a mine need to be adjusted,” he said.

“For our big iron ore miners and people with sophisticated supply chains, figuring out how to manage these operations holistically, using Artificial Intelligence I think, will drive significant outcomes.”

Predictive Maintenance a real opportunity

Mr Jenkins says an area where the advantages of AI and Machine Learning are being used is in Predictive Maintenance.

“Predictive Maintenance allows us to obtain more uptime out of our assets, by only taking them offline for maintenance when required,” he said.

“We are moving to what I would call just-in-time maintenance, which is probably the end goal there.

“I think there will also be big advantage through improved safety, by creating the ability to remove humans from potentially dangerous situations and the other big advantage, I think, is that we’ll see productivity increases.”

The prerequisites for including AI are a very solid set of data about existing operations.

“A lot of infrastructure has to be in place to ensure that the data related to operations is in place,” Mr Jenkins said.

“Without data, Artificial Intelligence will be useless.

“So that’s a key point that I would make, that without that underlying infrastructure and the technology to support it, the use of AI will be limited.

“In a way, it is better to introduce AI into an operating project as you will have the necessary data to make it worthwhile.

“I think in the future, with the increased uptake of AI and Machine Learning, we will see mine sites being run completely autonomously, but that is likely to still be decades off.”

Exploration a huge area of opportunity

Mr Jenkins says the use of AI in the exploration space is a real area of opportunity for local miners which will be worth billions and billions of dollars.

“For Australia, in particular, if we can develop new ways to approach exploration, there is significant upside here because it will help us find new assets and new resources,” he said.

“A lot of the techniques behind AI can be applied to examining statistics and geology in general to help us identify new assets.

“The resources that we are looking for now are more marginal than they used to be, deeper underground, harder to detect, you know, we have to use significantly more effort to find them.”

He pointed to the use of AI and Machine Learning in a recent “Gawler Challenge” prize winning entry.

The entry highlighted the use of people utilising Machine Learning and other techniques related to Artificial Intelligence to help bring together different data sources.

A data-driven methodology proposed systematically tested the relative importance of a large number of geological factors on known deposits.

The approach integrated a mineral systems approach that emphasised and analysed the processes operative in the passage of ore fluids and metals to their eventual deposition sites.

The entry identified essential ingredients in the different crustal levels of the theoretical mineral system as 100 mappable criteria in the exploration data.

These criteria were input into the machine learning algorithm and were used to make predictions about the less known areas.

Additionally, an analogue-driven methodology to test geological processes (geomechanical modelling) was applied that simulates rock failure and ore fluid flow at potential metal trap sites in the upper crust to hone district-scale targeting further.

“I think that this is definitely the future of mining exploration where we combine data from different sources to help us identify regions and drill drilling targets, expression,” Mr Jenkins said.

“Ultimately, these techniques are not a replacement for direct observation and drilling, like we do now, but what is interesting is what more we can infer about what is happening under the ground if we combine geochemistry, geophysical aerial survey, magnetic survey, and other detection techniques, in a kind of holistic way.”

The AI future is definitely an exciting one with its potential to help identify new resources and then improve their production.

NB: The WA Data Science Innovation Hub is a WA Government initiative, supported by Curtin University, aiming to ensure the State remains at the forefront of the digital revolution by increasing the uptake, education, training and awareness of data science in WA.