Whether as chief data scientists or machine learning project leaders, women in Canada are using artificial intelligence tools and techniques to revolutionize the country’s mining industry.

But even by harnessing the oft-touted power of data analysis and predictive technologies, AI is no “magic bullet.”

As described in recent mining industry conferences and meetings of AI practitioners, women in the industry have discussed how Canadian mining firms are using artificial intelligence and machine learning to analyze findings from geological surveys looking for new extraction opportunities, among other things. AI is being used in some cases to direct and control autonomous vehicles used (often in dangerous settings) in the mining process and to monitor environmental, health and safety conditions in and around a mine.

Women in Artificial Intelligence & Machine Learning (WinAI&ML) is an industry group based in Montréal citing some 300 members. Working in partnership with another tech entity, Women in AI, the goals of the group are to learn from one another, to share current knowledge and anticipate future developments in the field.

Mining is seen by many as a natural fit for AI: it is a data-driven industry where the end product is a very specific and definable asset. Data can very easily describe the scientific, environmental and historical parameters of a planned or existing mining activity.

And because the data gathered from mining is more scientific and technical — less personal — than that gathered by, say, online shopping sites, there are fewer concerns about data privacy and the very real ethical implications of data collection and analysis identified in other AI-related activities.

The Montréal chapter of Women in Mining Canada also held an event looking at artificial intelligence and the mining industry, hosted at office of law firm Dentons. WIM Montréal president Kimberly Darlington of Refined Substance, a marketing and communications firm for the mining industry, chaired the event, and Dentons partner Mira Gauvin moderated the AI discussion among panelists representing leading Canadian mining firms.

Sarane Sterckx, for example, is Project Manager at Goldspot Discoveries Inc. (the company has offices in Montréal and Toronto), with international education and industry experience in the field. Goldspot uses AI to put historically underutilized data to better use as a way to fully understand the potential of a particular property or resource.

As well, GoldSpot has developed its own artificial intelligence-driven trading platform to help reduce its own capital risk while increasing success rates in resource investment and exploration.

Another leading Canadian mining company, Rio Tinto, is making use of artificial intelligence, machine learning and autonomous vehicles as it incorporates more technological solutions into its global operations.

Maiko Sell also participated in the panel discussion; she’s an Earth Data Scientist with Rio Tinto. The company’s mining control centres look a lot like the control room for space mission. Its centre in Saquenay, QC, is an example where video screens and computer consoles help operators manage a fleet of driverless trucks, rock crushers, autonomous trains and other heavy mining equipment that can be thousands of kilometres away.

The company’s Mine Automation System (or MAS) is already operating at nearly all the company’s locations. It combines real-time data from various sources to deliver operational insights and opportunities for increased efficiencies, the company describes, working from a huge fusion database with artificial intelligence and machine learning capabilities built in.

Guy Desharnais, Director of Mineral Resource Evaluation, Osisko Gold Royalties Ltd.; and Caitrin Armstrong, Machine Learning Engineer at Aifred Health and MSc candidate in Computer Science in the Network Dynamics Lab at McGill University, were also panel participants.

Despite the many opportunities for increased efficiency in the mining sector from continued deployment of AI resources, the technology is not a magic bullet, reports from the meeting underscored.

As it does in many situations, the successful implementation of AI is a multidisciplinary task. Buy-in from data scientists, geologists, computer engineers, economists and senior management is needed.

AI tools must be carefully tested before deployment and user orientation is a crucial part of that. And while privacy concerns may be lessened, legal implications are not: who is responsible or accountable should an AI program not go as planned? Is it the programmer or software provider? A third-party provider? The mining company itself?

While the uses of AI in mining and resource extraction are bringing benefits today, industry leaders both female and male know new opportunities still need to be carefully evaluated, best practices still need to be widely adopted and perhaps legal or technical issues that haven’t yet been encountered may soon be uncovered.

Kind of like heading into a mine shaft.

 

# # #

 

-30-