Addressing the topic of innovation in mining and mineral exploration was the order of the day at The Northern Miner’s second annual Progressive Mine Forum (PMF) held at the MaRS Discovery District in downtown Toronto in mid-October.
The first panel discussion, moderated by Northern Miner staff writer Richard Quarisa, focused on “Clean Mining” — looking at how far we’ve come over the past few decades, but also how far we still have to go.
“Clean mining for me is actually about just getting better,” Vale Canada vice-president of corporate affairs Cory McPhee said in front of a packed room of delegates.
As an example he talked about Vale’s aim to dismantle the superstack at its Sudbury operation by 2020. After completing a $1-billion investment in clean technology, McPhee said the company would no longer need it.
“We’re reducing emission to 1% of what they were when the superstack was built,” he said. “We’re making huge progress. I just don’t think it’s fast enough.”
Canada Mining Innovation Council CEO Carl Weatherell agreed. For him, mine design principles need to change, so that producing waste rock and tailings, and failing to recycle energy, are no longer an option. He said it starts with a change in thinking.
“Mining tends to talk to mining, with mining, for mining, about mining,” Weatherell said. “So we have to look at this clean question from a humanities perspective, not ours.”
While everyone on the panel agreed the industry should do more to facilitate clean mining, they diverged slightly when discussing what — or who — would spark that change.
“The consumer first will drive change,” McEwen Mining managing director of innovation Nathan Stubina said. “Government second, third, someone outside the industry, and fourth, the people in this room.”
He said there is no shortage of good ideas about clean mining and that a lot of the technologies are affordable. However, he agreed a shift in the way the industry thinks needs to happen. Too often, he said, the terms “research and development” and “innovation” get used interchangeably.
“All the money we do spend tends to be on tinkering on the R&D side,” he said.
He’d like to see diesel trucks, ball mills and tailings all disappear.
“If in my lifetime, if I can ban one of those things, I’ll be happy. If I ban all three, that’s even better.”
Jason Goodhand, e-Zn vice-president of business development, said that over the long-term, government regulations would determine the future of clean mining. He also argued using wind and solar power can help cut costs.
His company makes power-storage units that could hold massive amounts of electricity, driving down costs in remote mining locations. He says non-carbon-generated electricity can significantly reduce carbon emissions.
“When you think of clean, it’s: ‘What are you leaving behind?’ This energy is powered on diesel and fossil fuels, and what you’re leaving behind is a lot of carbon dioxide,” he said.
Whittle Consulting partner Raziel Zisman argued that change should come from within the industry. He says that while the industry makes mistakes, it tries hard to be better.
“We should and ought to be leaders,” Zisman said. “We are on the verge of an enormous revolution in our industry. It’s as simple as that. You can wash your hands and face at the end of a shift. You cannot wash your lungs.”
At Goldcorp, vice-president of corporate affairs John Mullally helps put these kinds of ideas into practice. He said Goldcorp is conscious of commercializing clean technologies when they’re available.
“In the case of Borden, our all-electric underground mine in Northern Ontario, we’ve had workers who have left and actually come back to the mine because the air quality is so much superior, versus a diesel environment,” he said.
Still, the industry can always do more.
“Change is not something that you ever fully accomplish,” McPhee said. “There’s no finish line. We have to continue to pursue it.”
The PMF roundtable on automation, moderated by Canadian Mining Journal editor-in-chief Alisha Hiyate, heard from mining experts that discussed the promise and limitations of the technology.
The panel agreed that there’s a surprisingly low level of automation in the mining industry — especially considering that modern-era automation was first introduced into various aspects of mining more than two decades ago.
“Apart from technology that’s been around for 20 years, like remote scoop operation, I think there’s a lot of low-hanging fruit out there, said Jason Cox, a consultant and executive vice-president of mine engineering at Roscoe Postle Associates.
“Maybe 10% at best of operating mines that we see have some sort of automated equipment beyond that scoop.”
Even so, Mikko Koivunen, a business line manager at Sandvik Mining and Rock Technologies in Canada, said he has detected a shift in customer attitudes.
“In the past two to three years, it’s becoming [more than] an interest — customers believe in the technology,” said Koivunen, whose purview includes mine automation and digital solutions. “There are so many references around the world — it’s more about how you adopt it, how you implement it in your own applications, rather than ‘does the technology work?’ It does.”
But current technology automates many pieces of the process — not the complete process.
Also, the complete mining process needs to be rethought for big productivity gains from automation, said Doug Morrison, CEO of the Centre for Excellence in Mining Innovation.
“For me, the biggest step is to move away from a technology platform that was based on human interactions and begin to design automated systems that were designed from scratch to be autonomous. If we simply automate the equipment that we have, we’re locking in all the inefficiencies that we built up and relied on for the last 35 years.”
Still, companies that implement existing technology have positive results.
Goldcorp, for example, runs autonomous drills at its Penasquito mine in Mexico, and has brought them to its Porcupine mine in Ontario, said Daniel Lucifora, the company’s manager of mine automation.
“We can actually uplift the base machine from its present status to 1.34 times that manned machine through the introduction of automation, which is a big improvement for us, and it will translate to a lot of improved productivity,” Lucifora said.
At Hecla Mining’s underground Casa Berardi gold mine in Quebec, where it has introduced a Sandvik autonomous truck, the miner has seen a 30% reduction in maintenance costs, less equipment damage, and higher use, productivity and safety, Koivunen said.
Several of the panellists noted that automation requires “change management.”
Walter Siggelkow, president of Hard-Line Solutions, said it is “the buy-in at every level,” from executives, operations and everyone else that is key.
Training is also crucial, Siggelkow said. “The breed of people required to maintain these systems does not exist, so training these people from the ground up is very important, and that’s one of the things that’s difficult in implementing these new systems into the mines.”
Holding onto trained personnel is another issue, and “that is one of the biggest hurdles that we’re running into today — both on the vendor side and from the mining side,” he said.
The day’s third panel, moderated by Northern Miner editor-in-chief John Cumming, tackled the topic of big data.
Humera Malik, founder and CEO of Canvass Analytics, defined big data in mining as “looking at all aspects of all the information that’s coming out from every aspect of the business, or every aspect of different processes that you run. Consider big data like the cloud [network], where you have everything you’re doing today getting converted into some format of information that is being collected and is being measured.”
Gordon Stothart, executive vice-president and chief operating officer of Iamgold, was asked how big data affects his company’s global operations on a practical, day-to-day basis.
He responded that “for us, it’s still a little bit in its infancy. Certainly since the 1970s or 1980s, on the processing side, we’ve used elements of artificial intelligence and process control and feedforward, and more static systems of that nature to handle the processing of ores. But certainly the level of data is getting pushed further upstream, all the time.”
He noted that the other generator of large amounts of data has been on the exploration side.
“We sort of have both ends of the process, but in-between it hasn’t been quite so much,” Stothart said. “Equipment has become more sophisticated, so haul trucks have a thousand and one sensors on them, and they’re monitoring all parts of the equipment. Some of that data is collected and transmitted, and very often isn’t used. I think that’s where we’re starting to see it.”
“Where I see it these days is combining a lot of the measurement systems and creating data to where I can look at it and see how the operations are doing on a daily basis, year-to-date, quarterly, annually — all that good information, and try to track that and overlay it with cost information, so we can understand how we’re doing economically, as well as production-wise.”
Shelby Yee, co-founder and CEO of RockMass Technologies, said there is a lot of opportunity with respect to using more sensors in operating mines, and that miners can look to the automotive industry for new technology that could be adapted to use in mining.
“A big opportunity that we’re moving towards is applied data,” Yee said. “We have very specific processes within mining. How can we use these sensors to collect very specific information that’s useful, that you can put on a dashboard and immediately view without having to do a lot of manipulation and time-consuming work with the original data that they provide?”
Glenn Mullan, president of the Prospectors & Developers Association of Canada (PDAC), and president and CEO at Golden Valley Mines, noted that with respect to big data, the junior mining companies that make up most of PDAC members are interested in “manipulating large data sets in geoscience and trying to find something of value that has either been overlooked or hasn’t been explored before. So large data sets, but not on brownfields or mine sites, but on huge tracts of land. That’s where the opportunities lie — in the earliest exploration.”
Talia Dabby, director in the Technology Consulting, and People and Organization practices at PwC, said the composition of the mining workforce is changing as big data grows in importance, and discussed ways companies can refine their hiring and retention strategies to keep a cutting-edge workforce.
“It’s a theme that’s come up through all the presentations and panels: how big data and technology generally are really changing the people that we need, and people that are going to be supporting us from a mining perspective in any organization.”
She noted that the financial services industry over the last three years has hired more IT roles than financial roles, “and that’s not going to be any different from a mining perspective … but it’s really about up-skilling those operations, and the operational people that we have today.”
The panel was asked what they consider low-hanging fruit in terms of big data solutions to mining challenges.
Stothart said he thought it would be “understanding your orebodies better, so you can truly optimize them better. We have the concept of zero-waste or clean mining — obviously, ideally you’d like to extract only the thing of value and nothing more, and walk away with just that element. We’re still a ways from that, but that all goes to optimization.
“Every time you touch the rock in whatever form that is, from your initial drill hole through to the last time you pump it out to your tailings pond — the more useful information you can extract from that, the more opportunity really comes to optimize that process.”
Malik commented that big data can be used in several ways.
“You have to determine a path, and whatever you can measure, and can analyze, and derive results from, that’s how we look at big data,” she said. “Then you look at each process, and the first thing you look at is, first you’re collecting data, and now you can visualize things. But now can you start to optimize it? And from there you begin the journey towards, how can we automate it? So that’s the way we look at it … if you have data, can you collect it into rich information that can be used?
“You don’t go immediately into automation, because first you have to understand — it’s called ‘state modelling’. From there you go into optimization, and from there you go into automation. And then from automation there are things that can go into autonomous operations, where you can make inferences where no one has to go in and make certain changes.
“And we start to make certain changes in autonomous operations because in my opinion, humans are made for much higher-level functions. Just as we are creating autonomous cars, can we start to make autonomous operations as well, so that we are more engineers and IT folks, and more people who are building those functions into autonomous operations? In a perfect world, that’s the use of data.”
“We’re sort of big-hammer people,” Stothart added. “We tend to design our processes around the least common denominator. So you’ll design a mill around the hardest kind of rock you’ll ever see. But the mills today can be a little more sophisticated — you can feed more rock if it’s softer, and slow the mill down to get more efficient grinding on a certain type of rock.
“Historically, we haven’t had that subtlety, but [now we’re] using feedforward data to feed the information from blast-hole drilling, from mining, from fragmentation. With online analysis as it’s coming into the system, you can really start to tune the equipment so you’re putting efficient energy into the grinding process — which is the biggest consumer of third-party energy that any mine uses.”
Yee noted that a mine can have advanced processes, especially in milling, but at this same operation “you can have people recording data on pen and paper. These are very skilled engineers you’re paying a lot of money … and the time it takes to do the collection and transfer adds up to quite a bit. So if we can streamline this data-collection process and improve the data quality — whether it’s through consistency, through precision — to input into this model, that’s pretty low fruit, to just get away from pen and paper.”
— Look to The Northern Miner Podcast and YouTube channel in the coming weeks for full audio and video of the panel sessions.