The rise of autonomous mine haulage

I couldn’t write about mining and autonomy without delving into the field of autonomous haulage, so, I got in touch with Richard Price and Ben Miller.

Both are mining engineers with extensive experience of autonomous haulage systems (AHSs). The pair originally met at the International Intercollegiate Mining Games in Reno, Nevada, and have collaborated on multiple projects over the past 20 years. Miller is also currently completing a doctorate which focuses on functional safety for AHSs…

Basically, if there’s anything you need to know about AHSs, these two are your guys.

But, before we get into the juicy stuff, let’s start with the basics. AHSs have been around for a while, so why aren’t more mines using them?

Approaching the chasm

In his 1991 book, Crossing the Chasm, Geoffrey A. Moore suggested a revision to the well-known technology adoption lifecycle diagram (you know the one, Gartner based its famous Hypecycle on it) developed by Everett Rogers as part of his theory on the diffusion of innovations.

In the original, the bell curve is divided into five successive stages (innovators, early adopters, early majority, late majority and laggards) which describe the different groups making use of a new technology over time.

In Moore’s revised version, which focuses on marketing disruptive technologies, he suggests that there is a ‘chasm’ which exists between the early adoption and early majority stages. The chasm illustrates the difficulty in transitioning between these two phases… Bear with me, you’ll see where this is going.

“If you look at Moore’s curve… If you take the number of converted 200-300 ton class trucks, divided by the total number of those trucks in the world, we’re currently about halfway through the early adopters stage,” explained Miller, when I asked about AHS adoption across the mining industry.

“Companies like Codelco in Chile were the innovators, and then FMG, Rio Tinto and BHP in the Pilbara were the first early adopters. The Codelco one never transitioned into further deployments but, in the Pilbara, autonomous technology has been readily adopted. The three big players aren’t opening any new manned mines and they’re actively converting manned over to autonomous.

“And it’s not just iron ore, we’re starting to see this leap in other parts of the industry too. Highland Valley Copper is doing AHS trials in Canada, oil sands are all investigating the technology and some are incrementally converting their fleets.”

The beauty of this technology is that AHSs aren’t exposed to the commodity; bar oil sands deployments, the trucks, software and settings used are broadly the same in iron ore, nickel, copper operations… whatever, making them a widely applicable tool for driving cost improvements.

New deployments seem to fall mainly into two categories: operational mines that are using the technology to pull their prices down, and new projects – greenfield or brownfield – where autonomous technology is being used to drive the sizzle a little bit.

“If you look at, for instance, disseminated gold deposits where haulage is sometimes 75% of the mining cost, if you can drive down the haulage cost, it changes the pit shell and reserves substantially,” Miller said. “It’s a way of converting additional resource over to reserve by using, what is today now proven technology.”

That’s what this is all about; by converting to autonomous haulage, mines can increase their fleet utilisation and create an average 20-30% improvement in hourly operating costs compared to running a manned fleet.

“As a mine engineer, if you came in with a plan that changed the productivity of a mine by a couple of percent, you’re probably getting a promotion and a raise,” said Miller. “We’re talking about improvements in the tens of percent range with autonomous trucks. You might not get a plus 30% right off the gate; you might only get a couple of percent, but that’s how all industry improves, by taking these little bites.”

Making the most of autonomy

20-30% is a big number, but could we get more from AHS deployments? Are miners putting in the legwork and looking at the processes and systems that are needed behind these implementations to get the most out of them?

“Not yet,” said Price. “With any technology deployment in any industry, there are always companies that embrace the technology on a holistic basis, and there are those that do not.

“Two companies can roll out the same system on the same truck in the same region of the world, one embraces the technology fully and changes its culture to make the most of that system, and the other tries to fit the technology into their current way of working. Which one do you think will get the most out of the technology?”

Miller was more diplomatic: “I think groups like FMG are very public about how hard they’re working at it. Even companies that are quieter, like BHP, they’re working behind the scenes too. But there are examples that show FMG likely is having better success than others, and I think that’s because their culture is more accepting of that technology change than perhaps the other big boys are.”

Miller gave the example of refuelling and how FMG has changed its way of working to accommodate the needs of autonomous trucks versus manned units.

“In a traditional manned mine, refuelling occurs on the shift change” he explained. “Initially, at many autonomous deployments, fuelling happened on shift change again, and all the robots lined up at the go line, had an inspection, refuelled and went back out. It was soon realised that process wasn’t necessary.

“So, instead, the trucks now come into a refuelling bay within the autonomous operating zone when they’re under 15% fuel, generally speaking once per day. They get fuelled, they get their daily inspection and they get kicked back out in the fleet in a continuous process instead of the batch process that you would see for a manned fleet.

“Each time a mine lifts a constraint like that, the next constraint is probably harder to find and harder to lift. And there’s some level of diminishing return.”

Learning to trust the technology

“But do we think we, as an industry, are using these systems to the full potential?” Price asked. “The answer is no, and that’s partly because of mine design constraints.

“For instance, haul roads are currently 10% wider than they need to be. Initially this was because people were worried that the trucks would run into each other, but the fact of the matter is the first autonomous truck that I saw in 1996 was running wheel ruts in the road.

“Despite this, nearly 25 years down the line, there still exists some level of concern. With an AHS, haul roads could be made 20% to 40% narrower saving companies a huge chunk of money, and they will be narrower in the future.”

Of course, part of reducing that 10% has to do with sensor technology; the safety systems onboard autonomous trucks are less prone to falsely identifying objects that are in the way if you widen the roads.

“Now, we’re talking about the technical aspects of squeezing more out of the systems, and there are groups and companies, including myself, working on solutions to these issues,” explained Miller. “There are technical sensor engineers and mechanical engineers who work for the OEMs that are trying to find solutions, and external manufacturers that are making better sensors.”

The autonomous truck industry leans heavily on the automotive industry for its sensors and for the driving technology. Companies like Delphi that produce the radars used in the automatic emergency braking technology featured on most new cars; those sensors are also used in front radar arrays for autonomous Cat and Komatsu trucks.

“As those companies push car sensor technology further, they’re going to pull the mining industry along with them,” added Miller. “The automotive industry is investing billions of dollars in developing this technology.”

Price agreed: “Last year the number of motorcars sold globally was around 80 million. The number of off-highway haul trucks was 7,000. It’s a totally different scale.”

Safety comes first

Another hurdle the mining industry is currently tackling is functional safety.

“The GMG is pushing heavily on functional safety,” Miller said. “The purpose of this is to define methods to safely deploy these systems and to certify that they’re safe. That will allow mining regulators to approve them more readily, rather than treating each one as a one-off research project.

“We’re going to see mine sites within countries which follow international standards, like ISO, IEC, are probably going to find it easier to deploy AHSs than those in places like the US, which follows its own set of rules. US standards are much more inflexible than what you’d see in Australia, Europe or Canada.”

“There are AHSs operating in the Australian states of New South Wales and Queensland without any regulation whatsoever,” added Price. “The same in Canada. The province of Alberta does have some level of regulation, but it’s not mandatory. It’s the same in Western Australia.

“There are plenty of jurisdictions in the world that have AHSs operating now, whereby regulation is not going to be present at all for another decade or so. That just is what it is.”

It’s interesting that, yet again, the conversation has come back around to people and how their involvement is going to be a deciding factor in how far we can go with mining technology.

But I digress…

From leap of faith to common sense

To bring us back round to Moore and his eponymous bell curve, I asked: when will the mining industry cross that theoretical chasm? When will autonomous mine trucks become the norm rather than the exception?

Price was practical in his response: “I don’t know. Obviously, I don’t have a crystal ball, but we’re definitely past the innovators stage and the first early adopters. We’re starting to see fast followers. I think in the next 5-10 years, we’re going to be in the early majority stage. And, in the next 15 years, the late majority.

“That’s my take on it. There will come a point in time when you will call up Caterpillar or Komatsu and say ‘hey mate, can I buy a truck? And they’ll say, yes, no worries. Quick question, do you want a cab on it or not? It’ll cost more money to have a truck with a cab.’

“Of course it will cost more to have a truck with a cab, because a cab, from the perspective of the autonomous truck, is entirely superfluous. In the future, there’s absolutely zero reason as to why a truck needs a cab, apart from the fact that a human being wants one.”

Miller chipped in: “I think what we’re going to see is that the cost to operate gets driven down by mines that are putting AHSs in and that, to survive, other mines are going to have to drive their costs down too. I don’t know where else you’re going to find a 30% reduction in haulage costs.”

Price agreed: “The technology costs, LiDAR etc., those are all plummeting, but I think widespread adoption is going to be driven by the traditional mining situation; if you have a paradigm shift in cost, which this is, and your mine does not adopt that change, then you’re going to be pushed out of the market. You just can’t stay in the market if you have a big discrepancy in operating cost.

“Yes. Or you might have some market condition, like for instance, our current worldwide situation, that drives metal demand down,” added Miller. “Mines that are not producing at a low operating cost are going to slowly die out. Either die out or adopt a change to reduce cost.

“It’s nothing new. It’s the same thing that’s been going on for a couple of hundred years. That’s what’s going to drive the change.”

“When trucks become more expensive to buy because someone wants a cab, then clearly, that’s the sign we’re at that tipping point,” Price concluded.

You heard it here first folks.