Payload System Calibration

Sixty percent. That's roughly how many haul-truck loads read out of compliance the first time I got a clean look at the payload data at Kennecott Utah Copper, the world's largest open-pit copper mine. It raised some eyebrows.

I was an operations engineer at Rio Tinto's Kennecott operation from 2013 to 2015, and most of my job came down to one move repeated in different rooms. Find the number the operation runs on, figure out whether it's true, and put it in front of the person who can act on it.

The payload work didn't start as a data-quality project. I was analyzing how payload affects hauling speed, part of a truck evaluation for an upcoming pushback, when the compliance numbers came back absurd. That mattered because overloading a haul truck is not a paperwork problem. A machine that size carrying more than it was engineered for is a safety hazard, the kind that ends in catastrophic mechanical failure and rock coming off the load onto roads other equipment shares. And every extra ton grinds down components that cost a fortune to replace.

I started talking to everyone from operations to maintenance about the issue, and I kept hearing the same thing. We've been doing this for a long time, and we know we're not seeing overloading problems. Finally someone on the maintenance team dropped the answer right at my feet. The onboard payload systems require regular calibration, and it hadn't been kept up for years. Partly that was because the mine had recently gone through a major incident with the Manefay slide and priorities had shifted. But the primary reason was that the person in charge of the calibration program had been reassigned, there was no obvious successor, and so no one picked it up.

So I started calling vendors to see what a calibration study would cost, and the numbers that came back were staggering. Luckily, thanks to some cost-reduction efforts, our new head of technical services had a rule I've used ever since. Whenever you get a price, ask for a better price three times. I called the vendor the site had used last and got them down 30% over the next couple of calls, then decided to find a competing bid to force their hand a third time. But how does one find haul-truck scale calibration companies? In a stroke of genius I called the scale manufacturer to ask if they knew any vendors, and got a surprising answer. "We can come out and run the study ourselves." They would do volumetric work and additional reporting on top of it. How much extra did that cost? Nothing. Their quote came in at 80% of our lowest offer yet, and they could be there lickety-split. That's the company we went with, and yes, I haggled them down three times too. We ended up getting all of the work done for a 90% reduction from the original quote.

After the recalibration, out-of-compliance loads dropped 51% and measured payloads rose about 3% across the mine over the next four months. Some of that 51% was bad data getting cleaned up, and some was real overloading getting corrected. Both had been invisible.

I have two big takeaways here. First, don't expect something to be managed if no one is assigned to it. Second, always have a clear line of succession, because transitions come faster than you'd like and seldom go well. Cross-training could have saved this whole headache, and I've seen the same issue at mine after mine. If you pick up a new technology it needs a champion, and that champion needs an apprentice following along behind them. I remember heading to the SME conference while I was at Kennecott, and the procurement team asked me to get quotes on a fragmentation camera system for our shovels. I ran into the guys who make these systems and told them where I worked. They said great, how's your system working? System? Yeah, we installed a system on two of your shovels a few years back. Our shovels already had these systems onboard. The issue was that the people who had been looking at the data and maintaining them no longer were.

That interaction raised so many questions, but I'll save those for a future post.

P.S. Always ask for a better price three times. It really does work like a charm, and as long as you don't burn out your vendor relationship, the worst that happens is a simple no.