12 AUGUST 2019
Once viewed as optional, fleet management systems are now mission critical to mines. Jigsaw was the go-to FMS for many mines and is now known as OP Pro, a new name and greatly improved solution that is integral to Hexagon's end-to-end life-of-mine technology. Hexagon's Sergio Blacutt and Jonathan Olson explain.
JS: Thanks for tuning in. Hi, I'm Jose Sanchez from HxGN Radio. Once viewed as an optional, fleet management systems are now mission critical to mines. Jigsaw was the go-to FMS for many mines and is now known as OP Pro, a new name and greatly improved solution that is integral to Hexagon's end-to-end life-of-mine technology. Today we have Sergio Blacutt and Jonathan Olson from Hexagon's Mining division with us to discuss OP Pro.
We appreciate you guys taking time, and thank you for joining us, gentlemen.
SB: Thank you.
JO: Thank you.
JS: Before we discuss OP Pro, let's talk about its predecessor, Jigsaw, fleet management solution you helped create. Tell us how that happened.
SB: This is Sergio. I'll say, well, where we started Jigsaw, actually, several years after I left Modular, like, three years after having left another successful venture, I was thinking what to do? One day I decided to call Jon and we got together for lunch, and we talked about starting a new company that would meet the current demands of the mining industry. We both have worked with Modular before, so we knew where the industry was, how the systems were operating, and we saw a huge demand for more modern systems. So, I talked to Jon about that, so Jon was very happy and very, very receptive to this idea. So that's how we started Jigsaw Technologies. And Jon, I'm sure you can elaborate a little more about how this happened.
JO: Yes. My name's Jon Olson. I began my career in mine management back in 1981, when I graduated from the University of Arizona. I was one of the founders of Modular Mining Systems, and I developed all of their embedded systems that run on all of the beta processors in the trucks, as well as the central computing-optimization software that use linear-programming techniques to optimize flow patterns of trucks in the mine. Back then, we were using a VAX-11/780 computer that had about one-thousandth of the capabilities of your smartphone. The VAX-11/780 cost a million dollars, and it talked over 1,200-baud radios to all of the equipment out in the field. Nonetheless, with this technology, we developed a very workable mine-optimization solution where our operators were sent small text messages to the remote field panels that indicated the next shovel assignment or the next dump assignment. This was in the era before GPS, so you couldn't actually track where a truck was going all the time. We only knew where the truck was when he would arrive within transmitting distance of a low-powered RF beacon.
I stayed at Modular for 22 years, and I left Modular in August of 2003 to start Jigsaw Technologies with Sergio Blacutt. The idea that we had in Jigsaw Technologies was to develop a more modern fleet-management system that used distributed database technologies to be able to take a lot of the intelligence and the GPS monitoring and be able to continuously gather information onboard the vehicle, whereas previous systems were all based on a central computer, and the central computer only knew sort of vaguely where the trucks were going at any given time. By distributing the intelligence onboard the vehicles, the vehicles know continuously, second by second, exactly where the truck is on the road, when he's going to arrive, what he's doing. And by that, we could develop much more intelligence systems that integrated automated haul cycles, be able to detect when buckets dump in the back of the truck, and implement technologies that you couldn't do using the previous centralized computer design.
JS: Excellent. That's great. Fleet-management systems have gone from something nice to have to expected and even mission critical. Why do you guys think that is?
SB: Well, as Jon was saying before, when we started developing Jigsaw, we saw really the need for continuity on the operations. Before, the reliability or the equipment wasn't good, communications weren't that good, so communications were critical. Right? So, we wanted to make sure that the vehicle would store data at all times, regardless whether they'd have communications with the central computer or not. So that's kind of what we did.
And, well, nowadays, really what has been happening is that mining operations operate in a very competitive environment, where they have to keep up with their competitors. They have to be competitive in how much they spent producing their materials, the minerals or coal, whatever it may be, diamonds. So that is what's driving operations to utilize these fleet-management systems.
It's critical that in your operations you're monitoring what the equipment is doing. You need algorithms that can control the flow rate of material, the equipment, what the equipment is doing, how are you feeding the different dump rates, and stuff like that. And really the most important thing is that you can provide this information in real time, and you know that the computer can really react to changes. I mean, the algorithm can react to changes as they happen in the operation. That's the only way you can remain competitive. A human cannot really do that. So that is really what has happened, that they became from like, oh yeah, it's nice to have my little report that a computer generated, to a must-have application that really drives your mining operation.
JO: Another thing I would like to expand on is back when fleet-management systems were nice to have, they were also extremely expensive. You needed to have a huge investment in a central computer room with air conditioning and install one-plus-million-dollar central computer, and then you had repeaters to install, and you had very low-speed networks. Computing technology has really increased by four to five orders of magnitude in this time. Whereas it used to require a one-million-dollar system to execute one million instructions per second. Now you have a five-hundred-dollar system in your pocket that actually executes four billion instructions per second.
But at the same time, costs—ultimately all of this comes down to cost—costs for mining equipment, workers, fuel, maintenance, and all the costs that mining operations will incur have increased higher than inflation. So whereas computer technology to actually implement a fleet-management system is much more available and much less expensive, the things that we're managing are much more expensive. So it only makes economic sense to install a fleet-management system.
JS: That makes absolute sense. And I think one of the things that's really exciting to see is that the technology you guys have started, you guys have built, is now part of every mine, and every miner needs to consider that now it's no longer just a commodity.
You guys were instrumental in Jigsaw's success and also in the success of Guardvant, a fatigue-detection specialist company, acquired by Hexagon last year. How are you applying your experience and expertise to Hexagon's OP Pro?
SB: Yeah, Jose. Really, from our experience at having started Guardvant and all the years that we've worked at the mining industry with Jigsaw and different products, we see a very clear path for the future of OP Pro, with our experience from Guardvant. For example, one of the things that we see as materializing in the next version, so for the OP Pro, is that we're going to integrate the operator-alertness status into the assignment algorithm. So, in other words, we will be able to use this information in dispatching tasks to different operators. For example, if an operator is experiencing fatigue, he can be directed to refuel, for example, or he can be directed to do different tasks that are not like so—not very long hauls, for example, where it will take him a long time to do. So, basically, we're going to integrate all data information into the algorithms, making a unique system in the world. We'll be the first ones, really, to come up with that add-on to the application.
JO: In addition, the technologies that are used for basically managing a system like this are a distributed application server technology where we have components that are running on equipment and as well as components running on the server, and all of this infrastructure that we have built up for wirelessly communicating and sharing information between multiple pieces of equipment and server are really sort of the same type of technologies between OP Pro and the Guardvant fatigue-management system. And so going forward, a lot of the things that we've done with the fatigue management, our computer vision, we have cameras looking at the operator's face, and we're monitoring the operator's face. These same computer vision algorithms can also run on the external cameras so that we can take this technology that we've developed for computer vision and all of this video processing technology and integrate that into smart proximity detection systems as well.
And in addition, there is the integration because by integrating the two technologies, now you don't have to buy a fatigue-management system for one place and buy a fleet-management system. Now you have two computers and duplicated radios and duplicated GPS technology. We can integrate it all into one platform and greatly decrease both maintenance costs as well as hardware costs for the combined systems.
JS: You know, we often-times consider and talk about Hexagon Mining being puzzle pieces that have been put together, and we keep adding to that puzzle piece. Where does OP Pro fit into that Hexagon end-to-end life-of-mine vision, in your opinion?
SB: I think OP Pro is really a critical part of that puzzle, of the life-of-mine - Hexagon's vision. For example, I think it really fits the gap, right after mine planning, and it ensures that your mine planning is executed properly. In other words, it will make sure that your blending is executed correctly, that you're feeding the dumps at the correct rates, that utilization of your equipment is up to par to what you have planned. So, basically, it's critical, right after the mine planning. It fits right into that. And then the idea is to have the fleet-management deliver the product, ready for the next phase, which can be feeding the plant, different plants. It could be leaching facilities, or it could be the concentrator. So, basically, what the system does, it gets the material ready for the next phase.
JO: Life of mine indicates a lot of days, day-to-day operations. And OP Pro, as well as the fatigue-management systems, deals with day-to-day operations as every day in the life of the mine you have to be able to manage this mining environment, you have to be able to track your operations, you have to be able to track your equipment, you have to be able to know what everything's doing, and manage both time and material during this phase. And that's really what fleet-management systems do: they manage time, they manage material, and they manage workers. And without that, you don't have any end-to-end or life-of-mine vision.
JS: How does OP Pro make a mine smarter?
SB: You're right. It's like kind of what we've been talking about it before. I think—and what Jon was talking a little bit before—you have to manage time, materials, flows, and different things. It would really be impossible for a human to do. So, basically, what OP Pro does, it gathers all this data, it digitizes it, puts it, brings it into the algorithms, and it's able to make the right decisions. So really what it really does, it's able to react to changes in the operation. A mine really has to be smarter by reacting to change. If everything continues in a continuous process with no changes, then things are easy. But a mine's really a living entity where things are constantly changing, there's equipment breakdowns, and different things, so that's what an FMS does. It adjusts to changes by feeding all this data continuously in real time to the mathematical algorithms that are processing this and comes up with a new solution. So every time that there's a change in the operation, it comes up with a new solution. But it's sent to everyone in the operations, how to react to that. So that's really how it makes it smarter. It's just like something that no human really can do or no humans because of the amount of data that you're gathering.
JO: And just to illustrate the kind of things that we're monitoring with OP Pro. I mean, we're actually monitoring when every dipper hits the truck. We're actually measuring swing time, we're measuring spotting times, we're measuring travel times, we're continuously updating these travel times as well as expected arrivals. We're continuously updating a linear programming solution as well as a dynamic programming solution to attempt to maximize flow rates of material to the dumps and the crushers as well as minimize idle times of trucks and the shovels. And being able to monitor this continuously in real time is something that just requires computer technology to do it.
Back in the day, before there were fleet-management systems, a guy was just locked. He says, "Okay, you haul off of the shovel all day." And until maybe two hours later, and they notice, "Oh, well, there's six trucks over at that shovel. Maybe I should move one of them." You wind up wasting tremendous amounts of resources and money if you don't have computer technology to monitor this for you.
JS: What distinguishes OP Pro from other FMS solutions?
SB: I think that there's several factors, really. One of them, I think is that OP Pro was really designed from the ground up, with the newer technologies in mind. So it uses—it wasn't like something that was adopted, or it was really designed for distributed technology, this dual-computing technology. It was designed to make use of the latest GPS technologies and communication technologies. So that makes a huge difference, makes the system very scalable. It makes it very modern so it's easy to integrate new changes. It's easy to play with different devices with different computing systems, and the algorithms. By utilizing all this new technology, they get a lot more data that was not available before. For example, we have a really good interface with a lot of—with most of the equipment manufacturers, so we can gather the data from the equipment and analyze the vital signs in real time, that all of those will affect the algorithms as well. So, the system is designed to do that and is designed to change and to evolve with the times. One of the last examples, as we said, for example, it's ready to integrate with the Guardvant application, with the fatigue levels. So we can integrate the fatigue that has already the design to do that. It can integrate the mine planning as well. So it's basically lots of, what I would say, makes it a big difference.
JO: And I'd like to expand on that a little bit. The distributed nature of OP Pro really makes it totally different. In fact, the design that I came up with for the Jigsaw Technologies was specifically designed to alleviate some of the deficiencies of previous fleet-management systems. The previous fleet-management systems were designed such that the onboard equipment was relatively stupid. It was sort of like a text messenger. You could press a button that would send it to the central computer; the central computer would tell you what to do. And all of the intelligence was on the central computer.
That sort of worked using the technology, and it was the appropriate design for a system back in the 1980s because there were very low-bandwidth networks but they had very good coverage. In other words, the radio networks were always there. Just like your FM radio. You can drive around, and your FM radio is always working.
But, I guess you've noticed your Wi-Fi radio isn't. You have all these dark spots. And so, putting the intelligence in the central computer makes it not really work very well with Wi-Fi, with modern networking technology.
And so the distributed design, because the intelligence is now onboard the vehicle, you're not dependent on a network. You can actually make a decision onboard the vehicle, and you can gather data onboard the vehicle and store data, and then upload it whenever you have radio coverage. And that greatly enhances the scalability as well as the performance of the system. We can do things that we could never do before because all of the data acquisition is actually performed on the vehicle. We're, for instance, continuously monitoring an accelerometer at 40 hertz and being able to know when each dipper falls in. We know exactly when the truck starts moving. We can actually detect automatically as soon as he raises his bed by the vibration of the truck that he's actually tipping. And, these kinds of technologies, with a central computer, you couldn't do it. And so in the OP Pro, we implemented full-haul cycle automation, where the operator doesn't have to press buttons and tell the system what he's doing. The system onboard the truck actually knows what the operator is doing based on hardware inputs on the device.
JS: That's great. Thank you for that. You've also already touched on this. I think the developments that are coming in OP Pro, like you were saying, Sergio with integration into planning, integration into the safety part. How else do you see OP Pro developing?
SB: One of the things that we see developing, which is very important, is how OP Pro is going to play with the autonomous vehicles and the autonomous mining operations. Autonomous mining operations is basically—you still have the same issues that you'd have in a normal mine except you don't have operators. You don't have people operating the equipment. But you still need—it's critical that you have a robust and really dynamic fleet-management system that can control the fleet. Just in any operation, you're going to have equipment breakdowns, whether it's in excavators or in haul trucks. So you need to adjust the changes. So, you need a supervisory system that can direct that fleet, that can orchestrate how the operation is flowing. So that's really something, but it's already in the works. So we're doing that. It's basically one of the other evolutions. And again, like you said, that still doesn't imply that we're not going to interface to mine planning. We might not interface, and we might not use the fatigue module because there's no operators, but it's still, the mine planning is going to be critical. So, it's really basically, it's building in everything that we have towards achieving that.
JO: And certainly, in the case of autonomous systems, which are the future of a lot of mining technology, there's a lot of interface with video cameras, and that plays right into all of the computer vision algorithms that we're currently using for operator management. Autonomous vehicles require a distributed solution. I mean, a tele-operated mine isn't going to work, where you have a bunch of operators in a central office that are trying to use joysticks to drive vehicles. I mean, it's just not feasible. So ultimately, the autonomous systems of the future are really going to require computer vision systems. They're going to require intelligence on board. They're going to require a distributed system where the system is largely independent and can function just like an operator of a vehicle but doing it with algorithms rather than a human.
JS: The world over, OP Pro helps mines safely save time and money. What personal satisfaction have you derived from helping create the solution?
SB: Personally, I feel an immense satisfaction of seeing our products—products that have been built by our engineers, designed by our engineers—deployed and implemented by engineers all over the world, that they're functioning all over the world and that they're really considered the best in the world at what they do. That is really an immense satisfaction.
The other thing is that these products—counting the FMS and the fatigue applications—they're making a big difference in mining operations. They're making mines primarily safer operations and more productive operations. So, we're really helping both ways: helping a mine do better, and do things better safely. So that is a huge satisfaction.
And one of the things is that this satisfaction is shared by everyone in the company. Everyone is really, from all offices in the world and every level of people that work with us, is a satisfaction to see that we're building these products that are considered the best at what they do in the world and that they're deployed all over.
JO: I've personally seen great satisfaction in all of my career developing fleet-management systems. These systems have been critical for optimizing mine operations as well as the safety systems that we've developed at Guardvant have really been critical in saving potentially a lot of serious accidents in the mine. But from a personal standpoint, I really find a lot of satisfaction in the diversity of knowledge that I've been able to gain in technology because I've been very lucky in that I've been instrumental in hardware, firmware, operating system, design, algorithm design, as well as computer vision processing, video processing. Most engineers don't get the variety that I've seen in my career. And so it's been very challenging working in this field, and I think it's a very interesting field to pursue for any new engineers.
JS: Excellent. A big thank you to you, guys, for being our guests today. For more information about today's topic, visit hexagonmining.com. To listen to additional episodes or learn more, visit hxgnspotlight.com. And thank you for tuning in.