FLEET MANAGEMENT
Fleets on the fly
The most recent version of the Hexagon OP Pro FMS includes support for the Hexagon 64-bit onboard platform , helping to accelerate the company ’ s Power of One vision
From introduction of AI to new nimble agnostic players and entrenched players rethinking their approaches , Paul Moore looks again at mine fleet management systems , often referred to as FMS
Most people today are aware that AI is a technology that enables computers and machines to simulate human intelligence and problem-solving capabilities . And in mining that it is becoming an increasingly important part of next generation autonomous haulage systems ( AHS ). But it has huge potential in fleet management systems as well .
George Mavros , Komatsu Product Manager , Load & Haul told IM that AI fits well with fleet management optimisation in that it can help define where the truck needs to go and at what times : “ In some ways , we ' re already incorporating some of the technology now , and we plan to extend and expand on this , for example introducing optimisation in refuelling and other processes that we ' re currently not optimising . So it ' s it ' s more of a continuation with what we are already doing with AI and expanding AI capabilities into micro-optimisation or local optimisation as we call it . It is also important to note that AI is not going to replace the need for human supervisors monitoring mining fleets . But it will help the FMS dealing with real time events , especially those that are unexpected or unforeseen .”
Mavros says it will still need to set parameters – “ for example we know the restrictions of the mine that at this point AI may not be aware of , so the AI also needs to respect the boundaries that we set .” He says the AI is utilising the tremendous amount of data that is being captured anyway . Komatsu is using digital twins to make that data available in other containers . “ For example to optimise a refuelling schedule , or other examples like optimsing breaks and the park up process . We create these data containers so that the AI can focus and function based on those restrictions that are information specific .”
For refuelling , the AI would decide this is the
optimal time to refuel and then the driver would get a a message telling them to go and refuel , rather than the dispatcher having to make a separate decision . “ The algorithm will come up with the with the refuel schedule until the end of
the shift as we know the current levels and the AI , based on the utilisation and the availability of a truck , will say this is when that truck needs to go to refuel and where it should go – it is a window that reflects fuel-based availability . We also want to minimise the waiting time at the fuel farm and of course keep the truck out of production as little as possible , so the drive to the fuel base is also an input that we take into consideration . Leaving all these decisions to a human , means it ' s practically impossible for them to come up with a perfect optimised schedule but the AI will help us do that .”
As stated the same concept can be used for park ups and breaks so you can set specific time windows . Mavros : “ Right now , there ' s a lot of manual scheduling and use of relief modes where operators need to take a specific break every defined amount of time to limit fatigue and and all that . The AI would allow operators to take opportunistic breaks , for example if the shovel goes down for a short period of time .”
As this logic Komatsu is building is agnostic , as it further develop its power agnostic mining
iVolve , now part of Komatsu , is a technology company that provides fleet management solutions for small to mid-tier quarry , mining and construction operations trucks , Mavros says AI will also enable it to manage how charging is done for battery trucks , or when battery trucks should join a trolley line for dynamic charging for example . “ We ' re going to address recharging or refuelling with the same concept , but both obviously have different restrictions that we need to respect and meet .”
Mavros says it is already introducing these new applications and algorithms which work hand in hand with its Modular Mining DISPATCH FMS , which is responsible for feeding these applications with the information they need . “ These AI engines are going to make the decisions , and then pass them to DISPATCH to get executed . So the engine itself at this point is not native to DISPATCH . We have these applications currently at a couple of our mining customer operations in tests , focused on the refuelling optimisation .”
Currently Komatsu is using the digital twin with the engine making the decisions on its side – referred to as encapsulated dispatch . Komatsu is pulling the data it needs and the AI engine is currently on Komatsu premises . But it will have the ability to this on the Cloud as well . The outcome is going to run and the AI engine will provide that schedule that is then sent back to DISPATCH for execution .
The other big FMS news in Komatsu is that it has acquired iVolve Holdings Pty Ltd based in Queensland , Australia , which is led by CEO Kim Parascos . iVolve is well known supplier of FMS for construction and mining equipment . In particular it provides FMS for small to mid-tier miners , contractors , and quarries . iVolve offers systems that visualise operation management information , reduce running costs , and promote safe operations through access to real-time data .
Through this acquisition , Komatsu says it will globally deploy iVolve ’ s FMS , which has a unique IoT platform , as a new solution to further contribute to improving safety and productivity at customers ’ workplaces . iVolve will continue to provide its services as an independent group company , and Komatsu says it aims to become the world ’ s leading FMS provider in the construction and mining equipment market by capturing new business opportunities through Komatsu network . iVolve ’ s main FMS package is Mine4D , which focuses on operational efficiency improvements directly related to production and vehicle maintenance . This enables its clients to increase productivity , cut costs and minimise risk . The 4D refers to four key components . iVolve Mine4D Production enables real-time decision making in two ways : by providing real-time feedback to operators in-cab ; and by monitoring and recording in real-time all parameters of the haul truck load cycle . This data is sent over the Nexis network to the iVolve Server for real-time access
42 International Mining | MAY 2024