To do this heavy lifting, fleet management technology is introducing the dual strengths of artificial intelligence and digital twins to maximize the use of electric vehicles (EVs) against operational needs — using all of the real-time data connected to the internal workings of the vehicles and chargers, and that from outside influencers like route length and climate.
“Having a charge management system is essential for operating EVs effectively,” Brandy LeBeau, fleet manager for the city of Roseville, Calif., said via email. Roseville uses fleet management technology from BetterFleet, which has developed AI-driven optimization technology that creates a digital twin of the fleet operation.
The city is not yet using that component of the BetterFleet technology, but plans “to leverage this capability once our four EV buses are in service,” LeBeau said. “At this point, we anticipate utilizing digital twin modeling to further optimize our transit operations and ensure efficient service delivery to meet demand.”
Chargers don’t always charge at the same speed every day. Temperature, battery condition and other factors can affect charging speed. BetterFleet’s AI and machine learning technology takes all of these variables into consideration and provides fleet managers better insights into the true operational performance of vehicles each day.
The digital twin technology is able to mimic the topography of the region, the temperature on that particular day, and the heating and cooling demands of the vehicle, Daniel Hilson, BetterFleet CEO, explained.
“So, we know the mass of the vehicle, the performance of different situations, the energy consumption, the battery size, and yeah, is a battery degraded. So over time it might be degraded to x percent,” Hilson said, illustrating the many variables that make their way into the system’s modeling capabilities. “So all of those things will impact it, and the digital twin we’ve created is what effectively gives that pinpointed accuracy on what they can do effectively.”
Digital twin technology is increasingly becoming a companion piece to AI and smart city applications. Digital twins are being used to train “AI to do something,” Jumbi Edulbehram, director of smart spaces and local government at NVIDIA, said during a March 26 webinar hosted by TechConnect.
For example, digital twin technology can train AI-enabled cameras deployed in the real world to recognize smoke: “You can create the scenarios in the digital twin, and then train the AI to recognize those when it happens,” said Edulbehram.
Or, the digital twin could run any number of hypothetical scenarios at traffic intersections, which could lead to an accident or near-miss.
“And then of course, train the AI to cognize them quickly so that the appropriate response can be made,” Edulbehram said.
AI and digital twins are, experts say, able to bring about the next technology chapter in fleet management.
“So, when charge management 1.0 existed, it was really about the charges, like are the charges working? Are they turning on and off,” Hilson said. New developments in technology, he added, are offering deeper insights into the entire fleet system, collecting and analyzing data related to, yes, the chargers, but also the vehicles, their routes, climate and priorities set by the fleet organization.
Fleet management technology is essential for any organization wanting to get the most out of their vehicles, Roseville’s fleet manager said.
“Yes, the charge management system helps the city save money by prioritizing charging during off-peak hours when our rates are lower,” LeBeau said. “This not only reduces operational costs but also helps extend battery life by avoiding unnecessary fast charging, which can degrade batteries over time. Additionally, optimized charging ensures vehicles are ready when needed, reducing downtime and increasing reliability.”
