In this episode of Energy Insight, Benoit Lacroix, CEO of Relion, shares his thoughts on scaling operations and maintenance for EV charging infrastructure, and the need for a proactive approach to systems monitoring. Below is an edited transcript of the conversation.
SGO
For listeners who may not be familiar, could you describe the kind of work Relion does in this field?
Benoit
At Relion we help network operators of charging infrastructure to operate at scale. We’re the first platform that is dedicated to operation and maintenance. So we bring everything under one roof that goes into their issue resolution identification process. That includes monitoring, diagnostic, issue resolution, part management, work order and field dispatch. We also bring under one roof all the stakeholders, so the operators themselves but also manufacturers and service providers. So all these stakeholders can collaborate, share the same visibility and accountability, so you have a continuity in the process of operating and maintaining an infrastructure.
SGO
Your platform has recently been adopted across a wider range of charging environments. Last year, you were selected by Greenlane to support its medium/heavy freight charging network. Obviously, this is an area where reliable availability is especially important. How do you help customers set expectations around reliability?
Benoit
This is the most extreme application I think for reliability; trucks, or commercial vehicles have to be ready with the state of charge at the moment desired. The biggest gap is execution -- we like to call it the “now what?” It’s one thing to know that there’s a problem, but what do we do now? We still operate in an industry where granular detail on error codes and definition and process and what to do next is still hard to combine. Getting the knowledge, getting the information, mobilizing all the stakeholders that could be involved in solving or understanding the problem can be difficult. Plus, if there’s any field intervention involved, fetching the parts, determining who has the right training, who is available for dispatching effectively and efficiently, etc. So the gap is really in the execution, and of course at Relion that’s our bread and butter. But as an industry, making sure that the knowledge is available, is shared, that there are interoperability and partnerships to make sure that there’s a better flow when we know there’s a problem -- this is the biggest gap that we identified.
SGO
Since September, you’ve been working with Montreal to deploy your platform in select areas for curbside charging. Can you tell us about how you’ve approached that unique challenge of urban public charging networks versus private fleet customers?
Benoit
Yeah, completely different. With fleets the element of interest is not only the availability of the charger for a newcomer; it’s very often behind a gate, returned to base. You have your vehicles, you want them to be ready in the morning. This is what you’re monitoring as a platform. Now for curbside charging, the City of Montreal approached us to do a pilot project as they were reevaluating their technology mix and literally their business model for the future, and this is a totally different approach. Here we’re talking very often about simpler units, L2s, but they’re out there: they’re really in the jungle, they’re really distributed by definition. We’re talking here in Montreal close to 2,000 chargers distributed across the city. And the complexity comes with the number -- the number and the high amount of units, highly distributed. So the monitoring can become very tedious, and also the execution, again, a problem. Even if the units are simple, you have 2,000 pieces of equipment that you need to attend to. Reliability is always important, but it’s still not as critical as for fleets. With that huge number, there’s still a lot of citizens that are depending on this infrastructure. So really the complexity comes with numbers. We’re talking about more finding the needles in the haystack and bringing the problems up and making sure that workflows are better executed because again, with that huge number, you can imagine how OPEX and deploying field service providers can become a very tedious task.
SGO
We know from industry data that despite the advances that have been made, charging networks continue to have significant reliability issues. So what would you say are the biggest technical barriers to deploying predictive maintenance at scale? Where do you see the most progress?
Benoit
I like to say, let’s not even walk before we run, but let’s crawl before we walk. For predictive maintenance, we ought to go there at some point, but I think we’re at a point where we’ve fist got to be proactive, really reach a proactivity in monitoring. Giving you some examples of what we do at Relion, I like to say we don’t only look at the picture, but at the movie. So we really look at patterns of utilization. We will raise problems if a charger statistically hasn’t been used much lately. Curbside is a good example: this charger is being used 1.7 times a day, but it hasn’t been used for the last 2.3 days. This is statistically significant -- probably something is wrong; maybe the pistol is in the ice. So it kind of brings a flag to maybe proactively attend and look at those chargers. Same thing goes for error codes -- if you have patterns that come, even if the charger is not down, you might want to take a look. So before predictability, I think we should just really have a real proactivity.
I think a lot of the reliability is impacted by the time it takes to solve an issue. OPEX is one thing, but also shrinking the mean time to resolution. We’re still at a time where for some units an error comes up and the error is called “no error” or “generic error”. So right away you don’t know what it is. Maybe you don’t have access to the error codes or you have access to some information, but it’s not comprehensive enough to really understand what’s going on. So then you try a few things, try to soft reset; it still doesn’t work. Okay, so you have no more information. What you do? You send a technician on site to literally figure out what’s going on and do some tests there, only to find out that it was XYZ problem and maybe unfortunately the technician didn’t have the right part, the right firmware update, so he needs to go back home, wait for this new firmware update to come back, then do a second dispatch, and you can imagine how this process can take a lot of time -- maybe two or three truck rolls where you could have done just one if you had had a real idea of what was going on. That’s still the reality. So we’ve got to fix all those kinks before we can really predict eventual failures and proactively send a technician with the replacement part before it actually breaks. We have to get there, but we have to fix all those processes I spoke about.
I mean the machines themselves are getting more reliable, that’s for sure, but distributed assets always break and there’s always problems. The whole O&M process and monitoring won’t go anywhere. So definitely the industry is getting much more mature in terms of technology. I think we’re getting much more mature also in terms of partnerships, again, the way things are being done. There’s still a lot of work to be done, but definitely we’re getting there. We’re always keen on also tipping our hat for best practices being done in the industry, like how more and more manufacturers will share more rich information about what’s going on in the machine. We still have some way to go, but I think this is getting much better. We’re starting to do a lot of integration with other peripheral systems to share information and ultimately just enrich any diagnostic we could do. We need to do more training on how to fix those machines, how to diagnose them, how to fix them. Definitely heading in the right direction and I really think that over the next 18-36 months we’ll really see some progress on how this industry and infrastructure is being monitored and maintained.
SGO
Do you have any final thoughts on what to expect in 2026?
Benoit
2025 was surely an interesting year; I think 2026 will be also. I think what happens in tough times is, everyone just gets better at what they do. So I think collectively we will just make sure to have more traffic to charging infrastructure, be more efficient in operations, in how we select equipment, how we deploy them, how we operate them. So I think it’s obviously good for Relion, but I think it’s good for everyone as it will establish more solid footing for what’s coming up. And obviously, EVs are not going anywhere. So if things pick up this year or next year, we will all be in a much better position to really offer the infrastructure that this transition needs.
Benoit Lacroix will be speaking in the session “Effective Charger O&M from Detection to Resolution: Teamwork, Not Guesswork” at SGO’s EV Charging and Infrastructure Summit, North America West, February 24th to 25th in San Francisco. For information and to register, visit www.ev-charging-summit-na.com
