Making Charging Friction-Free: Interview with Robert Hollinshead of Hitachi ZeroCarbon

Catering to global commercial fleets, Hitachi ZeroCarbon strives to make vehicle readiness a priority through their managed charging services. At a webinar this year, Product Manager Robert Hollinshead explained how siloed EV charging can be and how eDRV’s AIOps can help clarify opaque charging transaction data and reduce friction among collaborators and partners in EV charging. 

What problems do Hitachi ZeroCarbon come across in the day to day operations of a charging depot?

Among our many services, Hitachi ZeroCarbon provides battery and charging management services for fleet operators. So if you've got a large fleet, regardless of what stage in your journey through electrification, we've got different levels of technology and managed services to help you along. 

At Hitachi ZeroCarbon, we measure ourselves on vehicle readiness. We have an active role to play when chargers aren't working or chargers are underperforming and we've got a vested interest in making sure that we can reduce errors that are stopping power from getting to vehicles and ensuring they are charged and ready for operations every day. 

In our work with a national UK bus fleet, we found that there were reports from our customers where charging wasn't finishing on time and it was almost like an opaque wall as to what was going on. We would get these vague tickets in our service management platform saying “charging not working”, “unacceptable” or “slow charging”. For our operations team this meant diligently and systematically work through large chunks of data and looking through diagnostic logs from chargers. It was something that involved large text files with tens of thousands of rows.

We were getting to the bottom of what the issues were but it was taking a long time. There was a lot of friction in this process. Because of the effort involved in every single ticket, we didn't have the time to look at the whole customer landscape. We weren’t able to observe the whole ecosystem to spot these patterns as everything was evaluated in isolation. 

When major errors crop up, it’s essential to coordinate among different vendors and players in the ecosystem. Is that true in your experience as well?

Our customers already have their own suppliers for vehicles as well as their own charging hardware as we position ourselves as hardware agnostic. With the help of eDRV, we can work with a whole variety of different charger vendors, but we all share one common customer that we're trying to please. The problem was when all the data was segregated, we didn't really see what each other's part of the system was doing. It can become quite defensive when problems crop up.What we really needed to understand was why there were high percentages of failed charging sessions. OCPP alone isn't enough to diagnose these errors. OCPP is there for communications between the charger and the CMS so it only has information about the charger. But for Hitachi ZeroCarbon, we're trying to manage the whole charging process, including the information between EV and charger. 

“You shouldn't need a whole organization worth of deep level charging experts in order to be able to manage charging if the tools are there.”

What are the effects of vast data streams when it comes to the operation of personnel at your managed charging locations?

You shouldn't need a whole organization worth of deep level charging experts working 24/7 in order to be able to manage charging if the tools are there. The question becomes how can we reduce the technical expertise that you might need in your support organization in order to effectively manage it? I think that's what we've discovered as part of our operations. ‍

How is eDRV AIOps being used at Hitachi ZeroCarbon today?

We've got AIOps monitoring sessions for our most popular charger model across our network, with multiple customer sites operating with multiple different EV types as well. We’ve integrated AIOps into a core part of our support workflow when issues have been raised and diagnosed, and we’re using it to spot patterns and trends. Beyond the user interaction of AIOps, the visualization of data allows you to pick out one particular site where, for example, you can identify that 50% of the issues occurred with the same EV. That really helps you narrow down the problem. 

AIOps also lets you measure the effects of what you've been able to fix. One of the success stories so far is where someone raised a ticket where we had previously had months of going around in circles between EV vendors and charger vendors. With AIOps we arrived quickly at a conclusion that there was an issue with the charger firmware. We were able to pull out logs, summarize across multiple sessions, narrow down on that particular issue and get the acceptance that we needed to push out a firmware update. ‍

Could you speak to the business value you’re seeing with the use of eDRV AIOps in your operations?

Now when we get vague requests, we can quite easily triage out things that aren't deep or low lying or quickly identify issues that can be easily explained. This is shifting where the support is done at our charging locations. It allows our technical experts to focus on those high value, big ticket issues, letting the software diagnose smaller issues that we would usually have to go back and forth with the customer to find the problem. ‍

What future directions are you most interested in monitoring as charging operations scale at Hitachi ZeroCarbon?

I think the complexity in the future will be over “how do we get the vehicle data?”. Right now they send the charger little information and they don't send the charger everything. 

How do we bring them into the house, so to speak, so that we can provide a broader overview? The complexity will be in integrating that data. The idea is that we can go back to our customers–the fleet operators–and really provide them with great information, a root cause analysis and recommendations on what vehicle types work best with what chargers as well as how we can reduce those issues. 

This interview was transcribed and edited for clarity from an eDRV Webinar on AI and EV Charging. To learn more about how eDRV AIOps can further your charging network, reach out to us directly at hello@edrv.io.

‍