EV Charging Operations Management

The ticket the driver never had to raise

Most charging operations run on the driver noticing. Something fails, the driver tries, the driver gives up, and some fraction of them call. The operation finds out late, from an annoyed person, with no detail, and usually after the site has already lost sessions.

Self ticketing inverts that. The session itself raises the ticket: the unit reports the fault it saw, the platform opens the ticket with the unit, the site, the fault and the recent history already attached, and the operation knows before the next driver arrives.

This is the part worth building first, because everything downstream depends on it. Analytics need events that were actually recorded. Response targets need a clock that starts when the fault happened rather than when a driver complained. A maintenance schedule needs a fault history that is not just the calls that were angry enough to be logged.

The three areas, and what we cover in each

Operations divides into three, and the tooling for each is different work.

1. 1

   Installation

   Checklist software that walks the installer through the site: wiring, cabling, earthing, accessories, the photographs, and the sign-off. A commissioning record that exists as data rather than as a paper form in a drawer is what makes every later fault diagnosable.

2. 2

   Running operations

   Self ticketing from the units themselves, the analytics built on those events, and the voice agent handling the calls that still arrive. This is where the day goes.

3. 3

   Maintenance

   A portal to report what was found, record what was done, and recommend what is due, so scheduled work is driven by the fault history rather than by the calendar alone.

What operations actually consists of

• Monitoring: knowing what is up, what is degraded and what nobody has noticed is down
• Support: answering drivers, site hosts and partners, mostly about the same six things
• Field work: dispatch, repair, replacement, and the visits that turn out to be unnecessary
• Maintenance: the scheduled work that prevents the unscheduled work
• Commercial: tariffs, invoicing, reconciliation, revenue share, disputes
• Customer: onboarding, credentials, refunds, and the relationship
• Compliance: records, safety, metering, tax

Most operators start with the first three and discover the other four when the network is large enough that they cannot be done informally.

The four sources of every ticket

Almost every operational event reduces to one of four causes, and the split between them is the single most useful number an operation can know about itself.

Our own unit failure rate runs near 0.5 percent, about one charger in two hundred, and it is a much smaller number than our call volume. An operation that has not separated these four is resourcing against a hardware problem it may not have.

Where the cost actually concentrates

Operators expect cost to sit in parts. It sits in travel.

• A site visit costs a multiple of most parts, and a large share of visits find nothing wrong
• Distributed sites make this worse in a way that is not linear: the tenth city costs far more per site than the first
• Support call volume is driven by things that are not faults, and answering them still costs an hour
• Reconciliation effort is invisible until it is a person's full-time job
• The expensive fault is not the difficult one. It is the trivial one that needs an engineer physically present, which is why a tripped breaker is the worst small fault in charging

Every design decision that avoids a visit is therefore worth more than it looks, and every one that requires a visit is worth less.

Response targets worth committing to

Ambitious targets in a contract become penalties in practice. These are worth separating.

1. 1

   Time to know

   How long before the operator knows something is wrong. This is the one worth being aggressive about, because it costs monitoring rather than people, and everything downstream depends on it.

2. 2

   Time to diagnose

   Remote, and it should be minutes for most faults. If it is not, that is a logging and telemetry gap rather than a staffing one.

3. 3

   Time to respond

   Contacting whoever raised it. Cheap, and it is what determines whether people feel looked after.

4. 4

   Time to restore

   The one customers care about and the hardest to commit to, because it depends on travel, parts and whether the fault is yours at all.

Committing to restoration times without separating the four causes is how operators end up penalised for utility outages and vehicle behaviour.

Shifting work from field to remote

The ratio of remote resolutions to site visits is the number that decides operating cost, and it is set by decisions taken long before the first fault.

• Firmware that records enough to explain itself after the event
• Traces retained long enough to investigate a complaint reported three days later
• Fault codes that distinguish causes rather than reporting that something failed
• Remote log access without dispatching anyone
• Remote firmware update, so a fix does not need a van
• Remote configuration, so a setting change does not either

None of these can be retrofitted cheaply. They are procurement requirements, which is why operations should have a voice in hardware selection and usually does not.

Team or outsource

The usual answer is a split: diagnosis and customer contact in house, because that is where knowledge compounds, and field attendance contracted, because that is where geography dominates. Outsourcing diagnosis is the error, because it is the part that gets better with repetition.

What we do and do not do

Worth being exact about, because operations is a word that covers very different commercial arrangements.

1. 1

   Managed monitoring, yes

   We can monitor your network, including through a management platform that is not ours, and alert you when something needs attention. With IoT enabled station management on site, that extends past what the charger itself reports: the cable, the bay, the panel, the environment.

2. 2

   Operating your network for you, no

   We do not take on running your network as a service today.

3. 3

   Site visits, no

   We do not attend sites. The field work stays with you or with your contractor, and what we provide is the tooling and the monitoring that tells them where to go and what to bring.

The value of remote monitoring is the same either way. Most of the cost of a charging operation is in visits, and most visits are avoidable if the team knew what was wrong before the van left.