Lighthouse Projects · Energy
Predictive maintenance for EV chargers and demand forecasting
See the degradation before the charger stops.
Problem
Failures detected only after they happen. Stalled chargers, unhappy B2B customers, outages visible in public apps. No ability to predict peaks or degradation.
Solution
Predictive models over sensor history, alarms and sessions. Early degradation signals. Daily automated ops briefing. Field team allocation recommendations.
KPIs
- % of incidents detected before customer impact
- Field team mobilization time
- SLA breach reduction
- B2B-perceived availability
How it is measured
Every KPI with a baseline and a target.
The numbers for this project come from your data. We measure the baseline during framing, agree the target with you before building and, in production, measure the same KPIs the same way.
- 01BaselineMeasured during framing, with the data of the process as it runs today.
- 02TargetAgreed with the case owner before building, starting from the baseline.
- 03In productionThe same KPIs, measured the same way, compared with the baseline.
- % of incidents detected before customer impact
- Incidents with an alert before the first failed session or complaint, divided by all incidents.
- Field team mobilization time
- From the alert to the team arriving on site, from the service orders.
- SLA breach reduction
- Breaches of the availability SLAs agreed with B2B customers, per month.
- B2B-perceived availability
- Availability of each B2B customer's charge points in the period, calculated as in the contract.
Before starting
Prerequisites and risks.
Prerequisites
- History of alarms, statuses and sessions per charge point, long enough to contain repeated failures.
- Records of past failures and interventions, to train and validate the models.
- A field team available to test the alert prioritization during the pilot.
Risks and how we handle them
False alarms
Too many alerts waste trips and the team stops trusting them. The alert threshold is calibrated in the pilot with the field team, against the failures that actually happened.
Short or incomplete history
Without enough failures in the history, the model cannot learn to anticipate them. Framing measures what exists before promising forecasts.
Other projects · Energy
Want to see this in your operation?
90 minutes with you and your team. We leave with 3 concrete opportunities and estimated payback.