TopN AnalyticsReal-Time Charger Uptime & Outage Monitoring
An eMSP platform serving multiple charge point operator (CPO) partners who need to monitor the operational reliability of their charging infrastructure.
Where things stood
Charging infrastructure reliability directly affects both revenue and customer trust, but raw connectivity data isn't a usable reliability metric on its own.
CPOs need a clean, consistent, daily view of how much time each charger spent online versus offline — without gaps in the record on days when a charger produced no events at all, which could mean it was fine, or completely dark.
What we built
OCPP connectivity event logs (BigQuery) feed a Python pipeline that pairs each 'connection opened' event with its corresponding 'connection closed' event to compute online/offline durations, handling edge cases at day boundaries.
The pipeline scaffolds a complete grid of every charge point against every date, so a charger with zero events still appears with an explicit uptime record, joined with station and connector metadata (zone, city, state, connector power rating, current type).
Output refreshes daily into a BigQuery dashboard table, orchestrated via Apache Airflow.
How it works
TopN Analytics built a pipeline that reconstructs per-charger daily uptime directly from OCPP connectivity event logs, making 'no data' an explicit, visible state rather than an invisible gap.
This methodology has been instantiated as a reusable pattern across multiple CPO brands.
What we delivered and learned
A naive uptime calculation built only from days with actual events would silently under-represent chargers that went completely dark — gap-filling against a full date x charge-point grid was the key design decision.
Want the full picture?
We're happy to walk through the details, numbers, and trade-offs directly.