Our intelligent system monitors your chargers 24/7, ensuring that they are always in the best possible condition. If a charger is predicted to fail or go offline by our custom machine learning models, our system will alert you so you can take action before it affects your network.
Although instant resolution is possible for most charge point faults, manual detection and handling still consumes significant time and resources.
Our system is trained on billions of fault-resolution data logs, enabling automatic handling of most cases, reducing the time-to-resolution by up to 99%.
Charge Point Models
Monitoring Active Chargers
OCPP Data Logs Analyzed
We automatically monitor and observe recently handled chargers to ensure they are working as expected and prevent fault-reset loops
Keep track of chargers post-reset, ensuring they successfully connect to the network and charge vehicles before being removed from the overview. This helps prevent fault-reset loops, where chargers are reset, assumed to be working, only to fault again, repeating the cycle indefinitely. Chargers that repeatedly fail after reset are automatically added to the Dispatch Technician or Investigate view, ensuring quick and effective resolution.
In most cases, chargers fail multiple sessions before an action is taken, due to faults slipping through existing monitoring systems, that only read the data at face value.
By aggregating and analysing data in real-time, we provide actionable insights, enabling a reduction of recurring faults by 75%.
Charging issues that aren’t inherent to the data, rendering them invisible to any operations team, account for over 90% of technical issues.
Through real-time data analysis and ML-model data processing, we detect errors such as zero kwh sessions and persistent idle faults, allowing these problems to be fixed.
Understand your locations performance and behaviour better with in-depth analytics from all chargers at a specific location, allowing you to identify any trends or issues.
In addition to invisible faults that can be revealed through various data processing methods, chargers also experience issues unexplainable through the data.
With dedicated ML-models, we identify chargers that deviate significantly from the norm — based on firmware, location or model averages — allowing for detection of issues exceeding data, firmware- and configuration recommendations, and identification of high-level negative performance trends.