Close
Telematics - ML-based Fuel Station Quality Rating System
Posted On February 12th 2025
Our customer, one of the leading Telematics data aggregators for commercial vehicles, wanted us to rate the refuelling stations based on the refuelling quality to optimise fuel usage, reduce costs, and enhance safety.
Challenges Faced:
● No explicit indicator of refuelling event was available as part of CAN feed
● Readings around fuel left in tank were noisy due to constant movement of the vehicle
Decision intelligence
● Linear reading of fuel available in the tank was obtained by applying convolve Normalising technique
● Refuelling points were identified with sudden change in fuel levels with higher variance
● Vehicle mileage was analysed before and after refuelling
● Fuel stations were classified into classes A, B and C based on the improvement/degradation of mileage after refuelling
● This fuel station quality rating was in-turn integrated with Transport trip planning to identify best refuelling stations in the route in which the vehicle is planned to travel
Outcome delivered
● Fuel Station quality derived from multiple vehicles data without any additional indicator
● Informed decision on where the vehicle is refuelled and the station’s quality
● Loss due to poor fuel quality is reduced
● Better route planning with information on fuel stations and their quality en-route
Get in Touch
Sales Enquiries : sales@datsy.io
Support Enquiries : support@datsy.io
Partner Enquiries : biz@datsy.io
