About our client
Our client is a Canadian software firm that serves automakers with personalized, user-centric SaaS solutions. End customers use our client’s software to process vehicle preventive maintenance requests from owners. Requests can be based on a maintenance schedule, advice from remote diagnostic systems, or predictions from a cloud-based machine learning system. Our client specializes in technology solutions for the automotive, energy, and banking sectors.
Our client came with the challenge
Our client acquired a large contract with an international auto manufacturer to help the manufacturer’s Canadian dealers schedule vehicle maintenance. To serve the large user base – all owners of vehicles by the manufacturer – our client started developing an online Software as a Service tool for predictive maintenance of cars.
Their biggest challenge for predictive maintenance software was handling a large volume of vehicle data, including data on the performance of individual vehicle parts. To analyze this data smartly, we proposed introducing machine learning algorithms as well as an online support system with high fault tolerance. Our client already had related experience with online systems but was looking for a partner with experience both in the automotive sector and with SaaS development to get things done with speed and quality. Their search led to Intellias, as our company is well known for its focus on the automotive sector and its experience with SaaS solutions across industries.
Intellias developed the solution
We developed an online support system based on Microsoft Azure cloud services that processes requests for appointments and matches drivers with nearby dealer service centers.
The online system also notifies drivers by phone message or email about the need to check parts that have exact maintenance schedules stated in the vehicle specifications, which are synchronized with the online system.
Machine learning algorithms will recognize:
- Starter motor malfunctions
- Drop of pressure in the fuel pump
- End of a battery’s service life
We achieved great results together
Together with our client’s team, we’ve achieved notable improvements to the notification system for drivers about car maintenance thanks to a comprehensive online support tool. We’re now preparing to develop a prediction system based on machine learning algorithms that collect data from steering and braking systems as well as the starter motor, battery, and fuel pump and send all this data to the cloud for analysis and diagnostics.
Here’s an example of the E2E predictive maintenance algorithm for battery life:
- In-car monitoring system checks battery status
- Data is transferred to the cloud
- Cloud-based ML algorithm predicts that the battery will run low
- System processes all inputs and prepares advice to the driver
- Notification system sends a message to the driver with instructions
- Driver avoids issues related to low battery thanks to the E2E system’s prediction