Today’s modern society is completely dependent on timely and sustainable transport solutions. Predictive maintenance plays its role in achieving this by mitigating many of the unplanned vehicle stops as well as contributing to a safer road environment with less roadside standstills. Existing methods to diagnose and foresee vehicle failures are based on analysing the current state of a vehicle. However, to take the next step and reveal the patterns behind the failures, a more complete picture is needed. This requires data for many different sources and with a high variety in structure, i.e. vehicle maintenance records, vehicle usage data, and ambient conditions, such as weather data.
The use of the EVOLVE testbed will allow developments in predictive maintenance both at the levels of model training and diverse dataset analysis, leading to the creation of innovative services.