Field test vehicle usage classification
Testing is a crucial part in the process of developing a vehicle. The motivations could range from evaluating the vehicle’s performance in different environments to making sure its functions work as intended.
The data gathered from Volvo Truck’s field test vehicles are logged under various circumstances, while being used in customer application, driver training, test by Volvo, service and repairs and so on.
However, information about the circumstances is not always available explicitly.
Manually classifying the data can be a tedious and time consuming task. In order to increase the efficiency and reliability in data analysis, machine learning techniques have been employed. Using a combination of feature engineering, unsupervised and supervised learning, a result of 91% accuracy was achieved.
The method is currently in use on Volvo’s analysis of field test data.