COMPREDICT was founded in 2016 in Darmstadt and currently employs 10 full-time and 7 part-time employees from 8 different countries. With our main product, the COMPREDICT Suite component failures can be predicted precisely and the design and construction of individual vehicle components can be optimized – completely without additional hardware costs.
AI-based virtual sensors are the main feature of our product. They are purely software-based. We use the virtual sensors to measure loads, like forces, torques, temperatures and electric currents. These loads damage components like powertrain parts, chassis parts, the steering assembly, and batteries of electric vehicles.
Based on the loads measured by our virtual sensors we calculate precise usage profiles for the main components of each individual vehicle of a fleet.
The data generated by the virtual sensors is enriched with meta-data (e.g. weather conditions). Finally, methods of machine learning are applied, which learn from all these features and can thus exactly predict why, how and when a component will fail.