Predictive Maintenance in Production

Existing machine learning concepts are evaluated in this project and tested for their applicability to the acquired information. Promising algorithms are prototypically implemented, and models for assessing the status of the plant equipment are created. The models are trained using the recorded datasets as well as the domain-specific experience of employees. Additionally, existing maintenance processes at TEST-FUCHS are evaluated, ultimately resulting in a concept for implementing predictive maintenance in a production environment: necessary process changes, personnel resources, technical practices, system expansions, and a requirements specification for the machine learning software. This project is co-financed by the European Regional Development Fund. Further information on IWB/ERDF can be found at www.efre.gv.at.