Industrial Analytics wins Deutsche Bahn Mindbox Challenge

With a great team and a great presentation by Dr. Richard Büssow Industrial Analytics wins the DB Mindbox Challenge Future of Maintenance 2020. The already in-depth knowledge of the analysis of machine data, which Industrial Analytics was able to contribute by bundling many years of experience, was convincing.

Building on this success, the first project collaborations between Deutsche Bahn and Industrial Analytics are now being planned. To this end, Industrial Analytics, together with the Deutsche Bahn Group, will advance a Proof of Concept (POC) for predictive maintenance using vibration analysis. The aim of this POC is the acquisition of vehicle data, which is to be installed in a test train with the help of selected sensors and subsequent steps are to be determined with the help of the analysis of Industrial Analytics. At this point, the experience of Industrial Analytics turned out to be helpful, since the team can give recommendations for the selection of sensors and at the same time incorporate their own experience with the acquisition of data into the project. Data to be collected in this part of the POC will preferably be vibration data, which will arise when objects in the undercarriage construction of the test train are driven over.

Based on the collected data, initial analyzes are then possible. Using vibration analysis, we will interpret the recorded data on different train types for asset localization and anomaly detection. Furthermore, vibration data can provide information about the chassis and possible effects caused by objects being rolled over. These can then be assigned to appropriate recommendations for action and thus enable autonomous maintenance and repair.

The cooperation should then form the basis for future projects in which artificial intelligence methods such as machine learning are also to be applied to trains. The real-time monitoring of rail vehicles can be mentioned as an overarching goal, but in the long term the autonomous driving and maintenance of trains based on occurrences and events that rail vehicles independently recognize and interpret.