pdmInsight – Predictive Maintenance Applications
Project Description
Predictive maintenance (PDM) is a method designed to find an optimal maintenance cycle for machines. Therefore, sensor values are included to find correlations between measured data and machine behavior.
This interdisciplinary approach includes knowledge about engineering, data mining, and management. We try to enrich manufacturing execution systems (MES) with the features of predictive maintenance for an advanced production control. The research focus is to include generic methods into production processes.
To connect machines to predictive maintenance application it is necessary to deal with machine communication protocols like OPC UA and MQTT. These techniques allow not only the information transfer but also the semantical modeling of machines.
Considering growing data logging, it is necessary to include data reduction techniques and distributed analyses. All collected information is used to train the application and forecast machine behavior.
Project Details
Project Team
| Ulm University | |
 |  Burkhard HoppenstedtUlm University, Institute of Databases and Information Systems |
 | Dr. Rüdiger Pryss Ulm University, Institute of Databases and Information Systems |
 | Klaus Kammerer Ulm University, Institute of Databases and Information Systems |
 | Prof. Dr. Manfred Reichert Ulm University, Institute of Databases and Information Systems |
Funding
The project is partially funded by atr Software GmbH.
Duration
The pdmInsight project has been running since 2016.
Publications
2019
| Hoppenstedt, Burkhard and Reichert, Manfred and Kammerer, Klaus and Spiliopoulou, Myra and Pryss, Rüdiger (2019) Towards a Hierarchical Approach for Outlier Detection inIndustrial Production Settings. In: EDBT/ICDT 2019 Workshops, Lisbon, March 26, 2019, Proceedings of the Workshops of the EDBT/ICDT 2019 Joint Conference. (Accepted for Publication) |  |
2018
| Hoppenstedt, Burkhard and Pryss, Rüdiger and Stelzer, Birgit and Meyer-Brötz, Fabian and Kammerer, Klaus and Treß, Alexander and Reichert, Manfred (2018) Techniques and Emerging Trends for State of the Art Equipment Maintenance Systems - A Bibliometric Analysis . Applied Sciences, MDPI, Vol. 8, pp. 1-29, 10.3390/app8060916. | |
| Hoppenstedt, Burkhard and Pryss, Rüdiger and Kammerer, Klaus and Reichert, Manfred (2018) CONSENSORS: A Neural Network Framework for Sensor Data Analysis. In: 26th International Conference on COOPERATIVE INFORMATION SYSTEMS (CoopIS 2018)), Valetta, Malta, October 22-26, LNCS, Speinger. (Accepted for Publication) | |
2017
| Hoppenstedt, Burkhard and Pryss, Rüdiger and Treß, Alexander and Biechele, Bernd and Reichert, Manfred (2017) Datengetriebene Module für Predictive Maintenance. ProductivITy, Vol. 22, pp. 21-23. | |
Master & Bachelor Theses
2018
| Väth, Thomas (2018) Exploring Temporal Data in a Mixed-Reality Application. Bachelor thesis, Ulm University. | |
2017
| Grabiec, Sebastian (2017) Developing a client-specific workflow for Predictive Maintenance. Master thesis, Ulm University. | |
| Grausz, Krisztián (2017) Log Analyzer for IoT Applications. Master thesis, Ulm University. | |
| Kuhaupt, Nicolas (2017) Conception And Analysis Of A Raspberry Pi Cluster With Apache Spark. Master thesis, Ulm University. | |
| Salonikidis, Georgios (2017) Minimization of Redundant Make-To-Stock Production in a Dental Factory: An Integer Linear Programming Approach. Bachelor thesis, Ulm University. | |
| Schwarz, Holger (2017) Evaluierung neuronaler Netze auf Maschinendatenbasis. Master thesis, Ulm University. | |