Webseite Fraunhofer-Institut für Produktionstechnologie IPT
The Fraunhofer-Gesellschaft (www.fraunhofer.com) currently operates 76 institutes and research institutions throughout Germany and is the world’s leading applied research organization. Around 30,000 employees work with an annual research budget of 2.9 billion euros.
At the Fraunhofer IPT in Aachen, we work with more than 530 employees every day to make the production of the future more digital, more flexible and more sustainable. In the department for production quality, we focus on the digitalization of production systems to increase quality, resilience, and sustainability of production.
The optimization of production processes is becoming increasingly important, but also more complex, due to rising demands on the sustainability of processes. Bayesian optimization (BO) – a particular machine learning method – represents a promising alternative to traditional process optimization methods due to its adaptive decision making and data efficiency. A key challenge is the configuration of the BO-method (i.e., surrogate model and acquisition function) depending on the characteristics of the optimization problem. To establish Bayesian optimization in production engineering practice, this thesis aims to analyze which BO-methods are well suited for different optimization problems with respect to an accurate and efficient optimization.
What you will do
- Literature research on Bayesian optimization and optimal experimental design in production processes
- Analysis of different BO-algorithms and their dependencies on process optimization properties
- Derivation of a set of properties for characterization of production processes
- Development of a methodology for optimal configuration of the BO-algorithm regarding the defined properties of the production process
- Verification and validation of the methodology for a freely selectable use case from optics production, laser processing or biotechnology
- Documentation of the results and writing of the scientific work
What you bring to the table
- You study mechanical engineering, computer science, mathematics or a comparable subject
- You are interested in the field of production engineering and the transfer of theoretical concepts into real-world applications
- You are familiar with the theory and the approaches of machine learning and would like to specialize in this subject area
- Your strength is analytical thinking, and you have good abstraction skills
- A high degree of motivation, willingness to learn, and independence
- Good language skills in German and/or English
What you can expect
- Application-oriented research in the integration of machine learning into industrial production processes
- Ideal framework for practical experience alongside studies in a renowned research institute
- Collaboration in a multi-member team within an exciting research project in cooperation with several Fraunhofer institutes
- Excellent equipment of machines and devices
Questions according to this position will be answered by:
Lars Leyendecker M.Sc.
Research Assistant Production Quality
Phone: +49 241 8904-314
Um dich für diesen Job zu bewerben, besuche bitte jobs.fraunhofer.de.