Design process is crucial

Electronic components are essential for digitalisation and are required in large quantities. Rare raw materials are often used in production. The aim is therefore to use resources as sparingly as possible. In her dissertation, Mona Fuhrländer is developing simulation-based processes that can be used to reduce rejects in the production of electronic components and their probability of failure during

Uncertainties can arise due to inaccuracies during production, natural material fluctuations or environmental influences. Using various machine learning methods with error control, the young scientist from Darmstadt has developed efficient methods for the simulation-based quantification of uncertainties and for design optimisation under uncertainties. Thanks to these mathematical methods, waste can be greatly reduced and resources saved.

In her dissertation, she looks at the design of the permanent magnets of an electric motor as a concrete application example. As these magnets are made of rare earth metals, it is particularly important to conserve resources. Furthermore, greater safety is guaranteed, as the robustness optimisation also reduces the probability of failure during operation. In addition to the reduction in waste, reliable running performance is guaranteed under increased performance requirements.

In contrast to conventional methods, the processes developed by Mona Fuhrländer are significantly less computationally intensive while maintaining the same level of accuracy, making them suitable for use in industrial product design.