Speaker
Description
In machining processes, measurement data collected during operation can provide valuable insights into the final quality of the manufactured components. This enables both the reduction of unnecessarily long process chains and the real-time adaptation of process parameters. Machine learning models trained on this data can capture and predict complex process characteristics. However, acquiring comprehensive measurement data is often expensive and technically challenging, making resource-efficient deployment strategies essential. One promising approach is the use of resource-aware rejection ensembles. By categorizing measurement data into low-cost, contactless measurements; medium-cost, integrated measurements; and complex but highly reliable measurements, it becomes possible to optimize rejection strategies, for example, through the targeted selection of measurement devices based on the uncertainty profiles of different predictive models.
