@inproceedings{girsule_data_2020, address = {Vienna (Austria)}, title = {Data {Acquisition} {Approaches} for {AI}-supported {Metal} {Processing}}, url = {https://ieeexplore.ieee.org/document/9211935}, doi = {10/ghs5v9}, abstract = {Due to increasing digitalisation, it is possible to digitally map the production of sheet metal profiles from configuration to production. In our research project we design, implement and evaluate a knowledge- and rule-based system in cooperation with a sheet metal profile manufacturer using a modern production machine. One big challenge hereby is the data acquisition to perform a producibility assessment. In many cases only the positive (producible) production data is stored and negative data is discarded during the production process. In this paper, we present approaches to generate and collect negative training data for a machine learning approach using a data generator and feedback from manufacturing experts to perform a predictive manufacturing assessment.}, language = {en}, booktitle = {{ETFA} 2020 - {IEEE} 25th {International} {Conference} on {Emerging} {Technologies} and {Factory} {Automation}}, author = {Girsule, Bernhard and Rottermanner, Gernot and Jandl, Christian and Kreiger, Mylene and Moser, Thomas and Fuchs, Patricia}, year = {2020}, note = {Projekt: Wikant}, keywords = {Eintrag überprüfen, FH SP Data Analytics \& Visual Computing, Forschungsgruppe Digital Technologies, Forschungsgruppe Media Computing, Green OA, Institut für Creative Media Technologies, Institutional Access, Smart Manufacturing, Wiss. Beitrag, peer-reviewed}, pages = {4}, }