We support Voestalpine Stahl Donawitz GmbH in its mission to build and maintain a value chain that is consistently focused on quality. High quality in the starting material ensures high quality in the product. However, defects (e.g., cracks, inclusions) can occur in the starting material during the production process.
The exact location of the source of the defect is not always possible. Using numerous sensors, models were developed with the aid of machine learning methods to identify the causes of defects. The aim was to predict errors already in the production process and, at best, to avoid them.
The partial or complete mapping of production chains increases transparency and traceability, especially in the case of errors. Downtimes are reduced and quality and throughput are improved.