ELAMU

Project motivation:

In many industries, it is essential to have precise information about the exact chemical composition of materials. This information is passed along the entire supply chain and continuously supplemented. Despite some efforts to represent these data in a standardized digital format, the information, which is crucial for companies, is often still transmitted in the form of unstructured and inconsistent material certificates on paper. Using Optical Character Recognition (OCR) and Intelligent Character Recognition (ICR), characters displayed in documents can be automatically identified. Companies use OCR/ICR to extract information from documents into structured formats, thereby partially automating the media break between paper-based documents and downstream processing systems.

OCR/ICR-based recognition of classic, well-structured documents such as purchase orders or invoices already works relatively reliably and can lead to significant cost and time savings by supporting or replacing the purely manual extraction of master data from invoices and receipts. However, due to the lack of standards and the high variability among different material certificates, the current challenge lies in applying OCR/ICR technologies to unstructured material certificates.

Objectives:

The main objectives of the project are to increase the level of digitization of material certificates. The focus is on significantly reducing the error rate and costs associated with the manual processing of material certificates.

At the same time, data quality and availability for complementary downstream processes are to be improved in order to minimize manual documentation efforts. To make the use of machine-learning-based methods manageable for end users at the user interface level, methods and experiences from the fields of End-User Development (e.g., visual programming, adaptability interfaces, collaborative tailoring) and appropriation support (e.g., usage discourse environments, extensible help systems) will be integrated into the project.

The project has an innovative character for metal-processing companies. To achieve the project objectives, an interoperable platform will be developed that processes complex and heterogeneous metal certifications as unstructured data. Follow-up systems such as ERP or CAQ systems will be connectable via interfaces. In addition, field-tested interaction concepts will enable intervention and supervision by employees during the processing workflow.rvention by employees.

The project is funded by the German Federal Ministry of Education and Research (BMBF).

Important partners in the project are NeurologIQ Engineering GmbH(https://neurologiq.com/de/), Robert Josef Wolf GmbH & Co. KG(https://rjwolf.de/), AVENTUM GmbH(https://www.aventum.de/) and the University of Siegen.

Additional partners are ifm business solutions gmbh(https://www.ifm.com/de/de), Bruse GmbH & Co. KG(https://www.bruse.de/), the Mittelstand 4.0 - Kompetenzzentrum Siegen(https://www.uni-siegen.de/smi/kompetenzzentrum/?lang=de) and the ZDW Südwestfalen(https://zentrum-digitalisierung.de/)