New working group on predictive quality: Reducing testing efforts

Together with the Fraunhofer Institute for Production Technology IPT in Aachen, the Machine Tool Laboratory WZL at RWTH Aachen University has launched a new industry working group for "Predictive Quality". The aim is to significantly reduce joint inspection efforts on a pre-competitive basis and to realize higher productivity by eliminating physical inspection processes, as well as to continue to achieve higher quality through a [...]

Predictive Quality
Predictive Quality Demonstrator at the Chair of Production Metrology and Quality Management at WZL. (©WZL)
Together with the Fraunhofer Institute for Production Technology IPT in Aachen, the Machine Tool Laboratory WZL of RWTH Aachen University has launched a new industry working group for "Predictive Quality". The aim is to significantly reduce joint testing efforts on a pre-competitive basis and to realize higher productivity by eliminating physical testing processes, as well as to continue to achieve higher quality through a reduction in scrap, end-to-end quality monitoring and knowledge generation from the models, and increased sustainability through more resource-efficient production.

Predictive quality reduces testing efforts

More and more data is becoming available to modern quality management faster and faster. At the same time, advanced algorithms are enabling ever more detailed images and models of production. These data and models form the basis for the field of predictive quality. Predictive quality describes the data-based prediction of quality characteristics. Using a learned relationship between process parameters and quality characteristics, complex physical inspection processes, which are often only carried out in random samples, can be replaced by a low-cost model-based 100% inspection. Predictive quality has already been successfully implemented in industry-related research projects, in which inspection costs have been significantly reduced and productivity increased. has been increased. At the same time, more and more data-based quality management tools are being developed and used by manufacturing companies in digitalization and Industry 4.0 projects, software companies are providing advanced infrastructures for data collection and storage and start-ups are creating business models by providing corresponding algorithms for data evaluation.

Faster dissemination of research results

The two Aachen-based institutes support companies from the manufacturing industry (e.g. automotive, metal processing, chemicals, pharmaceuticals, medical technology) as well as software companies that specialize in the acquisition, storage and processing of data (e.g. CAQ, MES, sensor manufacturers, cloud providers) in the working group with their many years of experience. The industry working group is financed by an annual membership fee and helps new members to and utilization of research results and networking and is based on three pillars. Two community meetings are held each year to facilitate exchange between the members of the working group. Current findings and results from industry and research will also be presented at the meetings. One study per year is carried out within the working group on specific topics in order to gain insights into the current state of technology in the companies, challenges and new approaches. The topics are chosen by a majority vote of the working group members. In one demonstrator project per year, new ideas and approaches are specifically tested by the Machine Tool Laboratory WZL and Fraunhofer IPT. For example, various algorithms for quality prediction or preprocessing can be implemented and compared. The demonstrators can either come from the halls of the Laboratory for Machine Tools and Production Engineering WZL and Fraunhofer IPT or be provided by a company. Joint project results are available to the partners without restriction. Source and further information: www.wzl.rwth-aachen.de

This article originally appeared on m-q.ch - https://www.m-q.ch/de/neuer-arbeitskreis-zu-predictive-quality-pruefaufwaende-reduzieren/

More articles on the topic