HANNOVER MESSE 2019, 01 - 05 April
Homepage>Exhibitors & Products >Predicitve Quality

Predicitve Quality

When deep learning takes over quality assurance...

Logo Predicitve Quality
Logo Predicitve Quality

Product description

Predictive quality is a new deep learning process that predicts component quality based on numerous types of available data. Many companies are already monitoring each component throughout the production process and checking it multiple times using sensors. All of this sensor data regarding flawless and faulty components is imported into the self-learning deep learning system from it-RSC over a period of several weeks. This allows error patterns to be detected with increasing speed, which results in faulty parts being sorted out early in the production process. At the same time, the system yields important insights for the technicians involved, thereby optimising the production process.

Predictive quality through deep learning is suitable for all manufacturing enterprises. It optimises value added by completely avoiding the unnecessary costs of refining faulty components. Predictive quality is the short cut to error-free quality so to speak.

The bottleneck for deep learning is often the IT systems in the companies. That is because the existing IT platforms are generally unsuitable for networking large volumes of data properly and performing meaningful big data analyses. In many cases, this data is therefore merely gathered and archived. To make it easier for companies to take the first step towards a predictive quality project, it-RSC GmbH simply provides its high-end vCAX virtualisation environment on loan in a flight case. In this way, the project can commence independently of the company s available IT infrastructure, without any major IT expenditure.
With the help of predictive quality and vCAX, it-RSC is already rocking the production processes of some major customers in the manufacturing industry.

Hall 6, Stand C30

Stored items


Server communication error: Item could not be saved.