Weidmüller industrial data analysis solutions help GEA

GEA, a long-standing supplier of condition monitoring, launched a "pilot" project in Erde, Germany, with the launch of Weidmüller's latest automated machine learning software, looking to develop and expand the services offered in the field of equipment.

Weidmüller industrial data analysis solutions help GEA

　　Digital skills and Industry 4.0 present challenges and new opportunities for companies in the mechanical and engineering professions: they need to adapt their production facilities to individual products and customer requirements. Service matters are becoming more and more important.

"For a long time, we have been working on machine condition monitoring and have developed a threshold profiling approach. We also know that there is more potential in this area. Kerstin Altensuer, Senior Vice President Service Product Management at GEA, explains, "We want to draw up process maps and optimize our processes together with our customers. Of course, we also want to establish new business forms and applications, such as machine rental or ordering. ”

GEA has an excellent reputation for 125 years of experience in the manufacture of liquid separators and decanters. These products are used in a variety of occupations, such as the food, chemical and pharmaceutical industries, as well as in the biological, energy, shipping and environmental skills industries.

GEA expects to discover more revenue sources through the creation of new business forms or applications. Kerstin Altensuer said: "We realised that we needed the expertise and assistance of data experts in these projects. It's not easy to find the right algorithm experts and recruit them to the team, which makes the work even more cluttered. ”

GEA expects to discover more revenue sources through the creation of new business forms or applications. Kerstin Altensuer said: "We realised that we needed the expertise and assistance of data experts in these projects. It's not easy to find the right algorithm experts and recruit them to the team, which makes the work even more cluttered. ”

GEA wanted to design its services to its customers from the ground up and build a range of smart services. Improving the quality and performance of its equipment is the basis for discovering new forms of business that will make GEA competitive in the market.

Tobias Gaukstern, De Miller's Industrial Anatomy Services Manager, explains: "It soon became clear that our primary requirement was to go through a proof of concept (PoC) to verify the intent of the project before we could assist GEA in developing and operating the ML model independently. ”

In the future, GEA's experts will be able to practice machine learning algorithms or models independently with active machine learning software services. Tobias Gaukstern explains: "The AutoML stuff makes the process of using ML faster and easier with programmers, and it doesn't require any ML expertise. ”

Mechanical engineering companies are often faced with the problem that their design, automation, and process experts can't easily translate their common sense into machine learning-based solutions. At present, it is difficult to bring together the expertise of these applications or in one software, let alone algorithms.

Matthias Heinrich, Division Manager Electronic Solutions at GEA, explains: "We were very interested in this solution because we have a lot of professional engineers who are very familiar with the machine and can interpret the data to a certain extent. With Weidmüller's help, we can now turn this common sense into an algorithm. ”

To test the practical application of theoretical observations in GEA's production environment, we conducted a Proof of Concept (PoC) using historical data in Erder to see if it is possible to proactively detect anomalous behavior in dairy occupational separators.

The success of this intention is due to close collaboration within the team. On the one hand, the proximity of the area is a great advantage, as the project team can easily meet at short notice to discuss individual issues.

Kerstin Altensuer explains: "From a data scientist's point of view, Weidmüller has a very broad sense of knowledge. At the same time, as a mechanical engineering company, you will feel like you are meeting a soulmate, because you will be meeting not only IT experts, but also engineers who really know the machine. ”

As part of this intent, GEA manages the data entry and requirements profiling, while Weidmüller is responsible for the proof of concept. Tobias Gaukstern explains: "The collaboration was a great success. Our orderly and positive harmony has achieved very good results, which has laid the foundation for the pilot application and final transformation into series production. ”

Weidmüller industrial data analysis solutions help GEA

Sections viewed

Contact

working hours