OEM&Lieferant Ausgabe 2/2021

90 Our Offer for SMEs and large enterprises https://t1p.de/vjsn Share development phase at the laboratory scale before they are implemented at an industrial scale and a further ramp-up phase before they are capable of production with the re- quired quality, productivity and cost-effec- tiveness. This is especially the case when new materials and production technologies are employed, when the manufacturing pro- cess has a vast number of parameters and high-dimensional state-spaces, and when only poor dynamical process models are available at the outset. Traditionally, domain experts accomplish this via extensive exper- iments based on their knowledge, experi- ence, and intuition. Nevertheless, acquiring experimental data from manufacturing pro- cesses is associated with large costs due to, e.g., downtimes of expensive infrastructure, resource consumption for non-marketable output and costs for destructive testing of output products in dedicated laboratories. Fraunhofer is convinced that the systematic use of AI leads to significant improvements in the maturation of manufacturing process- es, both in terms of the solution quality and the resource investment needed to arrive at a given maturity. To transform immature manufacturing processes into mature pro- cesses using modern ML algorithms and tools Fraunhofer e.g. has invested in the Karlsruhe research factory, offering a re- al-life testbed for AI in manufacturing. State-of-the-art measurement and control technology implies that the relevant pro- cess parameters and product features are captured in line, i.e. in pace with the current process using appropriate sensor technolo- gy. They are used to monitor machinery and equipment and to improve processes. This is based on the precondition that the applied sensors are appropriate for the process and the product and are interconnected so the data can be evaluated. In measurement technique, Fraunhofer supports machine builders by deploying image-based measurement technology for process intelligence and product-related quality assurance in their manufacturing operations, for instance. To inspect (partial- ly) reflective surfaces, we have enhanced deflectometry so it can be applied as an inspection technology in industry, capable of inspecting large and small object in line with a high degree of precision. Depending on the machine builders specific require- ments, we advise them even in the design phase of a newmanufacturing installation to support them in integrating state-of-the-art visual inspection methods in their system so they can achieve full quality and productivity right from the start. Fraunhofers neutrality as a research institute ensures that compa- nies receive the inspection and measure- ment technology that is just right for their requirements, independent of suppliers. In control technology, we model processes in an analytical, knowledge-based or data-driv- en way. The resulting models can then be used for process improvement, monitoring and control. Digitals twins is another focus of our proj- ects for the mechanical engineering sector. The concept is based on modeling assets with all their geometrical data, kinematic functionality and logical behavior using digital tools. The digital twin refers direct- ly to the physical asset and allows it to be simulated, controlled and improved. Today, digital twins are being discussed in Industry 4.0 working groups in the context of asset administration shells or Industry 4.0 compo- nents. From Fraunhofer’s point of view, dig- ital twins will become a major topic for de- velopment over the coming years because they are not single objects or monolithic data models, but include different aspects of digital representations, functionalities, models, interfaces etc. During the R&D efforts for Fraunhofer’s SmartFactoryWeb (SFW), an industrial platform for smart factories, which is at the same time the official test bed of the Industrial Internet Consortium (IIC) and a longterm cooperation project with the Ko- rean partner KETI, Fraunhofer engineers have come to the conclusion that these plat- forms can actually have disruptive effects on the manufacturing industry. SFW aims at improving the value chain by flexibly equal- izing the capacities between the smart fac- tories participating in the platform. To this end, the factories register with the SFW portal, allowing customers to find appro- priate production capacities. By now, SFW even provides features to manage supply chains and networks. Today, various man- ufacturing-as-a-service (MaaS-) platforms have established themselves in the market, offering the production of parts as a service – usually still in the form of NC chip mak- ing, 3D printing or sheet metal production. Manufacturers can join these platforms by offering their resources and thus their pro- duction capacities; the platform carries out all management activities. Based on the 3D data provided by the customer, the software automatically calculates the price and the delivery date and assigns the manufactur- ing order to one of the participating factories. Thus, end customers do not have any direct contact with the manufacturer any more, they only access the platform. From Fraun- hofer’s point of view, this scenario poses, as a matter of fact, a threat to small and medi- um-sized manufacturers as they increas- ingly depend on the relevant platform. They are no longer in direct contact with their customers and owing to maximum trans- parency, competition is reflected almost exclusively in the price. The main question is whether data from production lines and factories are passed on to third parties only in specific cases of application. The control of the use of the data has to remain with the owner of the geometries and/or the relevant manufacturer. For this reason, Fraunhofer supports the aspirations of the International Data Spaces Association (IDSA), which is establishing the Industrial Data Space (IDS) as a secure and sovereign network for data exchange. The proportion of software and automa- tion in machines and manufacturing equip- ment is increasing continuously, whereas the share of mechanics is decreasing in terms of value. Fraunhofer advises manu- facturers of machinery and equipment in the creation of future-oriented IT architec- tures as well as the selection of suitable IT technologies, tools and functionalities. Yet experience has shown that this is not sufficient. Just like the development of a machine, software development requires an engineering-related process including well-defined phases, release mechanisms, project management and documentation. For this reason, it is necessary to tailor the management of professional software engineering to the specific needs of the company. The mechanical and plant engineering sector has to reinvent itself Unlike many other industries, digitalization is still at an early stage in manufacturing. In the future, factories and their equipment will rely on realtime-compliant software that can be interlinked spontaneously, not only inside a company, but throughout the entire supply chain. This will have major conse- quences for the structure of added value in the mechanical and plant engineering sec- tor. Europe should aim at maintaining and strengthening the mechanical and plant engineering industry, for countries need a strong industrial basis for future economic turbulences. Furthermore, their manufac- turing equipment suppliers should support international standards for easy-to-use con- nectivity, data security by design and the generation and use of digital twins for cus- tomers around the world. 1) MES manufacturing execution systems, MOM manufacturing operation management Digital Twin https://t1p.de/ad97