Digital twin enables individual solutions from standard parts

Christoph Bayer manages ECAD-PORT, a company which supports manufacturers in developing and operating component databases. Exceeding the boundaries of individual product portfolios, such systems provide all the information required by users of MCAD or ECAD in order to design a product. HARTING supports the E³.Series software of Cadenas and Zuken by contributing interface data and ECAD-Port is a certified service provider to E³. In this interview, Bayer discusses the advantages of component libraries for digitized production and the role of digital twins.

tec.news: What are the current trends in Computer Aided Engineering and Online Part Libraries?

C. Bayer: “Configure-to-order” (CTO) procedures are becoming more and more popular. Here, the customer configures a product according to his requirements, using the manufacturer’s developments as a basis. The rest of the process is automatic: from ordering the parts, setting up the workstations and manufacturing the products, to digital quality control, sales and service processes. All these procedures result in new product configurations based on high-quality CAx libraries. Currently, many companies work on documenting their components and improving the quality of their data sets. This “individual” production needs consistent product master data, not only to meet the producer’s requirements but also to assist the users with digitisation.

tec.news: What role do digital twins play in industry today?

C. Bayer: At the moment, CTO exists mainly in the automotive industry, where the procedures are often associated with the use of digital twins. However, there are also other branches of mechanical engineering where it has become more and more important to provide design data for follow-up processes and to develop existing data models on the basis of experience. Within the company, the orientation towards the digital twin leads primarily to changes in the communication structure: departments such as research and development, electrics, mechanics, hydraulics etc. must agree on a common electronic database. Therefore, it has become more common that systems like PLM (development) and EAP (manufacturing) are integrated, assuming that they use the same standardised formats and codes, for example, ECAD.

tec.news: How do you define the digital twin?

C. Bayer: The digital twin is the pattern and the image of a product in data form. It grows during the lifecycle, simultaneously with the realisation of the product: from model development and customer delivery up to decommissioning and disposal. The “data double” not only contains the relevant technical characteristics of the product, but also the serial number, the batch, warranties, possible call-backs and lists of installed components. The twin can also be fed with performance data, which is important for machine diagnostics and predictive maintenance.

tec.news: Which requirements must manufacturers meet, if they want to provide the relevant product data for design automation to their customers?

C. Bayer: First, manufacturers have to eliminate the media breaks between their subdivisions as internal communication must be accessible for all departments. All company units must have access to binding, automatically generated parts lists. The intelligence of Product Lifecycle Management (PLM) is also incorporated. The digital twin feeds design automation programs such as MCAD and ECAD and forms the reference point for all the steps up to the provision of a product. There is only one “single source of truth”, which is held constantly up-to-date. Users only need to have the right development and data management systems to tap into that source.

tec.news: What are the benefits of working with a digital twin?

C. Bayer: The users can now concentrate on the developments in their core area of competence. They receive all the required data and can be sure that the components they choose, e.g interfaces, fit in the intended environment. In addition, the digital twin makes it easier to commission and produce new batches of existing parts. The users can also commission the manufacturer with services for the component. Thus, data twins speed up the removal of parts and make state of the art binding for everyone involved.

tec.news: What else do you think belongs to the digital twin? Which are the next development steps?

C. Bayer: The product libraries should be barrier-free within the communities of those involved and the manufacturers must keep sovereignty over their data. Content management systems allow the digital twin (and its parts) to be made available to parts of the community. Access to product libraries can be further filtered according to categories like “external” and “internal”. Communication with other systems requires a uniform data format, which helps to interpret even complex designs (e.g. .xml).

Digital twins do not only accompany their products up until the point of delivery, they live on. Starting from the 3D model, the “original” product design, they are enriched with data from all the production process. After ordering, customers can receive a compilation of all the data that is meaningful to them. Thus, they get the chance to enrich the digital twin further by adding their performance data, making it a powerful tool for predictive maintenance. Users can easily send feedback to the manufacturer and make their experience available for the continued development of the product.

tec.news: Where does the journey go with multi-vendor online product databases?

C. Bayer: In the future, the mass of companies in the industry will establish and use product data libraries. The goal is to design uniform descriptions of components, so that we have consistent coordinates and can improve the comparability of products. Electronic product databases exceed the limits of individual companies. The documentation of the parts purchased automatically flows into the development, production planning and control systems, meaning permanent feedback to the manufacturers is possible. Binding data quality and standardised communication flows ensure that vendor parts actually fit into the applications.

In the future, the workflow for design will look like this: via online portals, the designers of electrical components research the product ranges of component manufacturers. They select the parts they need and trigger the data transfer into their own design programs. All relevant information such as metadata, symbols and product designs are transmitted in standardised form. In order to integrate a component into their product, the users simply order and wait for the component to be delivered.

tec.news: What is the relationship between the digital twin and the benefits of AI mechanisms?

C. Bayer: It all depends on how far you push the question. With the help of the digital twin the predictive maintenance of a machine can be operated. At a certain point in time, the need for an oil change will occur, e.g. due to load changes or operating conditions. If the condition recurs, oil changes will be scheduled in a regular cycle. This cycle allows the machine to operate more effectively and efficiently. The basis for these processes is definitely Artificial Intelligence.

tec.news: Where exactly is the AI ​​located in such a manufacturing scenario?

C. Bayer: The learning unit, the Artificial Intelligence, can only be in the cloud. It has to learn continuously from all the “points of reference” of the product; regardless of whether this is the service workshop that documents unnatural mechanical wear, or the development department that initiates product recalls or determines load cycles.