Prostep | Newsletter

PROSTEP evaluates scan data in the DigiTwin project

By Dr. Josip Stjepandic

Digital twins make it possible to perform material flow simulations for plant layout and bottleneck analyses. Building digital twins for existing production systems is, however, extremely complicated. As part of the DigiTwin joint project, PROSTEP and three partners are developing a procedure for creating digital simulation models from the 3D scan data generated by production systems largely automatically.

Material flow simulations for bottleneck analyses, plant layout and inventory analyses help improve operational workflows. Up until now, developing corresponding simulation models was extremely complicated, making it difficult for small and medium-sized companies to use them. Digitalization, however, offers new possibilities for simulating and optimizing the real-life situation in production with the help of a digital twin. In the DigiTwin project, the Institute of Production Engineering and Machine Tools (IFW) at the University of Hanover, together with PROSTEP, isb - innovative software businesses and Bornemann Gewindetechnik, are examining how digital twins for existing production systems can be created more easily.

The research project, the full name of which is "DigiTwin - Effiziente Erstellung eines digitalen Zwillings der Fertigung" (Efficient Creation of a Digital Twin for Production), is being funded by the "SME innovation: Service research" initiative of the German Federal Ministry of Education and Research. Within the framework of the project, the partners are developing a service concept for deriving simulation models from scans of the factory floors largely automatically. The idea is to use object recognition to convert, with a maximum of automation, the 3D scan data from production into digital models that can be mapped one-to-one in the simulation software. The aim is to make both the layout of the production facilities and the logic of the production processes transparent.

In the project, PROSTEP is responsible for transforming dumb point clouds of machines, robots, transport equipment, etc. into intelligent CAD models that can then be used to simulate the manufacturing  processes. With the help of methods from artificial intelligence and machine learning, the solution uses the point cloud, or the network geometry derived from it, to identify similar system components, which are stored in a library together with their CAD models.

It is intended that system components for which there is no equivalent in the library be converted into CAD models and parameterized with the help of feature recognition so that they can be prepared for simulation. This means that the simulation models can easily be adapted to take account of company-specific characteristics. PROSTEP's data management team will make the services for object recognition, object harmonization and conversion available via the data logistics portal

Production systems generally vary from company to company. Company-specific machine configurations and special chaining logic cannot, of course, be derived directly from the scan data, which is why the scientists at IFW query this data using standardized forms. This minimizes the amount of time and effort needed to adapt the simulation models, thus also ensuring that the concept remains attractive to small and medium-sized companies. The project partners need only a few days to create a digital twin that can also be adapted quickly in the event of changes to production. As this is a service concept, no programming knowledge is required on the part of the customer.