A digital twin is a virtual representation of a physical object or process. It is primarily used when companies want to observe, analyze, simulate, and optimize their products or systems over their entire lifecycle. To do this, models, information, and data are continuously collected from the real object or process in real time, providing insight into its condition and behavior.
A digital twin has three basic components: A digital master, a digital shadow, and their intelligent linking.
The digital master contains all relevant models of the physical object or process. It uses appropriate data and information models to represent what is expected. Depending on the application, this may relate to the geometry, behavior, function, or other properties of products, machines, or processes.
A car is used as an example. All cars of a type have been produced using the same drawings and production instructions, and have the same average maintenance cycles. We call the models, simulations, and data from which all these cars are produced »master«.
The digital shadow, on the other hand, represents the actual. It consists of data that is collected over the life cycle of the system depicted. This can be operating, status or process data that is recorded by sensors, for example.
The data shows the differences between each car, even though they are identical in construction. This is due to the fact that each car has been subjected to special features during production, such as assembly variations. During operation, the car may have been used in particularly hot, cold or dry regions, requiring shorter maintenance cycles. We call this data, which represents the reality of the unique car, »shadows«.
The real value of the digital twin comes from intelligently linking the digital master and the digital shadow.
By comparing the vehicle's master and shadow data, it is possible, for example, to predict the optimum time for maintenance or to derive important improvement potential for future product generations (feedback to design).
Our researchers have developed a pilot system on which linked production systems can be designed cost-effectively and complex processes safely tested. The principle: A digital twin of the modular development environment ensures that the two robots are automatically configured and controlled for each assembly step.
using the example of the automation of an assembly process for professional tools
Time and cost savings of a robot-based system compared to a traditional pilot system:
The industry has recognized the promise of digital twins as a future technology. According to a Gartner study, only 13 percent of large companies with ongoing IoT projects in major industrialized nations are currently using digital twins. However, 62 percent of respondents have such projects at least in the planning stage.
But even in companies where digital twins are in place, they are far from being used to their full potential. So far, they are mostly used as data supply systems or for security and error analysis. This is the conclusion of the »Digital Twin Readiness Assessment« study conducted by Fraunhofer IPK and msg in 2020, which found that the provision of automated value-added services and the design of autonomous or adaptive systems have only been considered in a few concepts to date.
To fully exploit the potential of digital twins, we develop customized solutions in close cooperation with our partners and customers. We network digital twins with each other and thus enable the mapping of entire production lines with minimal time and financial expenditure. In doing so, we always keep an eye on the sustainability aspect, as the technology also offers new possibilities in this area.
Here we show some examples of problems from industrial applications that can be solved with the help of digital twins.
Fraunhofer Institute for Production Systems and Design Technology