Digital twins seem well suited to the manufacturing industry and could even revolutionize it in the form of more efficient production and better maintenance. But how do you deploy this technology as an SME? They often lack the knowledge and insight to invest in twins. However, Teade Punter sees plenty of possibilities and opportunities for the digital twin in SMEs.
A digital twin is the digital version of a physical asset, system, object or process, or asset: something worth collecting data about. A twin creates a digital representation of the real asset by using the data from the asset's sensors. The real asset and the virtual asset are the same. If something changes on the physical asset, the same is true in the twin.
Digital twinning is very similar to simulation in that it also imitates reality. Digital twinning adds an element to this; namely, a continuous connection between the virtual and real world. The digital representation is the real system, making it possible to directly see the system status and control it through the asset's actuators.
A digital twin can be used to develop new assets, as is done in aerospace, where the origin of the term also lies. This article assumes the digital twin as a derivative of existing assets, as developed in the manufacturing industry.
A digital twin is created by collecting data about the asset. This data comes from the asset's sensors and is visualized with the twin. Therefore, twin development starts with the data structure of the asset. In Digital Factory, methods such as RAMI4.0 and Asset Administration Shell (AAS) are being promoted for this purpose. Some companies are already picking this up, while others are more cautious. In addition to the data structure, it is often also important to look at the behavior of the asset. This might include "agents" that perform specific tasks of the asset, such as keeping track of inventory. Asset behavior requires modeling, a description of the states, with the conditions under which they occur.
Three phases are recognized for developing and applying digital twins: data logging, asset depiction and integration of asset and model.
Data logging forms the basis of any twin development: data is collected via (IoT) sensors. For many companies, this is often already a big step, and thus should not be underestimated.
A digital twin can be developed for a variety of reasons. For SMEs, the following are possible motivations for starting a twin.
Because the twin is a copy of the real system, the twin can serve as a remote monitoring tool for companies that have remote machines at customers' premises. Thanks to a digital twin, a service operator can watch directly when problems occur at one of its customers. Maintenance can thus be carried out more efficiently and possibly better informed.
The twin can also be developed to collect data on machine parts or machine usage to recognize patterns. For example, the wear of parts can be predicted and their replacement scheduled without interrupting the manufacturing process at undesirable times: so-called predictive maintenance.
The twin can also be used to collect data. The phase data logging is then done from the digital twin. Data is collected about (simulated) situations in which you don't want to, or can't, bring the real machine because otherwise it would break down or cause danger to employees.
Yet another motivation is to allow multiple people to work on the machine simultaneously by having them work with a twin, thus training the real machine.
Finally, one motivation may be to align the assets of different companies. Companies can thus form a chain of assets, exchanging machine information and allowing machines to hook up to each other, creating new forms of cooperation.
With digital twinning, the manufacturing industry is digitizing. In addition to the opportunities, as described above, and risks such as security and IT dependence, twinning should be seen as an investment. Here, of course, it is important to determine in advance why the twin is being developed. Here, a cost-benefit analysis can help. However, the benefits of digital twins cannot be fully predicted before the actual twin implementation, because twinning also changes the picture on the possibilities with the assets, machines and processes. Digital twinning is thus primarily a decision for companies with an eye on and confidence in a digital future.
Author: Teade Punter is lector High Tech Embedded Software (HTES) and leading lector Fontys Centre of Expertise High Tech Systems and Materials (HTSM).
Fontys Expertise Center High Tech Systems and Materials (HTSM) connects Fontys students and researchers with high-tech companies in the Brainport area.
Together they conduct applied research and develop innovative technologies that benefit both education and business. In addition, the Expertise Center HTSM offers various forms of lifelong learning for professionals working in the high-tech sector.
Willem: "However, we know that successful AI projects require more than just technology. You only achieve the best solution together with the customer!" And that, then, is the immediate challenge for all companies: get help!
Dutch