Digital twins enhance quality via simulation

​Industry is continuously seeking efficient solutions that save time and improve quality. A five-year project in the Wingquist Laboratory focuses on simulations using digital twins, which will lay the foundations for the self-compensating factory.

The Wingquist Laboratory at Chalmers University of Technology in Gothenburg has focused on research in the field of digital product realisation since 2001. A key area in this laboratory is geometry assurance, which aims to minimise the impact of geometric variation in the final assembled product.

- This area is a key one for industry and specific to mass production, where all the parts need to fit together and be interchangeable, says Professor Rikard Söderberg, Director of the Wingquist Laboratory, and continues:
- In industry, variation is traditionally handled by specifying tolerances, that is to say how much variation is permitted. Tight tolerances often make production expensive, since a high degree of precision is required in the production process.

The geometry assurance software RD&T, which was partially developed at Chalmers, has been used by Volvo since 1998 and also by a large number of companies globally.

- RD&T is now being further developed so that it can be used as a digital twin of the production system we would like to control. We can then regard RD&T as a digital copy of the actual system, with the capability for real-time optimisation and control, Söderberg says.

About a year ago the centre was awarded a large research project, Smart Assembly 4.0, by the Swedish Foundation for Strategic Research (SSF). The five-year project focuses on the geometry assurance process and has as its vision the self-compensating factory.

- By using simulation in RD&T along with new capabilities for rapidly scanning component parts, in future it will be possible to take into account geometric variation and adjust the assembly process to compensate for this, he says, and continues:
- This may involve matching the right bits together, adjusting fixtures and equipment and choosing the right sequence in spot welding, for example. By feeding a simulation model with real-time data, it can be used as a digital twin which will be used in practice to control the process.

Söderberg, who has carried out research on simulation support for geometry assurance over more than 20 years, takes a confident view of the wave of digitalisation that is now washing over the industry.

- This wave of digitalisation will increase access to data. And with more input data we will produce better simulation solutions. All companies face competition and want to benefit from smart working methods, Söderberg says, and continues:
- Industry as a whole is becoming more and more interlocked. The companies we work with have a number of subcontractors that manufacture components.

- In future we may receive measurement data relating to components for a Volvo from suppliers and can use the information to make preparations before they are assembled. In the complex networks leading to a business transaction, there is a great deal of data that allows us to get the best out of the end product. If we can communicate digitally and optimise production and the production system without physical meetings and manual data processing, we can save a great deal of time.

Photo: Anna-Lena Lundqvist

Smart Assembly 4.0

Published: Wed 30 May 2018.