Digital twins are an emerging concept with the potential for revolutionising the way we interact with the physical world. They have already been applied successfully in areas such as Industry 4.0 or traffic management. Digital twins are leveraging the digitalization of increasingly more application domains and are intensively using various models of different types (descriptive, predictive, and prescriptive) together with the ever-increasing available data to enable in silico inspection and manipulation of real-world systems (including physical systems and objects as well as social systems and organisations).

The concept has seen strong interest in industry, where there is a desire to control increasingly complex systems of systems, ensuring they behave as expected and to control their adaptation to the environment or any deviations with the initial plan. Digital twins can be used for improved analysis and understanding of complex systems (in silico experimentation) as well as for control and transformation of these systems. Digital twins are themselves complex software systems, posing novel software-engineering challenges, which have so far not been sufficiently addressed by the software-engineering research community.

There is a need for solid foundations to ensure the development of tools and methods according to well-established principles. We believe that Model-Driven Engineering (MDE), will be a key technology for the successful systematic engineering of Digital Twins. In this Dagstuhl Seminar, we aim to bring together both practitioners and researchers to (i) reflect on the concept of Digital Twins and the software-engineering challenges posed, (ii) identify relevant existing MDE approaches and technologies that can help tackle the challenge of systematically engineering digital twins, and (iii) define an academia–industry research roadmap for systematic engineering of digital twins based on MDE.

As the intended primary goal of the seminar is to create a community and establish a research roadmap, we are interested in discussing the following topics:

1. Challenges faced in real-world development of Digital Twins.
2. Opportunities offered by MDE.
3. Active exploration of collaboration opportunities.

A key outcome of the seminar could be a research roadmap for the new softwareengineering discipline for Digital Twins.

Creative Commons BY 4.0

Benoit Combemale, Vinay Kulkarni, Bernhard Rumpe, Steffen Zschaler, Tony Clark and Manuel Wimmer