Due to the ever-growing importance of flexibility in industrial automation (think about shorter innovation cycles, rapidly changing customer needs, more sustainable systems, continuous changes in legislation, and emerging cutting-edge technologies), the design, realization, operation, and maintenance of industrial automation systems are becoming more and more challenging.

As an answer, digital twins are emerging and gaining attention to support different phases of the system lifecycle such as design-space exploration, runtime adaptation, and predictive maintenance. While benefits of using digital twins are already demonstrated for different settings, digital twins are complex software systems themselves. Thus, their development, operation, and evolution pose novel engineering challenges.

In my talk, I will summarize different research endeavors we have followed to tackle the challenges in engineering digital twins with model-driven methodologies to close the gap between the design and operation of systems.