The technical basis for collaborative systems in smart factories and other “smart” areas (Autonomous Vehicles, Smart Grids) are cyber-physical systems (CPS). Modern CPSs are not only able to monitor their own environment and communicate with other systems, but they can also form dynamic networks whose functionality goes beyond the functionality of the individual CPS. They are dynamic because individual CPSs join or leave the networks during operation. In a smart factory, for example, individual robots come together for specific tasks or leave a network, e.g., to refuel energy at a charging station.
When developing such CPS, it must be considered that not all possible network constellations are reasonable or permissible. They are subject to certain constraints, which must be clear to all stakeholders involved in the development of the CPS. Until now, these constraints have been specified primarily textually by means of Object Constraint Language (OCL) or by additional comments in other development artifacts. This may have disadvantages for the common understanding of the stakeholders, for example, when the boundary conditions are discussed in the context of non-formal validation activities such as the FMEA (Failure Mode and Effects Analysis).
Therefore, Prof. Pohl's team, together with the industrial partner InSystems Automation, has developed an explicit model that clearly depicts the constraints for the dynamic constellations of the CPS networks. They have tested the applicability of this "Orthogonal Dynamicity Constrained Model" using the example of the autonomous transport robots from InSystems. Cooperation with other industrial partners from the automation industry ensured that this use case is representative of robots and smart factories from other manufacturers.
The model was deliberately designed to be easily related to other development artifacts. It was particularly useful to put the model in relation to the goal models of the individual CPS as well as the higher goal model of the dynamic CPS network. This approach is particularly useful in early stages of development, where the potential scope for solutions is still large and needs to be limited by a goal conflict analysis. The explicit modeling of constraints provides a common understanding base, which is not only important in situations where different organisations develop individual CPS for dynamic networks.
The work was carried out as part of the research project CrESt, funded by the Federal Ministry of Education and Research (BMBF), and was published in the Journal of Systems Architecture:
Jennifer Brings, Marian Daun, Torsten Bandyszak, Vanessa Stricker, Thorsten Weyer, Elham Mirzaei, Martin Neumann and Jan Stefan Zernickel: Model-based documentation of dynamicity constraints for collaborative cyber-physical system architectures: Findings from an industrial case study.J. Syst. Archit., Volume 97, 2019, 153-167. [DOI]
Information about CrESt (Collaborative Embedded Systems) can be found here:
Constraints between goals and the CPS network morphologies