Editing Status: 🚧 (under construction)
Last Modification: 09.05.2021
2️⃣ Applicability analysis of association mechanisms in automation based on presented adaptability use cases
Adaptability refers to structural changes of the production system. Structural changes are preceded by changings of the system in terms of design and automation. From a design point of view, a modular production system can be separated into four different physical levels:
The use cases in Figure 1 and Figure 2 to adapt the PEA by changing a COMP or changing a FEA, should be preferably solved by decentral orchestration. The COMP respectively the FEA are orchestrated by the PEA. For plant automation (see Figure 4), a central orchestration is the best solution due to the amount of services and the complexity. Both types of orchestration have a centrally located orchestrating node that is responsible for execution and interconnection of the automation functions with respect to the three logic levels. The use case in Figure 3 pointed out, that with regard to stricter latency requirements a choreography-based mechanism should preferably be used. An orchestration always has higher latency times due to the broker-like communication and interaction structure. This leads to disadvantages regarding regulatory and interlocking logic associations. A mechanism following the principles of choreographies results in a more interactive association to define the collaboration between each of the involved automation services. This leads to smaller latency times regarding regulatory and interlocking logic and enables a wider range of applications. As published in [8], choreography is a rather unknown and unexplored approach compared to orchestration in the context of industrial automation systems. For this reason, this paper presents a first concept in which the principles of a choreography is investigated and implemented according to determined requirements for industrial automation.