AbstractThe discussion paper “I4.0 language: vocabulary, message structure and semantic interaction protocols of the I4.0 language”, published by the working group “Semantics and Interaction of Industry 4.0 Components” of the GMA, also known as UAG of the AG 1 of the platform Industry 4.0 (I4.0), presented a concept for the language between I4.0 components. The main conclusion is: The increasing networking and cooperation of components enable new forms of organization and control. A clear understanding of machine interactions paves self-organized and self-optimized value creation in I4.0. Agent-based systems are an option for the realization of such I4.0 architectures. Due to their features, software agents are particularly well suited for representing I4.0 components and enabling I4.0 interactions. Agents are not only able to understand the necessary machine languages, but also the essential mechanisms for self-organization and self-optimization in value creation. The paper focuses on I4.0 scenarios described by the Platform I4.0 that describes challenges for the industry towards its digital future and demonstrates how emerging challenges in the area of I4.0 can be met with the help of agent-based systems.

The Cyber-Physical Production System (CPPS) is a concept derived from software (cyber) and hardware (physical) applications and is based on global information exchange between such systems. The CPPS is known as a trend of Industry 4.0 (I4.0) focusing on flexibility regarding new products and adaptability to new requirements. This paper focuses on two I4.0 scenarios described by the Platform Industrie 4.0 that describe challenges for the industry towards its digital future. First, it looks at the Order Controlled Production (OCP) scenario that deals with flexible and self-configuring production networks. It describes the dynamic organization of production resources required to execute a production order. Second, the Adaptable Factory (AF) application scenario is discussed, which focuses on the configuration of production resources and describes the adaptability of an individual facility through (physical) modification. This paper first provides a detailed analysis of the requirements from these scenarios. Furthermore, it analyses the current Multi-Agent System (MAS) architectures and agent-based planning and decision support systems requirements. MAS can be used to create application-independent I4.0 systems with arbitrary hardware automation platforms. To create a scalable communication network that also supports application independence and enables the semantically machine-readable description of the exchanged data, the OPC UA standard was adopted. As a result of the study, the concept shows how different and independent automation platforms can be seamlessly connected via OPC UA. The proposed MAS concept has been evaluated in different use cases, namely OCP and AF.

Small-scale manufacturing often relies on flexible production systems that can cope with frequent changes of products and equipment. Transports are a significant part of the production flow, especially in the domain of large and heavy workpieces that requires explicit planning to avoid unnecessary delays. This contribution takes a detailed look at how to create feasible integrated schedules within a decentralised or even heterarchical architecture and which information the agents have to exchange. These schedules incorporate constraints such as the blocking-constraint. They also consider dynamic setup and operation durations while finding a good-enough solution. The proposed agent-based solution applies to a wide variety of scheduling problems and reveals positive properties in terms of scalability and reconfigurability.

Agentenbasierte Steuerung ist ein möglicher Lösungsansatz zum Umgang mit steigenden Ansprüchen an Flexibilität und Robustheit sowie der inhärenten Komplexität von Produktionssystemen im Kontext von Industrie 4.0. Agentensysteme nutzen in vielen Fällen Auktionen, um Aufgaben und Ressourcen zuzuordnen. Die dafür notwendigen Gebote werden von den Agenten unabhängig voneinander berechnet. Im Beitrag wird ein Ansatz zur Optimierung vorgestellt, der einen Mehrwert für ein Unternehmen generiert, indem dieses mehrere Agenten innerhalb seines Einflussbereichs vereint und diese koordiniert auf einem unternehmensübergreifenden Marktplatz agieren lässt. Zur Ausgestaltung der Optimierung wird ein lernender Algorithmus genutzt, der es ermöglicht, aus einer Auftragshistorie auf die Wahrscheinlichkeit für den Erhalt künftiger Aufträge zu schließen. Anhand einer Simulation werden die sich daraus ergebenden Wettbewerbsvorteile herausgearbeitet.