Engineering a Multi-agent Systems Approach for Realizing Collaborative Asset Administration Shells

In recent years, there has been a high demand for flexible and reconfigurable processes to meet the fast-changing market conditions. In this context, Industry 4.0 is promoting the digitization of traditional production systems towards intelligent factories with highly automated and rapidly adaptable capabilities based on Industrial Cyber-Physical Systems (ICPS). This digitization process is currently being leveraged by the so-called Asset Administration Shell (AAS), a standardized digital representation of an asset that provides the uniform access to the asset information. Additionally, the AASs offer the digital basis for future autonomous systems, where intelligent AASs may perform collaborative functions to enhance industrial processes. However, currently such solutions are still at an early stage of maturity. In this sense, this paper explores the adoption of a Multi-agent Systems (MAS) approach to provide the required intelligent and collaborative aspects for the traditional AAS. The proposed MAS-based AAS solution was applied in an industrial automation case study to analyze the feasibility of MAS to perform intelligent and collaborative functions in the AAS context.