Semantic Knowledge Representation in Asset Administration Shells: Empowering Manufacturing Utilization

Within the context of Industry 4.0 and the Reference Architecture Model Industrie 4.0, the Asset Administration Shell (AAS) framework has emerged as a critical component for implementing Digital Twins that facilitate a seamless data exchange within a manufacturing ecosystem. The concept enables asset communication across various application domains, organizing the design and interaction amongst manufacturing components while capturing key information relating to assets such as operational parameters, intrinsic properties, and technical functionalities. Each category of information is stored within the AAS in a structure called a sub-model, while specific properties are stored as sub-model elements. Moreover, the literature presents a gap in post-processing the information contained within the sub-models to generate data-driven decisions and the bidirectional data exchange with the AAS. Additionally, the developing concept behind AAS requires demonstration with manufacturing cases of what can be done with the information in the submodels of the AAS. This research explores the practical implementation of AAS in a manufacturing context and the tools used in their development. An approach is developed to map a semantic ontology representation from the AAS structure. The semantic structure of the information enables querying and reasoning about the data, which contributes to better understanding among diverse manufacturing components, enabling enhanced monitoring and decision-making. The use case utilizes parameter data from the AAS to estimate a key performance Indicator, Overall Equipment Effectiveness (OEE), and stores the value back in the AAS. Utilizing this approach contributes to optimizing manufacturing efficiency and productivity within manufacturing production.