Real-time anomaly detection in Industry 4.0 with the help of the Asset Administration Shell.

Abstract

The rapid advancement of Industry 4.0 necessitates robust and interoperable digital twin technologies supported by structured semantic frameworks such as the Asset Administration Shell (AAS). This paper systematically explores aspects of Industry 4.0 implementations, including semantic interoperability via AAS, standardized data formats such as AASX, time-series data management, and middleware solutions. Emphasis is placed on unsupervised anomaly detection techniques—Median Absolute Deviation (MAD) and Mahalanobis Distance—within industrial streaming data environments. Utilizing a case study, sensor data were analyzed through a developed Eclipse BaSyx plugin integrated with InfluxDB and MQTT, demonstrating effective real-time anomaly detection. The findings underscore the importance of adaptable and standardized semantic integration for achieving optimized operational efficiency in Industry 4.0.