Integration of Digital Twin, Machine-Learning and Industry 4.0 Tools for Anomaly Detection: An Application to a Food Plant

Giovanni Paolo Tancredi, Giuseppe Vignali, Eleonora Bottani

This work describes a structured solution that integrates digital twin models, machine-learning algorithms, and Industry 4.0 technologies (Internet of Things in particular) with the ultimate aim of detecting the presence of anomalies in the functioning of industrial systems. The proposed solution has been designed to be suitable for implementation in industrial plants not directly designed for Industry 4.0 applications. More precisely, this manuscript delineates an approach for implementing three machine-learning algorithms into a digital twin environment and then applying them to a real plant. This paper is based on two previous studies in which the digital twin environment was first developed for the industrial plant under investigation, and then used for monitoring selected plant parameters. Findings from the previous studies are exploited in this work and advanced by implementing and testing the machine-learning algorithms. The results show that two out of the three machine-learning algorithms are effective enough in predicting anomalies, thus suggesting their implementation for enhancing the safety of employees working at industrial plants.