https://www.mdu.se/

Industry 4.0 presents nine technologies, including the Industrial Internet of Things (IIoT), Big Data and Analytics, Cloud Computing, etc. The progress of Industry 4.0 places a new demand for maintenance. Some of the nine technologies of Industry 4.0, such as IIoT, Big Data and Analytics, Cloud Computing and Augmented Reality (AR), as well as Cyber-Physical System (CPS) and machine learning, play an important in the development of smart maintenance technologies. In smart maintenance research, it is presented how IIoT can be used for machine connection and maintenance data collection, machine learning models for maintenance data analysis and failure prediction, AR for maintenance instructions, etc. Although previous smart maintenance research presents many technologies for smart maintenance, the manufacturing companies still face many implementation challenges when implementing and using to add benefits to maintenance organizations in line with companies main goals. Many manufacturing companies still utilize reactive maintenance and are experiencing too much downtime. In this paper, I have shown how two machine learning models, Isolation Forest and Regression Learner, and the statistical technique, Interquartile range (IQR), can be applied in order to detect anomalies in an unsupervised dataset, consisting of travel time for a linear guide, and temperature, in a drill station, which is part of a Cyber-Physical production system, located at a smart production laboratory in Sweden.