The aim of the Industry 4.0 initiative is to secure Germany’s future as an industrial location and to strengthen its competitiveness compared to other countries. In contrast to large companies, it is more difficult for medium-sized ones to implement the migration from Industry 3.0 to 4.0, as they do not have the financial and human resources to fully replace all systems currently in operation. Therefore, the migration needs to be executed evolutionarily by retrofitting existing production facilities so that new acquisitions can be avoided. In this paper, such a retrofit will be analyzed based on a machine for forming copper rings, which is part of a process for manufacturing valve seat inserts for combustion engines. Since production is carried out under a high cycle time and the rings meet tight tolerances, condition monitoring is to be implemented to detect failures at an early stage. For this purpose, the approaches Design of Experiments (DoE) as well as Machine Learning (ML) are considered. Both options are evaluated based on a complexity analysis using the environment concept of an intelligent agent by RUSSEL & NORVIG. Finally, suitable supervised ML algorithms are selected.
CITATION STYLE
Thelen, F., Theren, B., Husmann, S., Meining, J., & Kuhlenkötter, B. (2023). Method for a Complexity Analysis of a Copper Ring Forming Process for the Use of Machine Learning. In Lecture Notes in Production Engineering (Vol. Part F1163, pp. 600–610). Springer Nature. https://doi.org/10.1007/978-3-031-18318-8_60
Mendeley helps you to discover research relevant for your work.