The machining demonstrator showcases the use of AI for the visual analysis of chips during the turning process. The analysis results provide specific insights into how cutting parameters, such as rotational speed or feed rate, can be optimised. This improves the quality of the process and reduces tool wear. The demonstrator also offers the potential to integrate machine data into the analysis, thereby enabling real-time analysis during the process.

The approach to brainstorming and developing the demonstrator was based on a practical problem relating to process monitoring in machining. Implementation was carried out using the CRISP-DM approach. As part of this, relevant chip shapes were identified, produced and photographed in order to train an AI model for visual chip analysis. The analysis results are incorporated into a user interface that provides targeted recommendations for optimising the cutting parameters..

The demonstrator aims to improve the quality and efficiency of turning processes through AI-supported visual machining analysis. It offers the added benefit of enabling the targeted optimisation of cutting parameters such as rotational speed and feed rate, in order to reduce tool wear and stabilise production processes. Furthermore, the demonstrator promotes an understanding of data-driven process optimisation and machine data analysis, for example in workplace training.

Chair of Production Systems, Ruhr University Bochum