Resource-efficient machining through physics-informed machine learning

Resource-efficient machining has been at the forefront of sustainable manufacturing for many years. However, the total life of the cutting tool is often not fully exploited to avoid the severe consequences of excessive wear. This conservative strategy increases production time, cost per part, and the carbon footprint of the machining process. The present contribution focuses on resource-efficient machining through physics-informed machine learning, exploiting the benefits of both physics-based and data-driven modelling to accurately predict cutting tool life. The results demonstrate the utility and superiority of this hybrid approach relative to traditional methods and the recent trends towards purely data-driven techniques.