Insights into the metal cutting contact zone through automation and multivariate regression modelling under the framework of gear skiving

The modern time of Industry 4.0 requires an enhanced prediction process for reliable and sustainable manufacturing. It is essential to understand the relationships between various process parameters of machining for better optimization. Digitalization offers the opportunity to accelerate the prediction process using different modelling such as numerical and data-driven models. Improvements in the knowledge of thermo-mechanical variables and the use of finite element method (FEM) tools and machine learning approaches for thorough thermo-mechanical analysis are noteworthy contributions to the area. However, an ideal standardized approach remains to be resolved. Therefore, this research proposes a development process of an automated FEM tool to simulate the tool-chip interaction for AISI4140 material, coupled with a hybrid multivariate regression model for fast prediction of non-linear relationships between the cutting parameters and the contact properties. Consequently, the study also interprets the tool-chip interactions in the secondary deformation zone, facilitating process optimization for improved machining performance.