Feature Reduction and Selection for Use in Machine Learning for Manufacturing

In a complex manufacturing system such as the multistage system, maintaining quality of products becomes challenging task. It is due to interconnectivity and dependency factors that can affect final product. With increasing availability data, Machine Learning (ML) approaches are applied assess predict quality-related issues. this paper, several ML algorithms, including feature reduction/selection methods, were publicly available dataset characteristic output measurements in (mm). A total 24 prediction models produced. The accuracy execution time evaluation metrics. results show uncontrolled variables most common features have been selected by selection/reduction methods suggesting their strong relationship performance was heavily dependent on algorithm.