Using Machine Learning Algorithms for the Prediction of Industrial Process Parameters Based on Product Design

Abstract

In the present paper, a method of defining the industrial process parameters for a new product using machine learning algorithms will be presented. The study will describe how to go from a final product characteristics till the prediction of the suitable machine parameters to produce a good quality of this product, and this is based on an historical training data set of similar products with their respective process parameters. In the first part of our study, we will focus on the ultrasonic welding process definition, welding parameters and on how it operates. While in second part, we will present the design and the implementation of the prediction models such as multiple linear regression, support vector regression, and we will compare them to the artificial neural networks algorithm. In the following part, we will present a new application of Convolutionnal Neural Networks (CNN) algorithm to the industrial process parameters prediction. In addition, we will propose the generalization approach of our CNN to any prediction problem of the industrial process parameters. Finally, we will deploy our models into a physical device with an interactive graphical user interface. This prediction device allows the user to move freely on the manufacturing field and perform process development operations.