Modeling a New Approach to Quality 4.0 Using Deep Learning

Quality 4.0 refers to the application of advanced artificial intelligence technologies in the domain of quality. It brings significant advantages in terms of efficiency, flexibility, prediction failure, customer satisfaction, and data-driven decision-making based on real-time data collected from sensors installed on production lines. This article focuses on the development of a new quality 4.0 model that aims to propose an intelligent quality control system by adapting deep learning methods. First, we started with an introduction to deep learning, specifically neural networks, and their modeling steps. Next, we explain the evolution of the concept of quality from its initial appearance to quality 4.0, as well as the different techniques used for quality management. Then, we construct the proposed quality 4.0 model based on neural network methodology to create an intelligent quality control system that can predict defects and failures on a production line before they occur. In this step, we study a real case of a cardboard compactor machine and we compare the results obtained from different neural network optimization methods such as Stochastic Gradient Descent, Gradient Descent with Momentum, RMSprop, and ADAM. Finally, we demonstrate the effectiveness, performance, and accuracy of the results through the testing phase of the neural networks.