Intelligent Quality Management For Production Enhancement: A Review Of Ensemble Machine Learning Techniques

Abstract

Nowadays, Artificial intelligence technology led to a rise in manufacturing innovation, including improvements in quality management. The expansive realm of artificial intelligence (AI) encompasses various branches, among which machine learning (ML) has evolved into a distinct science, notably, the field of ensemble learning (EL) that has gained heightened interest.This paper attempts to explore the novel concept of ensemble learning and its application in quality management with a narrow focus on quality control and quality assurance. In fact, we examine the performance of the three most popular tree-based algorithms (Random forest, Extream gradient boosting, and Adaptive boosting). Through an evaluation process, we select the most used models based on previous works and researches in order to reveal their underlying qualities. This research reveals that Random forest is the most used algorithm that can outperform not only the basic machine learning algorithms but also the deep learners due to its properties especially its simplicity, capacity and ability to handle multidimensional data.