Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : September 11, 2022 - September 14, 2022
Machine learning offers powerful algorithms for various application domains and objectives. Therefore, this paper presents a systematic overview of applications in injection molding science and industry. Scopus and Web of Science were used to search for articles in which authors indicated they had used machine learning. 192 papers spanning nearly three decades were reviewed and analyzed, with 68 papers selected for this review. All papers were analyzed in terms of their objective, machine learning algorithms used, learning type, and data source. The most common objective was quality or defect prediction, with neural networks being used in most cases. Supervised learning was applied to the most common data source, machine and process data. Although there are already a large number of publications on this topic and this area has grown rapidly in recent years, there are only a few industrial applications in injection molding so far.
