AI Techniques for Anomaly Detection in Polymer-Based 3D Printing: A Review

The advent of 3D printing technology has transformed the manufacturing landscape, particularly in the field of polymer processing. However, maintaining consistent quality and detecting anomalies during the printing process remain key challenges. This review paper explores the application of Artificial Intelligence (AI) for anomaly detection in polymer-based 3D printing. It discusses a range of AI methodologies, including traditional machine learning approaches such as Support Vector Machines and Decision Trees, as well as advanced deep learning techniques like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). The paper also examines data collection and preprocessing strategies essential for training robust AI models. Through selected case studies, we highlight implementations where AI has significantly improved defect detection, print quality, and process efficiency. Challenges such as data quality, model interpretability, and real-time deployment constraints are addressed. The review concludes by summarizing the key findings and recent advancements in AI-driven anomaly detection, offering a consolidated perspective on the current state of the field.