Exploring Materials Extrusion in 3D Printing: Real-time Monitoring, and Machine Learning for Predicting Mechanical Properties

Materials extrusion (MEX) is a prominent additive manufacturing technique due to its capability to create complicated shaped components, availability of diverse materials, user-friendly operation and affordability. Nevertheless, MEX has certain limitations, including the inherent uncertainty in the fabrication process, stemming from variations in material consistency, heat source, bed adhesion and cooling conditions. In order to enhance process consistency, there arises a compulsive need for the implementation of in-process monitoring and control systems. This paper provides a brief overview of monitoring applications within the MEX method and explores the alternative sensors that can be effectively employed for real-time process monitoring. Additionally, the paper covers the utilization of datasets derived from the MEX process to develop a machine learning-based model, aimed at predicting the mechanical properties of biological scaffolds. The Gradient Boosting Regressor algorithm was employed for the prediction of compressive strength in biological scaffolds, and the proposed model achieved a remarkable accuracy rate of 99%.