Machine Learning Modeling to Predict Mechanical Properties of Nylon Parts Fabricated via 3D Printing

Abstract The aerospace industry consistently seeks lightweight and high-strength thermoplastics materials such as polyamide (PA) to enhance performance and fuel efficiency. This present study aims to explore the relationship between 3D printing process parameters (Nozzle Temperature (NT), Layer Height (LH), and raster deposition angle (DA)), and the tensile strength of PA specimens. Four machine learning (ML) models, including multilinear regression (MLR), XGBoost regressor, support vector machine (SVM), and random forest regressor (RFR), were used. The results revealed that among the proposed ML models, the RFR algorithm exhibited the highest performance with R2 = 0.82, RMSE = 0.42, and MAE = 0.11 in the test set. Furthermore, to showcase the efficiency of the considered models in prediction of tensile strength of PA specimens, the results of these models are compared by means of statistical metrics. Optimizing the 3D printing parameters using the proposed ML models can offer a pathway to manufacturing 3D printing thermoplastic parts with improved mechanical properties, making them valuable tools for optimizing thermoplastic materials design in the aerospace sector.