Predicting Three-Dimensional (3D) Printing Product Quality with Machine Learning-Based Regression Methods

This study examines how printing parameters affect the roughness, tensile strength, and elongation of 3D-printed parts used in various applications. Machine learning-based regression models were employed to optimize product quality. The open-source "3D Printer Material Requirement" dataset obtained from the Kaggle platform was utilized to predict product quality. This dataset includes input parameters such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, printing material (PLA and ABS), and fan speed. These parameters were analyzed for their impact on the product's roughness, load resistance, and elongation under tensile force. Based on these evaluations, product quality was estimated according to its intended use. Parameters such as layer height, wall thickness, infill density, infill pattern, nozzle temperature, bed temperature, print speed, printing material, and fan speed were identified as key factors influencing output performance. Within this framework, prediction models including Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Gaussian Process Regression (GPR), and Multi-Layer Perceptron (MLP) were developed, and their performances were assessed using metrics such as accuracy (R²), error rates (RMSE, MSE, MAE), and computational time. Among these methods, GPR demonstrated the highest prediction accuracy for elongation, tensile strength, and roughness, with respective values of 0.98, 0.9, and 1. The findings indicate that machine learning applications are effective tools for quality prediction and optimization in the production processes of 3D printers. Furthermore, this study provides a novel perspective on quality control and design optimization in 3D printing processes.