Prediction and Optimization of PETG Part Hardness in 3D Printing: A Comparative Study of Experimental Design Methods

Abstract

This study compared Box-Behnken Design (BBD), Definitive Screening Design (DSD), and Taguchi design methods to predict and optimize surface hardness in polyethylene terephthalate glycol (PETG) parts fabricated by fused deposition modeling (FDM). Critical printing factors including layer thickness (LT), infill density (ID), nozzle temperature (NT), and printing speed (PS) were analyzed to develop accurate prediction models while minimizing experimental runs. The comparative analysis revealed that all three methods provided reliable hardness predictions, with BBD showing superior accuracy (3.74% error), followed by DSD (4.25%) and Taguchi (4.84%). ID emerged as the most influential factor on surface hardness across all methods. BBD required 27 experimental runs, while DSD and Taguchi needed only 13 and 9 runs, respectively, demonstrating significant efficiency in experimental design. The optimal parameter combinations were validated through confirmation tests, achieving maximum hardness values of 62.47 Shore D (Taguchi), 57.16 Shore D (BBD), and 59.95 Shore D (DSD). These findings provide practical guidelines for industrial applications, enabling manufacturers to select the most suitable design method based on their specific requirements for accuracy versus experimental efficiency in PETG part production. © 2025 Elsevier B.V., All rights reserved.