Multi‐Objective Optimization for Strength, Porosity, and Surface Roughness of Extrusion‐Based 3D‐Printed ZrO₂ Ceramics Using Response Surface Methodology and Nondominated Sorting Genetic Algorithm‐II

Flexural strength, porosity, and surface roughness are three critical properties in extrusion‐based 3D‐printed ZrO 2 ceramic components. However, these properties are usually contradictory in extrusion‐based direct ink writing (DIW) process. In this work, taking nozzle diameter, filling rate, and layer height/nozzle diameter as process parameter variables, response surface methodology, and nondominated sorting genetic algorithm‐II (NSGA‐II) are used to perform multi‐objective optimization for the above‐mentioned properties of DIW‐printed ZrO 2 ceramics. The predictive models of flexural strength ( σ b ), porosity ( P ), and surface roughness (Ra’) are constructed. The applicability and reliability of the predictive models were verified through analysis of variance, and the interactive effects between the two variables are analyzed. The results show that the optimal parameter combination optimized by NSGA‐II was a nozzle diameter of d = 0.51 mm, a fill rate of r f = 73%, and a nozzle diameter/printing layer height ratio of d / h = 0.4. Experimental validation reveals that the optimal parameters effectively increased flexural strength and porosity, while reduced surface roughness. The actual values obtained are flexural strength σ b = 7.54 MPa, porosity P = 64.17%, and surface roughness Ra' = 26.16 μm, with prediction errors of 5.38, 1.03, and 3.19%, respectively, confirming the accuracy of the constructed predictive model and effectiveness of the optimal parameters.