Integrating Response Surface Methodology and Machine Learning for Multi‐Objective Optimization of Fused Deposition Modeling Parameters in <scp>3D</scp> ‐Printed Polylactic Acid–Bronze Composites

ABSTRACT In this study, a polylactic acid–bronze (PLA–Br) composite was fabricated using the fused deposition modeling (FDM) technique. The effects of key FDM process parameters on tensile properties were investigated through Response Surface Methodology (RSM) based on a Central Composite Design (CCD). Printing speed (PS), extruder temperature (ET), and raster angle (RA) were selected as input variables, while maximum failure load (MFL), elongation at break (EB), and Young's modulus (YM) were considered as response variables. Artificial Neural Network (ANN) and RSM approaches were employed to develop predictive models. The experimental results and analysis of variance (ANOVA) revealed that PS, ET, and RA significantly influence the tensile properties of the PLA–Br composite. Specifically, a decrease in PS combined with increases in ET and RA led to improvements in maximum failure load and Young's modulus, while reducing elongation at break. A comprehensive statistical evaluation demonstrated that the ANN model outperformed the RSM model in predictive capability across all tensile responses. The ANN model achieved higher coefficients of determination ( R 2 , adjusted R 2 , and predicted R 2 ) along with lower error metrics, including mean absolute percentage error (MAPE) and root mean squared error (RMSE), indicating superior accuracy and generalization performance. Finally, the FDM process parameters were optimized using the ANN model in conjunction with Pareto‐based multi‐objective optimization and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The optimal conditions were identified as a printing speed of 65.66 mm/s, an extruder temperature of 222.69°C, and a raster angle of 87.98°. Under these conditions, the predicted maximum failure load, elongation at break, and Young's modulus were 225.88 N, 1.81 mm, and 10.71 MPa, respectively.