Multi-response optimization of 3D printing parameters by Taguchi method and machine learning

Abstract Fused Deposition Modeling (FDM) is widely used in additive manufacturing, although the mechanical performance of printed components is highly dependent on process parameters. This study optimized the printing parameters of ABS PA-747H using the Taguchi method, analysis of variance (ANOVA), and Random Forest Regression (RFR). An L9 orthogonal array evaluated the effects of layer thickness, build orientation, infill density, and printing speed on tensile and flexural properties. The optimum parameter combination (A1B1C1D1: 0.2 mm, 0°, 100% infill, and 10 mm/s) increased tensile strength from 30 MPa to 40 MPa and flexural strength from 69 MPa to 92 MPa, while tensile and flexural moduli reached 1.90 GPa and 2.15 GPa, respectively. The RFR model, validated using 5-fold cross-validation, accurately predicted mechanical properties. Fractographic analysis revealed fewer voids, stronger interlayer bonding, and shorter crack propagation paths, confirming improved structural integrity under optimized printing conditions.