<p>This study aims to enhance the compressive strength of 3D-printed PLA/Carbon fiber composites by optimizing key Fused Deposition Modeling (FDM) process parameters, specifically layer height, infill density, infill pattern, and raster orientation. Using a full factorial design with 81 unique parameter combinations, this research systematically explores how each parameter and their interactions affect the compressive strength of the composite material. Statistical Analysis of Variance (ANOVA) was employed to identify significant parameters, revealing that infill density is the most critical factor, accounting for 74.13% of the observed variation in compressive strength, followed by infill pattern (6.4%) and its interaction with density (4.6%). The raster orientation and layer height showed negligible impacts, contributing only 0.03% and 1.3% of the variation, respectively. The experimental results demonstrated that the compressive strength increased significantly with higher infill densities, achieving maximum values of 82.51&#xa0;MPa with a 100% infill density using a linear infill pattern and optimal raster orientation. Additionally, Artificial Neural Networks (ANN) were employed for predictive modeling of compressive strength. Among the two backpropagation algorithms tested, the Levenberg-Marquardt (LM) algorithm outperformed the Scaled Conjugate Gradient (SCG) algorithm, achieving an R² of 0.97548 and a minimum Mean Squared Error (MSE) of 28.304, compared to the SCG's R² of 0.93862 and MSE of 54.8527. This dual approach—combining statistical analysis for insights on parameter significance with machine learning for predictive accuracy—offers a robust framework for FDM parameter optimization. The findings hold practical implications for industries requiring lightweight, high-strength components, such as aerospace, automotive, and structural engineering, where enhanced mechanical performance of 3D-printed composites is essential.</p>