<p>This research investigates the optimization of process parameters in Fused Deposition Modeling (FDM) 3D printing using artificial intelligence techniques. FDM printing quality is highly dependent on multiple interdependent parameters including layer height, print speed, extrusion temperature, infill density, and build orientation. Traditional optimization methods based on trial and error or design of experiments are time consuming and often fail to capture the complex nonlinear relationships between parameters. This study employs machine learning algorithms, specifically Random Forest and Artificial Neural Networks (ANN), to predict and optimize mechanical properties and surface quality of printed parts. Experimental validation was conducted using Polylactic Acid (PLA) material with systematic variations in five key parameters. Tensile strength, flexural strength, impact resistance, and surface roughness were measured according to ASTM standards. Results demonstrate that AI based optimization achieved 23.7% improvement in tensile strength, 18.4% enhancement in flexural strength, and 31.2% reduction in surface roughness compared to default printing parameters. The ANN model showed superior prediction accuracy with R² values exceeding 0.94 for all tested properties. This research provides a data driven framework for rapid parameter optimization in FDM printing, significantly reducing development time while improving part quality.</p>