As a leading additive manufacturing technology (AM), fused deposition modeling (FDM) exhibits exceptional versatility in processing a wide range of thermoplastic polymers for industrial applications. Nonetheless, achieving optimal part quality is inherently governed by a multitude of process variables, whose intricate relationships dictate the final structural and functional performance. The consequence is that measuring 3D printed components has metrological limitations that are inherent to the layer-by-layer material extrusion process and strongly influenced by thermal shrinkage gradients and the anisotropic behavior of the deposited material during the printing process. Thus, this study developed a Taguchi – fuzzy logic model (T – FLM) as a hybrid model that aims to enhance the surface quality of the printed parts, thereby ensuring uniform surface characteristics across complex components, especially including rotational, horizontal, vertical, and inclined surfaces. The specimen geometry was specially designed to feature these distinct characteristics. So that, it was required to design a geometry of the tested specimen then manufactured it by fused deposition modeling. Orthogonal array (OA) L27 experiments were conducted using PLA material. Experimental design parameters included layer thickness, nozzle temperature, print speed, bed temperature, and top/bottom layer number. The developed model utilized the experimental design to determine the responses of surface roughness (Ra) and deviation in dimensional accuracy (∆L), which are subsequently formulated into the multi–response performance index (MRPI) for optimization. The results confirmed the effectiveness of the hybrid model in optimizing the targeted response, wherein the enhanced uniformity in the printed parts’ surface quality was statistically quantified through central tendency and variability indices. This was evidenced during by the statistical analysis of the fuzzy model, which consistently yielded elevated MRPI values across all 27 experimental runs, achieving a notable improvement of 15.2%. Furthermore, a distinct improvement in the standard deviation of the overall surface quality was successfully demonstrated in approximately 78% of the runs. Additionally, an analysis of variance (ANOVA) test was performed to identify the statistically significant process parameters, and the validity of the optimal combination was successfully verified through a confirmation test.