<p>Metal matrix composites find wide applications in various construction fields. However, the main challenge lies in their fabrication: the random distribution of fibers leads to their overlap and the formation of voids. This study evaluates the occurrence of these defects, particularly voids, as well as the effect of fiber ratio and size on the mechanical properties of aluminum/graphene composites. To this end, a script based on the sequential random adsorption algorithm was developed. We used a numerical homogenization approach based on representative volume models, combined with finite element analysis. The results showed that the presence of voids significantly reduces the mechanical performance of the composite by promoting stress concentration and decreasing the overall strength. In parallel, the addition of graphene nanoparticles significantly improves the stiffness and shear modulus, which is attributed to the exceptional intrinsic properties of graphene. These results were verified using both the Mori-Tanaka model and the rule of mixtures. Furthermore, a hybrid model, combining numerical simulation and analytical regression, showed a strong correlation with the numerical results, thus verifying its reliability for predicting composite behavior.</p>