In this study, for the first time, a fuzzy logic model was used to elucidate and optimize the friction stir welding of pure copper. For microstructural characterization, light microscopy, electron backscattered diffraction (EBSD), and X-ray diffraction (XRD) were employed. The tensile WA was used to measure the ultimate tensile strength (UTS) and elongation of the joints. Furthermore, the tensile fractured surfaces were analyzed using scanning electron microscopy (SEM). The fuzzy prediction revealed that using a rotational speed of 1136 rpm, a traverse speed of 46.75 mm-min(-1), and an axial force of 3.34 K N resulted in maximum UTS of 276.1 Mpa and maximum elongation of 44.6 %. It was found that the amount of deformation was the dominant factor affecting the joint properties. At insufficient deformation, some voids were formed in the stir zones. However, at larger deformations, i.e., higher rotational speed, lower traverse speed, and larger axial force, the defects were eliminated, and the grain sizes were reduced according to the Zener-Holloman factor. By conducing friction stir welding at optimum condition, a synergic increase of strength and ductility was obtained. Moreover, it was found that the mechanism of grain structure formation was continuous dynamic recrystallization (CDRX).