This work provides a method to help optimize the underwater friction stir welding (UFSW) process of pipes based on the fuzzy multi-criteria decision making (FMCDM) methodology. In the optimization phase ambiguity and vagueness are dealt with using linguistic variables parameterized by triangular fuzzy numbers. In the UFSW an important consideration is the cooling effect during the process on the surrounding water. A study is being conducted to conclude and optimize the effects of the UFSWoP process parameters on the welded joint's mechanical properties. Experiments were performed at three levels of three parameters: tool shoulder diameter tool, tool rotational speed and travel speed. For optimization, the MCDM methods were applied as an approach for selecting the optimal values of the parameters. Using MCDM in optimizing the UFSW process is rare and has a limited number of publications. The MCDM techniques used are the Evaluation Based on Distance from Average Solution (EDAs), VlseKriterijumska Optimizcija I Kaompromisno Resenje (VIKOR), hybrid Fuzzy – EDAS and hybrid Fuzzy -VIKOR to find the optimum process parameters which maximize the values of responses: the UTS and the VHN of the welded joint. Results showed that hybrid Fuzzy-EDAS and Fuzzy-VIKOR theory can solve the limitations of using EDAs and VIKOR when uncertainty problems exist in the data. For the used parameter ranges the optimum values were 20 mm, 1800 RPM and 4 mm / min respectively for shoulder diameter, rotational speed and travel speed.