Metal Matrix Composites (MMCs) play a vital role in replacing many structural materials due to their superior mechanical and metallurgical properties. Processing those composites has always been a major factor influencing their application. Al-TiB2 MMCs are used as structural members in the marine industry. Friction Stir Welding (FSW) has revolutionized the process of joining those composites. In this paper, an attempt has been made to friction stir weld the Al-TiB2 MMCs and to develop a regression model for predicting the tensile strength of the weldment. The process parameters considered for FSW are tool rotation speed, tool traverse speed, axial load and weight percentage of TiB2 in Al matrix. The regression model is used to optimize the process parameter using Desirability Optimization Methodology to improve the tensile strength of the FS welded Al-TiB2 composites. The effects of process parameters on the tensile strength of the welded composites are analyzed and presented.