Abstract
Recent advances in the computational power, and at the same time, the software that is capable of taking advantage of the new hardware architecture promote numerical modelling activities for the Friction Stir Welding (FSW) process as in parallel with other engineering applications. All these developments provide a stronger basis for understanding of the FSW process by enabling inclusion of more detailed multi-physics phenomena, i.e. complex interaction among material behaviour, microstructure evolution, material flow, heat generation, etc. A source of motivation behind all these efforts is the increasing demand for the FSW process mainly in aerospace and automotive industries. However the list of unknowns for the success of the process (e.g. defect free welds) is not yet cleared up. Tool design and process parameter optimization studies are in general limited by design of experiments and those rarely supported by the statistical analysis tools. One of the main reasons for the lack of the autonomous optimization studies, in which the numerical FSW simulations are used for response evaluations, is still the high computational cost. Here in this review paper, a brief overview of remarkable achievements together with the discussion of the limitations in the numerical FSW optimization studies are laid on the table.