A comprehensive review of multi-objective optimization of friction stir welding processes

FSW is a solid-state joining process for similar/dissimilar materials which employs a rapidly rotating non-consumable tool, eliminating solidification problem of the conventional fusion welding processes. However, requirement of skilled operator, poor surface quality, wear and tear of the welding equipment, limited joint designs etc. are some of its limitations. Its superior performance can be achieved through complex process optimization involving multiple welding parameters and weld characteristics and nonlinear relations between them. Several optimization techniques have been proposed in the literature to resolve the issue. In this paper, 151 research articles available in Scopus and other scholarly databases during 2014–2023 are content-wise analysed. Based on application of different multi-objective optimization tools, they are divided into four categories, i.e. desirability function approach, metaheuristics, multi-criteria decision-making methods and grey relational analysis (GRA). Their contents are reviewed with respect to the experimental design plan implemented, works material(s) welded, pin material employed, welding parameters and weld characteristics considered. It is observed that the past researchers have mostly employed Taguchi’s L₉ orthogonal array as the design plan, various aluminium alloys (mainly AA 6061) as the weld materials and H13 tool steel as the pin material. On the other hand, tool rotational speed and welding speed are the maximally preferred welding parameters, while ultimate tensile strength, percentage elongation and hardness are observed as the main weld characteristics. Among the multi-objective optimization tools, GRA is noticed to achieve maximum popularity among the research community. The extracted information would be quite useful as the training dataset for developing machine learning-based prediction models for future use.