In search of the best multi-criteria decision making-particle swarm optimization-based hybrid approach for parametric optimization of friction stir welding processes

Although primarily developed for aluminium alloys, friction stir welding (FSW) has nowadays emerged out as a ‘green’ effective joining process for other light weight metallic alloys owing to its solid-state nature. It has been experimented that quality of the weld mainly depends on selection of the optimal combination of various welding parameters, like tool rotational speed, welding speed, axial load, tool shoulder geometry, tool tilt angle, pin geometry etc. In this paper, seven multi-criteria decision making (MCDM) techniques, i.e. weighted aggregated sum product assessment, technique for order of preference by similarity to ideal solution, grey relational analysis, VIekriterijumsko KOmpromisno Rangiranje, multi-objective optimization on the basis of ratio analysis, complex proportional assessment and preference ranking organization method for enrichment evaluation are separately hybridized with particle swarm optimization (PSO) algorithm to identify the best parametric combinations of two FSW processes. The corresponding polynomial regression (PR) models are developed to be the inputs to these hybrid optimizers. They are later compared with the traditional weighted sum multi-objective optimization (PR-WSMO-PSO) approach, showing their superior performance. Among those MCDM techniques, preference ranking organization method for enrichment evaluation hybridized with PSO evolves out as the best method with respect to improvement in the corresponding performance metric. It is observed that during FSW of double phase α/β brass plates, an optimal combination of tool rotational speed = 1097 rpm, traverse speed = 90.8 mm/min and axial force = 2.5 kN, and during FSW of AZ31-AZ91 magnesium alloys, an optimal intermix of tool rotational speed = 986 rpm, welding speed = 50 mm/min and tool shoulder diameter = 21 mm would lead to simultaneous attainment of the most desired weld characteristics. Moreover, there is approximately 2.02–14.19% saving in computational time for all the MCDM-PR-PSO approaches as compared to PR-WSMO-PSO approach.