The aim of the present work is to optimise the welding parameters for friction stir spot welded non-heat-treatable AA3003-H12 aluminium alloy sheets using a Taguchi orthogonal array. The welding parameters, such as the tool rotational speed, tool plunge depth and dwell time, were determined according to the Taguchi orthogonal table L9 using a randomised approach. The optimum welding parameters for the peak tensile shear load of the joints were predicted, and the individual importance of each parameter on the tensile shear load of the friction stir spot weld was evaluated by examining the signal-to-noise ratio and analysis of variance (ANOVA) results. The optimum levels of the plunge depth, dwell time and tool rotational speed were found to be 4.8 mm, 2 s and 1500 rpm, respectively. The ANOVA results indicated that the tool plunge depth has the higher statistical effect with 69.26% on the tensile shear load, followed by the dwell time and rotational speed. The tensile shear load of the friction stir spot welding (FSSW) joints increased with increasing plunge depth. Additionally, examination of the weld cross-sections, microhardness tests and fracture characterisation of the selected friction spot welded joints were conducted to understand the better performance of the joints. All the fractures of the joints during tensile testing occurred at stir zone (SZ), where the bonded section was minimum. The tensile shear load and tensile deformation of the FSSW joints increased linearly with increasing the bonded size. The finer grain size in the SZ led to the higher hardness, which resulted in higher fracture strength. When the tensile shear load of the joints increased approximately 3-fold, the failure energy absorption of the joints increased approximately 15-fold. (C) 2014 Elsevier Ltd. All rights reserved.