The task of obtaining suitable welding parameters for the friction stir welding process is often a difficult one, due to the lack of published data and the fact that the exact mechanism by which the process operates has not yet been fully determined. Therefore, suitable welding parameters often need to be obtained by using extensive, time consuming and expensive experimental methods. The work detailed in this paper pertains to the use of the Taguchi method as a mean to reduce the disadvantages of these experimental methods, more specifically, their cost. The Taguchi method accomplishes this task by substantially reducing the number of welding trials that are needed to obtain suitable welding parameters. This reduction leads to the parameters being obtained more rapidly and at a substantially smaller cost. In this paper a procedure for applying the Taguchi method to the friction stir welding process is presented as well as its application to the welding of a specific component. The method was applied to the welding of 4mm thick AA5083-H111 plates in a butt joint configuration, which constitutes one of the most common industrial welding scenarios. The purpose of the experimental tests was to maximize the welding speed whilst ensuring an acceptable welding quality. The quality of the welds was determined through visual inspection and tensile and bending tests. The application of the Taguchi method allowed, with a relatively small number of experimental welds, to provide some insight into the manner by which the parameters should be altered in order to optimize the process.