In this paper, a new algorithm for the selection of process parameters for obtaining the optimal back-bead width in the Gas metal Arc (GMA) welding of mild steel is presented. To consider these quality characteristics for the selection of process parameters, the Taguchi method is adopted to analyze the effect of each process parameter on the back-bead width, and then to determine the process parameters with the optimal back-bead width. The algorithms are obtained by multiple regression and neural network methods. Finally, Confirmation tests are performed to make a comparison between the results calculated from the developed models and the experimental results.