Considering the problems of oversize drawing force in pipe forming and the excessive stresses in pipe surface, a concept of multi-objective optimization was presented based on “FEM-ANN-MOGA” method. Orthogonal design, finite-element simulation and BP neural network were combined together to build the nonlinear mapping relations between drawing stress, residual stress and forming parameters. In the process of optimization, the multi-objective genetic algorithm based on vector evaluation technique and niche technique was adopted to obtain the evenly distributed Pareto-optimal solutions. By defining a satisfactory degree function, the satisfactory solution was selected, and the FEM simulation verification was carried out. The simulation results accorded with the optimized ones perfectly, which showed that the method is feasible and credible.