JCO forming is widely used in production of large diameter submerged-arc welding pipes. However,more total number of bending deformation and low production efficiency of JCO forming have been the bottleneck problem that restricts the further development. To resolve this problem,this study investigates JCO forming process by multi-objective optimization. JCO forming is simulated using the FE code ABAQUS and the FE model is validated experimentally. A response surface method based on the radial basis function is used to construct a surrogate model; the genetic algorithm NSGA-II is adopted to search for Pareto solutions; and grey relational analysis is used to determine the most satisfactory solution from the Pareto solutions. The obtained optimal design of parameters shows good agreement with the initial design and remarkably improves the JCO forming quality.Thus,the results of this study provide an effective approach for improving production efficiency and reducing total number of bending deformation in JCO forming.