<p>In the manufacturing industry, laser welding is an important production process, including cutting and welding designs. Obtaining optimal designs is critical to improving the manufacturer’s efficiency and competitiveness, especially in their product performance. A multi-objective nonlinear programming model is formulated for the laser welding design problem to minimize the length of the welds and the material consumption of the laser-welded blanks. An encoding mechanism is proposed to represent the decision variables. A non-dominated sorting genetic algorithm based on local search and incremental support vector regression (NSGA-II-ISVR), as a multi-objective evolutionary algorithm, is developed. Candidate sets are periodically selected for local search based on decomposition in order to locate promising areas of the feasible region for further exploitation. The trained SVRs are used as surrogate models to find approximate fitness values of the trial solutions in the offspring population and a selection rule is used to select the well performing solutions in the offspring population into the candidate set so as to reduce the function evaluations of the actual criterion vectors. The experimental results using four real-world instances of laser-welded blanks demonstrate that the developed NSGA-II-ISVR is very effective in finding good solutions as compared to four state-of-the-art baseline procedures.</p>