The main objectives of this study were to aerodynamically design and optimize a winglet for a wind turbine blade by using computational fluid dynamics (CFD) and to investigate its effect on the power production. For validation and as a baseline rotor, the National Renewable Energy Laboratory Phase VI wind turbine rotor blade is used. The Reynolds-averaged Navier–Stokes equations are solved, and k–ε Launder–Sharma turbulence model was used. The numerical results have shown a considerable agreement with the experimental data. The genetic algorithm was used as the optimization technique with the help of artificial neural network to reduce the computational cost. In the winglet design, the variable parameters are the cant and twist angles of the winglet and the objective function the torque. Multipoint optimization is carried out for three different operating wind speeds, and a total of 24 CFD cases are run in the design. The final optimized winglet showed around 9% increase in the power production. Copyright © 2013 John Wiley & Sons, Ltd.