The work aims to study the thickness distribution of paint film so as to achieve automatic spray painting on huge ship outer panel, optimize surface spraying trajectory and find out the best lap width of paint film. The orthogonal experiment was designed with 3 factors, including width of spray nozzle, distance and speed of spray gun to obtain the thickness distribution data of paint film. β distribution function was used to characterize experimental data, and Genetic Algorithm (GA) was applied to optimize 3 key parameters of the β distribution model. The thickness distribution prediction model of paint film was established based on BP (Back Propagation) neural network, and the prediction results were analyzed reasonably. The overlap rules of each paint film were studied and NSGA-II algorithm was used to solve the best lap width of paint film. The corresponding Tmax, w and β of the β distribution model were obtained for the 15 groups of painting experimental data. The prediction model could accurately predict the thickness distribution of paint film under different spraying parameters. For example, in the 16th group of painting experiment, the best lap width of paint film was 41.76 cm. The mapping model between the spraying process parameters and the paint film thickness distribution is obtained. The mapping model can accurately predict the coating thickness distribution under different process parameters. The best lap width of paint film calculated based on this model can obtain the coating with uniform thickness, thus making good preparations for the automatic spraying of ship outer panel.