To address the issue in the practical application of the designed intelligent spraying robot for two-component road traffic markings, where the relationship between marking width/thickness and spraying process parameters is unclear, making it difficult to precisely control the quality of the applied marking dimensions. First, by establishing an experimental platform and combining with the mechanistic analysis of the spraying process, this article adopts an experimental design that integrates single-factor and orthogonal experiments to investigate the influence patterns and interactions of various spraying process parameters on marking width and thickness. Then, the least squares method is employed to fit the experimental data, obtaining the optimal fitting forms for the relationship between each spraying process parameter and the marking width/thickness. Subsequently, multivariate nonlinear regression models are established to predict the width and thickness of the two-component road traffic markings, respectively. Furthermore, machine learning algorithms, including the back propagation neural network (BPNN), Gaussian process regression (GPR), and generalized regression neural network (GRNN), are introduced to compare and evaluate the prediction performance. Finally, prototype testing is conducted to validate the accuracy of the predictive models. The experimental results indicate that the coefficients of determination of the selected spraying process parameters on the road traffic marking width and thickness are 99.4% and 99.3%, respectively. The established multivariate nonlinear predictive models effectively overcome the overfitting defects of machine learning models under small-sample engineering conditions, with their predictive fitting degrees (R2) reaching 0.932 3 and 0.978 4, respectively, which meets the construction requirements for two-component road traffic markings. The findings provide a theoretical basis for the intelligent spraying robot to achieve precise control over marking width and thickness by regulating spraying process parameters, offering scientific support for on-site technical decision-making and contributing to the enhancement of the spraying process automation.