To improve the uniformity of asphalt mixtures, optimizing the particle distribution is essential, and the screw conveyor of paver plays a critical role in this process. In this regard, a parameter optimization strategy based on the discrete element method and response surface methodology (RSM) is proposed. First, a discrete element model is developed for spiral material separation and asphalt mixture particles. Next, uniformity evaluation indicators are used to analyze the impact of individual parameters on particle uniformity. Then, simulation experiments are conducted under varying material level coefficients and rotational speeds, and a regression model is developed to analyze the impact of coupling factors using RSM. Afterward, different algorithms are employed to optimize the parameters of the model, and the results show that the particle swarm optimization algorithm is better than the traditional hill climbing algorithm. Last, field tests are conducted to verify the feasibility of the proposed optimization strategy. The optimal theoretical parameter of the material level coefficient is 0.95, and the rotational speed is 53 rpm. Overall, the proposed strategy offers an effective and accurate model for simulating the process of particle separation, which can lead to improved uniformity of asphalt mixtures.