Abstract: Traffic forecasting and accurate forecasting are central to highway feasibility studies and have important implications for traffic planning and management research. As the number of vehicles increases, transportation issues become more complex. A hot issue that theoretical researchers are concerned about is traffi c forecasting, as traffi c problems become more complex and the number of vehicles continues to grow. In this article, when optimizing a GM (1,1) model using regression analysis and gray model theory, we first combine intelligent calculations. Reasonable weights can be obtained by combining the regression model with the optimized GM (1,1) model. The establishment of a combination prediction model for linear traffi c volume is mainly based on the gray model and the regression model. The actual data testing is mainly done on the Lima highway, and this model can eff ectively improve the accuracy of short-term traffi c forecasts.