Highway Capacity Prediction Based on Neural Network

Traffic congestion has become an important factor that reduces the quality of people's livelihood and economic development. At present, the main way to solve the traffic congestion is to direct the traffic artificially, which has lag and is relatively inefficient. It is of great significance to accurately predict highway transport capacity for short-distance transport scheduling. After proving the nonlinear relationship between all relevant factors and highway capacity by spearman correlation coefficient, with the number of people, the number of motor vehicles, highway area as input variables, highway passenger volume and highway freight volume as output variables, a back propagation neural network model was established to predict highway capacity. The results show that the predicted results of the adjusted model are more accurate, the training <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R^2</tex> for highway passenger volume and highway freight volume is 0.9943 and 0.9815 respectively, and the test error under non-interference condition is also within 0.15. This study provides a reference for forecasting highway transport capacity.