在公路工程可行性研究和路网规划中 ,由于经济指标受诸多因素影响 ,常用经济预测方法不能够反映各因素对经济指标的综合影响。神经网络是模拟人脑结构和功能的一种以计算机为工具的信息处理系统 ,具有自学习特性 ,能建立网络输入、输出变量之间的非线性映射关系 ,因而常用于识别、预测等问题。该文将神经网络方法应用于公路工程可行性研究中的社会经济发展预测 ,通过与常规方法预测结果的对比 ,表明该方法在这一领域的应用是可行的 In the feasibility study of road engineering and road network planning, due to the influence of many factors on economic indicators, the common economic forecasting methods can not reflect the comprehensive impact of various factors on economic indicators. Neural network is a computer-based information processing system that simulates the structure and function of the human brain. It has the characteristics of self-learning and can establish the nonlinear mapping relationship between network input and output variables. Therefore, it is often used in identification and prediction. In this paper, the neural network method is applied to the prediction of socio-economic development in the feasibility study of highway engineering. The comparison with the prediction results of conventional methods shows that the method is feasible in this field