根据公交换乘枢纽换乘量的生成特点和影响交通换乘量的主要组成因素,研究发现交通换乘量具有“小样本、贫信息”的灰色特征,为此提出了一种基于灰色软计算的换乘需求量预测方法。该方法利用灰色系统原理建立灰色神经网络系统预测模型,通过采用遗传算法改进神经网络的性能,提高系统预测的精度。以兰州市市区公共交通枢纽规划为例,结合实际的道路交通调查数据,运用该方法对提出的交通枢纽方案进行了实证分析与评价。结果表明:改进的灰色神经网络能有效地改善预测精度;运用GA-GNN模型求解道路交通中的非线性问题对提高决策水平具有较大的现实意义。 According to the generative characteristics of transfer capacity and the main factors that affect the traffic transfer, it is found that the traffic transfer has the gray characteristic of “small sample, poor information ”. Therefore, a gray Soft Computing Transfer Demand Forecasting Method. The method uses the gray system principle to establish the gray neural network system prediction model, and improves the prediction accuracy of the system by adopting the genetic algorithm to improve the performance of the neural network. Taking the planning of urban public transport hub in Lanzhou as an example, this paper analyzes the proposed transport hub scheme through empirical analysis and evaluation based on the actual road traffic survey data. The results show that the improved gray neural network can effectively improve the prediction accuracy. Using the GA-GNN model to solve the nonlinear problem in road traffic is of great practical significance to improve the decision-making level.