一种基于多元Probit模型的随机网络配流加载算法

为了改善基于Logit模型的随机网络配流加载算法的两个不足之处，本文提出了基于正态分布的Probit模型的随机网络配流算法。假设路段的广义交通时间服从正态分布，然后用蒙特卡洛模拟法对提出的算法进行求解。最后，用一个简单的交通网络例子对提出的算法进行验证分析。在算法求解过程中，运用Matlab对路段广义交通时间随机抽样，再对网络进行全有全无分配，将交通流量分配到起终点间的每一O-D对的最短路径上。应用Probit模型进行配流加载，其路径选择概率更接近实际更加合理，是一种值得推广的交通配流加载模型。

Algorithm of Random Network Flow Distribution Based on Multivariate Probit Model

For improving the two disadvantages of Logit model based on a random network flow distribution algorithm, a stochastic traffic assignment algorithm on the basis of normally distributed Probit model is proposed in this paper. Assuming that the generalized travel time of section follows the normal distribution, Monte Carlo method is used to solve this algorithm, then a simple traffic net to validate this algorithm was used. Matlab was used to randomly select the sample of the generalized travel time of the se...