基于可变客流的接运公交网络优化

Feeder Bus Network Design Based on Variable Demand

在充分考虑目的地非地铁站的变动客流和搭乘固定公交的既有客流基础上,基于多对多客流模式,以管理者、出行者和社会运营费用的总费用最小为目标,构建接运公交线路的优化模型.模型考虑了原本私家车出行客流和固定公交出行客流选择接运公交出行的可能性,应用Logit模型进行流量分配,并采用遗传算法对问题求解,获得了最优的接运公交网络,变动客流在接运公交网络中的第一公交站和换乘的地铁站.研究结果表明,接运公交线路方案与其占全程广义出行费用的比例密切相关,故有必要将其从全程视角进行优化.

This study takes account of the private car demands whose destinations are not subway station and the demands taking the fixed bus lines originally, develops a feeder bus network optimal model with the minimum cost of passengers, operators and social for many-to-many travel demand. Logit model is applied to redirect private car demand between private car and feeder bus line, fixed bus line demand between feeder bus and fixed bus line. A genetic algorithm is developed to solve the problem. The results show that the model constructed in this paper and the applied genetic algorithm can obtain optimal feeder bus lines, the first bus station and transfer subway station for originally private car demand. The solution result shows that the feeder bus line strategies have a close relationship with the percentage of the whole traveling chain costs.