高速铁路具有运营成本高、运输需求波动大、供给网络优化复杂等特征，客流分配是一体化客运计划优化的技术瓶颈，既有客流分配方法以静态列车时空局域服务网络为研究视角，在识别网络拥挤状态、获取需求转移、揭示供给与需求双方策略互馈下客流分布等方面呈现出局限性，亟需解决供需动态环境下客流分配问题。课题组前期发现旅客有策略购票行为和运营者的票额分配策略都会影响客流在服务网络中的分布，因此客票预售阶段成为供需动态匹配的核心环节。本项目旨在探究服务供给网络能力瓶颈、明晰需求转移机制、分析供给策略与需求策略互馈机理，提出扩展性三维座席服务供给网络建模方法、严格座席能力约束下弹性异质型需求选择行为模型、基于需求和供给的双策略下多日动态客流分配模型，设计不动点迭代算法，开发旅客全出行链仿真实验平台。本项目成果为高速铁路推行“动态定价”和“一日一图”等供给决策优化提供科学依据，为一体化客运计划优化提供理论和技术支撑。

High-speed railway has many characteristics such as high operation cost, large fluctuation in transportation demand and complex supply network optimization. Passenger flow assignment method is the technical bottleneck of integrated passenger transportation plan optimization. Many existing passenger flow assignment methods take static temporal-spatial trains service network as a perspective, showing some limitations in identifying the network congestion state, obtaining the demand transfer situation, revealing the distribution of passenger flow under strategic interaction between supply and demand. Therefore, it is urgent to solve the problem of passenger flow assignment under dynamic environment of supply and demand. In early stage, the research team found that the strategic ticket purchase behavior of passenger and the ticket allocation strategies of operators both affect the distribution of passenger flow, so the ticket pre-sale stage becomes the core link of dynamic matching between supply and demand. The purpose of this research is to explore the bottleneck of service supply network capabilities, clarify the demand transfer mechanism, analyze the interaction between supply and demand. In this project, the relevant mathematics models of expansible three-dimensional seat supply network and selection behavior of elastic heterogeneous demand under strict seat capacity constraints will be proposed. Besides, the model of day-to-day dynamic passenger flow assignment based on strategies interaction between demand and supply will be formulated. Furthermore, the fixed-point iterative algorithm will be designed. In addition, the experiment simulation platform of passenger’s the whole trip chain will be established to form the theory and methodology of dynamic passenger assignment. The results gained from this project are the scientific basis for operator carry out supply decisions such as dynamic pricing and timetable, and will provide theoretical and technical supports for integrated optimization of passenger transportation plan.