基于期望-超额行程时间与有限理性的城市交通分配模型  

Urban traffic assignment model based on mean-excess travel time and bounded rationality

作　　者：李洁[1] 张萌佳 陈思妤 LI Jie;ZHANG Meng-jia;CHEN Si-yu(College of Civil Engineering,Hunan University,Changsha 410082,Hunan,China)

机构地区：[1]湖南大学土木工程学院,湖南长沙410082

出　　处：《长安大学学报（自然科学版）》2024年第3期125-137,共13页Journal of Chang’an University(Natural Science Edition)

基　　金：国家自然科学基金项目(51878264);湖南省科学技术厅重点研发项目(2022SK2096);河南省交通运输厅科技项目(2020G11)。

摘　　要：为完善现有交通分配方法,综合考虑路网与出行者两方面不确定性因素对随机出行行为的影响,建立基于代理的城市路网交通分配模型。首先,采用期望-超额行程时间反映路网行程时间的客观不确定性,并构建通行能力随机退化条件下的出行成本函数。随后,引入出行者对路径成本的感知更新过程,采用有限理性理论描述出行者的主观不确定性,并结合Logit公式确定路径决策算法。接着,采用轮盘赌随机算法生成出行者每日出行选择,完成城市交通分配模型构建。最后,将所建立的模型分别应用于双路径网络、九点网络与Sioux-Falls网络,分析模型优势及参数变化对路网性能的影响。研究结果表明:路网达到稳定所需的天数随路网退化程度、出行者对成本差异的敏感程度和经验学习能力的提高而变长;稳定状态下的路网总成本随路网退化程度、出行者对行程时间可靠度要求的升高而变大;追求个人行程时间的可靠性会导致稳定状态下的路网总成本升高,可靠度越高,所需的缓冲时间越多;路网退化越剧烈,这种“升高”越明显,因为超额时间也在增大;相比于用户均衡(UE)模型、有限理性(BR)模型和期望-超额行程时间(METT)模型,所建模型兼顾主客观不确定性,能有效克服传统模型低估较长路径上流量的状况,分配结果更加贴合实际;该模型适用于城市大型交通路网的交通分配,其不确定性参数可通过相关交通管理政策的制定而改变,从而实现缓解交通拥堵、提升路网整体运行效率的目的。An agent-based traffic assignment model of the urban road network was established to refine current traffic assignment methods,considering the influence of uncertain factors of the road network and travelers on random travel behavior.Firstly,the concept of mean-excess travel time was employed to encapsulate the objective uncertainty of travel time in the road network.The travel cost function under the condition of random capacity degradation was constructed.Then the bounded rationality theory was used to describe travelers' subjective uncertainty,considering the updating process of travelers' perception of route cost.The route decision algorithm was determined by combining the Logit formula.Subsequently,the stochastic roulette algorithm was used to generate travelers' daily travel choices,and the urban traffic assignment model was constructed.Finally,the established model was applied to the dual-route,nine-point,and Sioux-Falls networks to analyze model advantages and the influence of parameter changes on the network performance.The results show that the number of days required for the road network to reach stability increases with the enhancement of road network degradation degree,travelers' sensitivity to cost differences and experiential learning ability.The total cost of the road network under stable conditions increases with the increase of road network degradation degree and travelers' requirement for travel time reliability.Pursuing individual travel time reliability can expand the network's total cost under stable conditions because the higher the reliability,the more buffer time.The more severe the degradation of the road network,the more noticeable this lift is,as the excess time also increases.Compared to the UE,BR and METT models,the established model considers both subjective and objective uncertainties can effectively overcome the limitation of traditional models,such as underestimating the traffic on a long route,making the results more accurate.Moreover,the proposed model is suitable for assigning

关 键 词：交通工程 交通分配 有限理性 期望-超额行程时间 城市道路网 主客观不确定性 

分 类 号：U491[交通运输工程—交通运输规划与管理]