考虑出行行为的机场轨道交通服务水平优化方法  

Level of Service Optimization Method for Airport Rail Transit Considering Travel Behavior

作　　者：张锐[1,2] 薛紫薇[1] 刘昕睿 ZHANG Rui;XUE Ziwei;LIU Xinrui(School of Transportation Engineering,Chang'an University,Xi'an 710064,China;Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]长安大学运输工程学院,西安710064 [2]北京交通大学综合交通运输大数据应用技术交通运输行业重点实验室,北京100044

出　　处：《交通信息与安全》2024年第6期143-151,共9页Journal of Transport Information and Safety

基　　金：国家自然科学基金项目(72371035);陕西省重点研发计划项目(2023YBGY143);综合交通运输大数据应用技术交通运输行业重点实验室(北京交通大学)开放基金(ZHJTDSJ202203)资助。

摘　　要：为实现机场轨道交通运能利用率与服务水平、运营收益之间的有效平衡,提出了考虑机场旅客出行方式选择行为的机场轨道交通服务水平优化方法。融合反映行为惯性的显示性偏好(revealed preference,RP)依赖与陈述性偏好(stated preference,SP)依赖变量,以及与行为惯性相关的其他变量(如常住人口、换乘难度等),构建考虑行为惯性的嵌套Logit(nested Logit,NL)模型模拟旅客前往机场过程中的陆侧出行方式选择行为。在NL模型的基础上构建机场轨道交通服务水平优化模型探讨轨道交通票价、发车间隔、高峰小时满载率等服务水平要素与轨道交通分担率、轨道交通运营企业收益之间的互动关系,提出相应的机场轨道交通服务水平优化方案。以西安市为例进行实证研究,结果表明:(1)受访者的SP问卷调查结果受行为惯性的影响,即受访者实际发生过的出行方式选择行为和先前的SP问卷调查选择结果会对受访者当下的SP问卷调查结果产生正向影响,且前者的影响程度大于后者;(2)考虑行为惯性的NL模型相较传统不考虑行为惯性的多项Logit(multinomial Logit,MNL)模型,可有效提升模型精度40.86%;(3)优化模型可经枚举法求解,若将案例城市当前机场轨道交通的拥挤程度(“有些拥挤”)优化至“临界状态”评价标准,为使轨道交通分担率不降低的同时客票收入最大化,则发车间隔缩小1 min前提下,地铁票价增幅建议为3元。This paper aims to balance the utilization rate of transport capacity,level of service and operational profit for airport rail transit,a level of service optimization method for airport rail transit is proposed,considering travel behavior.Specifically,a nested Logit(NL)model considering behavior inertia is established to simulate the landside travel mode choice behavior of passengers accessing the airport.The model includes variables related to behavioral inertia,such as revealed preference(RP)dependence,stated preference(SP)dependence,local residency,and trans-fer difficulty.On this basis,a level of service optimization model for airport rail transit is built to investigate the rela-tionships among level of service factors,such as ticket price,departure interval,peak hour load factor,as well as market share of rail transit and operation profit of rail transit.Corresponding level of service optimization proposals are then presented.The results of case study conducted in Xi’an city show that:①SP survey results are influenced by behavioral inertia.Specifically,the current SP questionnaire survey results of respondents are positively depen-dent on their actual travel choice behaviors and the previous SP survey results,while the dependence on actual trav-el choice behavior is higher than that on previous SP survey results.②The proposed NL model,which considers be-havioral inertia,outperforms the conventional multinomial Logit(MNL)model that does not account for behavioral inertia.The R-squared value of the NL model is improved by 40.86%compared to the MNL model.③The proposed optimization model can be solved using enumeration method.If the congestion degree of the current airport rail tran-sit(“somewhat crowded”)in the case city is expected to reach the"critical state"threshold,to avoid a decrease in the market share of rail transit and maximize ticket revenue,it is recommended that the ticket price be increased by 3 yuan,provided that the departure interval is reduced by 1 minute.

关 键 词：轨道交通 机场旅客 服务水平优化 嵌套Logit模型 行为惯性 

分 类 号：U121[交通运输工程]