基于随机参数Logit的市域旅客出行方式选择影响研究  

Research on the Impact of Urban Passengers’Travel Mode Choice Based on Random Parameter Logit

作　　者：马瑞洋 王兴仁 张显辉 Ma Ruiyang;Wang Xingren;Zhang Xianhui(School of Traffic and Transportation,Xi'an Traffic Engineering Institute,Xi'an 710300,China;Shaanxi University Innovation Research Institute of the Future Industry in Rail Transit,Xi'an 710300,China;School of Civil Engineering,Heilongjiang University of Science and Technology,Harbin 150022,China)

机构地区：[1]西安交通工程学院交通运输学院,西安710300 [2]陕西高校轨道交通未来产业创新研究院,西安710300 [3]黑龙江科技大学建筑工程学院,哈尔滨150022

出　　处：《科技通报》2024年第10期62-68,76,共8页Bulletin of Science and Technology

基　　金：陕西省教育厅科学研究计划项目(22JC048);西安市科技计划项目(23GXFM0046);黑龙江科技大学引进高层次人才科研启动基金项目(HKD202134);西安交通工程学院中青年基金项目(2023KY-20);西安交通工程学院中青年基金项目(2024KY-39)。

摘　　要：为了探究市域火车新线开通对旅客出行路径选择的影响,以西安鄠邑区至主城区市域出行旅客为研究对象,分析市域火车开通对旅客出行方式选择的影响。传统多项Logit模型无关方案独立性,无法定量研究旅客出行方式选择的差异性,故本文应用随机参数Logit模型分析市域旅客出行方式选择。选取旅客的个体属性、市域出行方式属性、到离站属性设计问卷,采用线上线下相结合的方式开展调查,获取了1205份有效数据,基于此建立并求解随机参数Logit模型。结果表明:在旅客选择大巴出行时,市域出行距离、职业、购票方式、到站方式和到站时间对旅客出行方式选择影响显著,市域出行时间服从(-0.15,0.146^(2))正态分布的随机参数特征;对市域出行时间随机参数进行边际效应分析,市域出行时间超过30 min后,旅客选择大巴和小汽车出行的概率降低,选择市域火车出行的概率增加;对旅客出行方式选择影响显著变量进行弹性分析,到站费用每增加1%,旅客选择大巴的概率增加0.394%,选择高铁的概率下降1.888%,选择市域火车的概率增加0.569%,选择小汽车的概率降低0.030%。结果证实,旅客对市域出行时间的偏好具有异质性,市域出行时间过长将降低旅客选择大巴和小汽车的概率,减少到站时间可以提高旅客选择市域火车的概率。In order to explore the impact of the opening of new urban train lines on passengers'travel route choices,this paper took passengers traveling from Huyi District to the main urban area of Xi'an as the research object to analyze the impact of the opening of urban trains on passengers'travel mode choices.The traditional multinomial logit model has nothing to do with scheme independence and cannot quantitatively study the differences in passengers'travel mode choices,therefore,this paper uses the random parameter logit model to analyze the city-wide passenger travel mode choices.We selected passengers'individual attributes,urban travel mode attributes,and arrival and departure attributes to design a questionnaire,we conducted a survey using a combination of online and offline methods,and obtained 1205 pieces of valid data.Based on this,we established and solved a random parameter Logit model.The results show that when passengers choose to travel by bus,city travel distance,occupation,ticket purchase method,arrival method and arrival time have a significant impact on passengers'choice of travel mode,the city travel time obeys the random distribution of(-0.15,0.146^(2))normal distribution parameter characteristics;Perform marginal effect analysis on the random parameters of urban travel time,after the urban travel time exceeds 30 minutes,passengers are less likely to choose buses and cars to travel,and more likely to choose urban trains to travel;Conduct elastic analysis on the variables that significantly affect the choice of passenger travel mode,for every 1%increase in arrival cost,the probability of passengers choosing the bus increases by 0.394%,the probability of choosing the high-speed rail decreases by 1.888%,the probability of choosing the urban train increases by 0.569%,and the probability of choosing the car decreases by 0.030%.The results confirm that passengers have heterogeneous preferences for urban travel time.Long urban travel time will reduce the probability of passengers choosing buses and cars,and

关 键 词：市域出行 出行方式选择 随机参数Logit模型 边际效应分析 弹性分析 

分 类 号：U121[交通运输工程]