摘要
为准确地统计和预测城市轨道交通客流,以乘客路径选择行为为研究对象,分析乘客路径选择行为的半补偿性和异质性,将属性切除点引入传统补偿模型框架,在效用函数中增加切除点违反量惩罚项,同时也考虑乘客的偏好异质性,采用随机模型参数,构建半补偿ML路径选择模型。以北京城市轨道交通为例,结合乘客路径选择行为调查数据,采用极大仿真似然法对所建模型进行验证评估,并与补偿型模型进行对比。研究结果显示,半补偿ML路径选择模型对样本数据的拟合效果最优,能够提高模型对路径选择行为的解释力度。
To accurately estimate and predict passenger flow in urban rail transit, the semicompensation and heterogeneity of passenger route choice behavior are analyzed taking passenger travel behavior as the research object. The attribute cutoff point is introduced into the framework of traditional compensation model, the penalty term of cutoff point violation is added to the utility function, and the passenger preference difference is also considered, which leads to the formulation of a penalized utility function. Considering the heterogeneity of preferences among passengers, a semi compensated Mixed Logit path selection model is established by using random model parameters. Taking Beijing urban rail transit as an example, combined with the survey data of passenger travel behavior, simulated maximum likelihood method is used to validate and evaluate the model, and compared with the compensation model. The results show that the semicompensated Mixed Logit path selection model has the best goodness-of-fit and contributes to the explanation of observed choice.
作者
赵凯华
李海鹰
ZHAO Kai-hua;LI Hai-ying(School of Traffic and Transportation,Beijing Jiaotong University,Beijing 100044,China;State Key Lab of Rail Traffic Control ＆ Safety,Beijing Jiaotong University,Beijing 100044,China)
出处
《铁道运输与经济》
北大核心
2018年第11期110-115,共6页
Railway Transport and Economy
基金
国家"十三五"重点研发计划课题(2016YFB1200402)
