数据融合视角下限行政策替代策略研究  

Study on Alternative Strategies of Driving Restriction Policy from the Perspective of Data Fusion

作　　者：漆威 宋钰元 黄恒君 QI Wei;SONG Yuyuan;HUANG Hengjun(School of Finance,Lanzhou University of Finance and Economics,Lanzhou 730020,China;School of Statistics,Lanzhou University of Finance and Economics,Lanzhou 730020,China;Key Laboratory of Digital Economy and Social Computing Science of Gansu,Lanzhou University of Finance and Economics,Lanzhou 730020,China)

机构地区：[1]兰州财经大学金融学院,甘肃兰州730020 [2]兰州财经大学统计学院,甘肃兰州730020 [3]兰州财经大学甘肃省数字经济与社会计算科学重点实验室,甘肃兰州730020

出　　处：《统计与信息论坛》2023年第8期101-110,共10页Journal of Statistics and Information

基　　金：国家社会科学基金一般项目“城市空气质量影响因素分析及环境政策评价”(18BTJ001);国家社会科学基金西部项目“面向城市计算的多领域数据融合方法研究”(20XTJ005);中央引导地方科技发展项目“城市计算方法体系构建及甘肃智慧城市应用”(YDZX20216200001876)。

摘　　要：数据融合在大数据研究中至关重要,在城市计算背景下讨论数据融合方法具有典型的意义。随着社会经济的发展和城市化进程的加快,交通拥堵、大气污染等已成为城市治理中的棘手问题。已有研究表明,机动车限行政策对于交通流量控制和大气污染防治短期有效,但长期失效;而交通拥堵费政策因技术要求高,难以在全球范围内推广。鉴于此种情况,试图吸纳限行政策便于实施的特点,规避拥堵收费政策的技术普及难题,通过概率模型融合传统调查数据与行政记录数据,讨论差异化停车费率作为机动车限行政策的经济激励型替代策略的有效性,为城市计算应用提供新的研究场景。首先,依据实际停车支付意愿随停车费率上升而递减的特征,提出以广义帕累托分布作为停车支付意愿分析工具,依据弹性理论,得到停车需求量对价格变动反应敏感的结论,并给出停车费定价的政策替代原理。其次,根据数据特征,融合停车支付意愿调查数据和停车收费订单记录微观数据,并基于价格歧视理论,提出融合停车费率和停车时长因素的差别化停车费率定价模型,用以分析机动车限行政策的替代效果。最后,作为应用,针对兰州市尾号限行和单双号限行的情景开展分析,给出与机动车限行政策等效、与当前停车收费标准相衔接的停车费率三阶梯定价设计和定价建议,从静态交通管理视角为中国城市的停车管理提供一种思路。Data fusion is paramount in big data research,and it is of typical significance to discuss data fusion methods in the context of urban computing.With the development of the social economy and the acceleration of urbanization,traffic congestion and air pollution have become important issues of urban governance.In practice,vehicle restrictions and congestion charging are considered as important traffic demand management policies to deal with traffic congestion and air pollution.However,previous studies have shown that driving restriction on vehicles is effective for controlling traffic flow and preventing air pollution in the short term,but ineffective in the long run;While congestion charging policy is difficult to implement worldwide due to its higher requirement for technical equipment.In view of these situation,this paper aims to design an easy-implemented economic incentive measure that makes full use of the existing urban infrastructures,as an alternative strategy to the vehicle restriction policy.Based on the fusion of traditional survey data and administrative record data,using the probabilistic mode,the effectiveness of taking the differentiated parking fee is discussed,as an economic incentive alternative to the policy of driving restriction,which provides a new scenario for urban computing application.Firstly,according to the feature that actual willingness to pay for parking decreases with the increase in parking rates,a generalized Pareto distribution is proposed as the analysis tool.On the basis of the elasticity theory,the conclusion that parking demand is sensitive to price changes is drawn,and the principle of parking fee pricing as a policy alternative for driving restriction is elaborated.Secondly,a fusion procedure of the survey data on willingness to pay for parking and the micro-data of parking charge order records is implemented based on the data feature,which effectively deals with the problem that data from different sources cannot be combined due to the differences in sample size.Then a di

关 键 词：停车费率 支付意愿 数据融合 限行政策 交通拥堵费 城市计算 

分 类 号：O212[理学—概率论与数理统计] F505[理学—数学]