【作者】 康振； 马成元； 杨晓光； 张力楠；

【Author】 Kang Zhen;Ma Chengyuan;Yang Xiaoguang;Zhang Linan;Key Laboratory of Road and Traffic Engineering of the Ministry of Education,Tongji University;

【机构】 同济大学道路与交通工程教育部重点实验室；

【摘要】 本文研究了利用高速公路收费数据与稽查数据作为数据资源进行车辆逃费行为预测的可行性,并完整分析实现了相关算法。选取广东省高速公路网2017年7月1日—2017年7月31日的高速公路收费数据与4—7月的稽查数据作为基础,介绍了数据的分析处理过程,包括进行数据准备,执行各类数据质量控制规则等。本文采取贝叶斯分析的方法,对影响车辆通行行为的指标进行筛选,并采取最大熵模型与贝叶斯模型结合的方法确定指标权重,实现了基于历史数据对车辆的逃费行为进行概率预测。更多还原

【Abstract】 This paper studies the feasibility of using the expressway toll data and the inspecting data as data resources to predict the traffic behavior of the vehicle, and completes the design of the algorithm. This paper is based on the toll data of Guangdong expressway network in July 1, 2017 —2017 July 31 st and April —7 month check data. The process of data analysis and processing is introduced in this paper., including data preparation, implementation of various data quality control rules and so on. In this paper, the method of Bias analysis is adopted to screen the index of vehicle traffic behavior, and the method of maximum entropy model and Bias model is adopted to determine the index weight. In this way, the traffic behavior of vehicles based on historical data is predicted.更多还原