西安建筑科技大学学报(自然科学版)

对智能公交系统数据进行分析和挖掘,能够为城市公共交通系统的规划和管理提供科学的决策依据.一票制的公交IC卡收费系统仅在上车时刷卡,乘客下车站点的信息无法直接得到,而现有方法只能判断出部分乘车记录的下车站点,这成为公交IC卡数据应用的制约因素.根据公交乘客出行的时空规律性,提出基于通勤出行模式和关联出行模式判断下车站点的思路.利用通勤出行规律判断公交乘客的通勤OD,再根据其居住地和工作地站点信息对通勤出行的下车站点进行估计.依据公交乘客长期出行具有一定时空关联性的规律,通过将当日出行与他日的关联出行数据进行匹配来判断下车站点.对提出的方法进行算法实现,并应用大规模实际数据对算法进行了实验和精度分析.研究结果表明:该方法能够有效提高下车站点的判断比例,准确率达到87.90%,能够应用于实际的公交系统规划和管理工作.

Analyzing and mining the data of Advance Public Transportation Systems can provide scientific decision-making basis for urban public transportation planning and management. In flat-rate automated fare collection system, alighting stops information of the passengers is not available directly. The existing method can only determine part of alighting stops, which is a limiting factor for the application of smart card data. According to time-spatial regularity of transit travel, this paper presents the method of estimating alighting stops based on commuting travel patterns and related travel patterns. Commuting OD is judged by commuting regularity, and then the alighting stops of commuting trips are estimated on the basis of commuters' residences and work places. Based on long-term time-spatial correlation of travels, alighting stops are estimated by matching the related travels. The algorithm was implemented, and the experiment and accuracy analysis were carried out using large-scale actual data. The results show that the method can help improve the ratio of estimating alighting stops effectively, and the accuracy ratio reached 87.90%. This can be used in practical work of public transportation planning and management.

引言
1 研究数据描述
2 基本思想
3 算法实现
4 算法实验
5 算法验证
6 结论

图1 4种数据源的关联关系[10]<br/>Fig.1 Association of four data sources[10]

图1 4种数据源的关联关系[10]
Fig.1 Association of four data sources[10]

表1 研究数据字段描述<br/>Tab.1 Fields description of research data

表1 研究数据字段描述
Tab.1 Fields description of research data

表2 出行链方法计算结果统计<br/>Tab.2 Statistics of trip chain method calculation results

表2 出行链方法计算结果统计
Tab.2 Statistics of trip chain method calculation results

表3 通勤OD运算结果统计[12]<br/>Tab.3 Statistics of commuting OD calculation results[12]

表3 通勤OD运算结果统计[12]
Tab.3 Statistics of commuting OD calculation results[12]

图2 刷卡1次记录的判断结果<br/>Fig.2 Calculation results of the records that only used once in a single day

图2 刷卡1次记录的判断结果
Fig.2 Calculation results of the records that only used once in a single day

图3 出行链方法推算失败记录的判断结果<br/>Fig.3 Calculation results of the records that cannot be estimated by trip chain method

图3 出行链方法推算失败记录的判断结果
Fig.3 Calculation results of the records that cannot be estimated by trip chain method

表4 出行模式方法计算结果统计<br/>Tab.4 Statistics of travel pattern method experiment results

表4 出行模式方法计算结果统计
Tab.4 Statistics of travel pattern method experiment results

表5 算法验证结果<br/>Tab.5 Verification results of the algorithm

表5 算法验证结果
Tab.5 Verification results of the algorithm

[1] BAGCHI M, WHITE P. The potential of public transport smart card data[J].Transport Policy, 2005(12):464-474.
[2] PELLETIER M, TREPANIER M, MORENCY C. Smart card data use in public transit: a literature review[J]. Transportation Research Part C,2011, 19: 557-568.
[3] 褚浩然, 郑猛, 杨晓光,等. 出行链特征指标的提出及应用研究[J].城市交通,2006, 4(2):64-67.
[4] BARRY J, NEWHOUSER R, RAHBEE A, et al. Origin and destination estimation in New York city using automated fare system data[J]. Transportation Research Record, 2002,(1871):183-187.
[5] ZHAO J H, RAHBEE A, WILSON N. Estimation a passenger trip origin-destination matrix using automatic data collection systems[J]. Computer-Aided Civil and Infrastructure Engineering, 2007, 22(5):376-387.
[6] FARZIN J M. Constructing an automated bus origin-destination matrix using farecard and GPS data in Sao Paulo, Brazil[J]. Transportation Research Record, 2008, 2072: 30-37.
[7] CHU K K, CHAPLEAU R, TREPANIER M. Driver-assisted bus interview(DABI): passive transit travel survey using smart card automatic fare collection system and its applications[J]. Transportation Research Record, 2009, 2105: 1-10.
[8] KUSAKABE T,IRYO T, ASAKURA Y. Estimation method for railway passengers' train choice behavior with smart card transaction data[J]. Transportation, 2010, 37(5):731-749.
[9] CHEN J, WANG Z F. Algorithm of estimating alighting bus stops of smart card passengers based on trip-chain[C]. 2nd International Conference on Civil Engineering and Transportation. Guilin: TTP, 2012: 1918-1921.
[10]陈君,杨东援.基于智能调度数据的公交IC卡乘客上车站点判断方法[J].交通运输系统工程与信息,2013,13(1):76-80.
[11]刘靓. 普适计算环境下居民交通行为特征研究[D].上海:同济大学, 2008.
[12]陈君,杨东援.基于APTS数据的公交卡乘客通勤OD分布估计方法[J].交通运输系统工程与信息,2013,13(4):47-53.

备注

引言

1 研究数据描述

2 基本思想

3 算法实现

4 算法实验

5 算法验证

6 结论

期刊信息