基于IC卡综合换乘信息的公交乘客上车站点推算
作者简介:宋晓晴(1991-),女,硕士生,研究方向为时空数据挖掘与交通地理信息系统。E-mail:xqsong@whu.edu.cn
收稿日期: 2015-08-05
要求修回日期: 2015-11-14
网络出版日期: 2016-08-10
基金资助
国家自然科学基金项目(41231171、41371420)
中国科学院资源与环境信息系统国家重点实验室开放基金(2013)
武汉大学自主科研项目拔尖创新人才类资助项目(2042015KF0167)
A Method of Deriving the Boarding Station Information of Bus Passengers Based on Comprehensive Transfer Information Mined from IC Card Data
Received date: 2015-08-05
Request revised date: 2015-11-14
Online published: 2016-08-10
Copyright
现有研究在缺少公交车运营信息的情况下,利用公交线间正交换乘信息的方法来识别公交车持卡乘客的上车站点,但在实际运用中很多班次无正交换乘的乘客,容易导致公交乘客上车站点匹配困难。因此,本文提出了一种基于IC卡综合换乘信息的公交乘客上车站点推算方法,该方法充分利用IC卡数据中的综合换乘信息(公交线间的正交和非正交换乘信息、地铁和公交线间的换乘信息)与公交网数据确定班次行驶方向,对班次内的乘客进行分组,充分利用综合换乘信息确定基准组及其对应的上车站点;然后,通过最小偏差规则匹配待定组的上车站点,实现公交乘客上车站点的推估。利用换乘信息确定班次行驶方向弥补了现有研究无法确定班次行驶方向的不足,使公交乘客上车站点推算方法更科学;最后,基于2011年8月的“深圳通”IC卡数据进行实例分析,对比只利用正交换乘信息确定的基准组数量和利用综合换乘信息确定的基准组数量,结果表明:该方法可使确定上车站点的组数占总组数的比值提高,克服了只利用正交换乘信息推算上车站点方法中存在的单个班次无基准组从而无法进行站点匹配的困难。本文方法比只利用正交换乘信息匹配站点,具有更高的可行性与准确率。
宋晓晴 , 方志祥 , 尹凌 , 刘立寒 , 杨喜平 , 萧世伦 . 基于IC卡综合换乘信息的公交乘客上车站点推算[J]. 地球信息科学学报, 2016 , 18(8) : 1060 -1068 . DOI: 10.3724/SP.J.1047.2016.01060
Due to the lack of bus operating data, some scholars put forward a method, which uses the orthogonal transfer information between different bus routes, to derive the boarding station information of bus passengers. However, a lot of the bus single trips lack the orthogonal transfer information in practical applications, which lead to the difficulty in deriving the boarding station information of bus passengers. To solve this issue, based on the existing research, this paper put forward a method for deriving the boarding station information of bus passengers based on the comprehensive transfer information mined from IC card data. This method uses the comprehensive transfer information between the subways and buses information mined from the IC card data to confirm the driving direction of buses, and then, divide the passengers of the same single trip into groups. These groups would match the corresponding boarding stations more accurately. We fully take use of the orthogonal transfer information and non-orthogonal transfer information between different bus routes and the transfer information between subways and buses to recognize the fiducial groups and their boarding station information, then, we use the minimum error rule to derive the pending groups′ boarding stations. Using the transfer information to confirm the driving direction could enlarge existing research on how to determine a single trip's driving direction, which makes the method of driving the boarding station information of bus passengers more complete and scientific. In the end, through example analysis, this study greatly increases the number of fiducial groups. This method has a higher reliability and accuracy compared to the method which only uses the orthogonal transfer information between different bus routes to derive the boarding station information of bus passengers. In this study, we use the transfer information to confirm the driving direction of some single trips with the absence of the location data. In addition, we use the transfer information mined from three transfer approaches to derive the boarding station information of bus passengers as well as put forward a series of constraint rules to make the result more accurate and feasible, which are considered to be the most significant innovation and contribution. Deriving the boarding station information of bus passengers is the up-front work in getting the bus passengers' demands and the bus service ability data, which has a guiding significance to urban planning.
Key words: IC Card data; traffic data mining; transfer information; boarding station
Fig. 1 Three ways of transfering to the target bus line图1 换乘到目标公交线的3种换乘方式 |
Fig. 2 The process of confirming the boarding stations of bus passengers图2 公交乘客上车站点的推算流程 |
Tab. 1 The boarding station information of some passengers in the same single trip (initially)表1 同班次中部分乘客上车站点信息(初步) |
乘客编号 | 卡编号 | 上车站点 | 上行编号(U) | 下行编号(D) | 上车时间(T) | 班次号 |
---|---|---|---|---|---|---|
1 | 0235***91 | 科技园② | 38 | 16 | 07:53:03 | 1 |
2 | 3243***02 | 竹园宾馆① | 13 | 41 | 08:51:53 | 1 |
… |
Tab. 2 Transfer station information of some passengers (partial information)表2 乘客的换乘站点信息(部分) |
乘客编号 | 卡编号 | 换乘前路线 | 班次号(tr) | 分组号(j) | 可能的换乘站点序号(i) |
---|---|---|---|---|---|
1 | 0203***54 | 80路 | 1 | 5 | 8、9、10、11、12 |
2 | 3266***86 | 74路 | 1 | 13 | 9、10、14 |
Tab. 3 Transfer station information of some other passengers (partial information)表3 乘客的换乘站点信息(部分) |
乘客编号 | 卡编号 | 换乘前路线 | 班次号(tr) | 分组号(j) | 可能的换乘站点序号(i) |
---|---|---|---|---|---|
1 | 3252***49 | 211路 | 2 | 12 | 13、14、15 |
2 | 0208***83 | 82路 | 2 | 12 | 15、16 |
Fig. 3 The process of deriving the boarding stationsfor the pending groups图3 待定组上车站点的匹配过程 |
Tab. 4 Shenzhen IC card data record (partial information)表4 深圳市IC卡数据记录(部分) |
卡编号 | 交易时间编码 | 闸机编号 | 交易类型 | 实价/分 | 原价/分 | 路线 | 站点 |
---|---|---|---|---|---|---|---|
0209***79 0209***85 3236***68 | 20110801101456 20110801082026 20110801081948 | 220001279 268025118 268012115 | 31 21 22 | 120 0 270 | 200 0 300 | 836路 1号线 1号线 | 无 桃园站 车公庙站 |
注:交易时间“20110801101456”表示“2011年8月1日10时14分56秒”;交易类型“31”表示公交车上车刷卡,“21”表示地铁入站,“22”表示地铁出站 |
Tab. 5 The statistics of the result for determining the fiducial groups of different routes表5 不同路线识别基准组结果统计表 |
线路名 | 班次数 | 基准组数量 | 正交换乘基准组数量 | 非正交换乘基准组数量 | 地铁换乘目标车基准组数量 |
---|---|---|---|---|---|
79路 | 4 | 13 | 2 | 2 | 9 |
101路 | 3 | 6 | 0 | 1 | 5 |
113路 | 6 | 18 | 5 | 3 | 10 |
223路 | 10 | 19 | 7 | 2 | 10 |
311路 | 6 | 11 | 4 | 0 | 7 |
320路 | 6 | 23 | 10 | 6 | 7 |
323路 | 4 | 12 | 4 | 4 | 4 |
338路 | 6 | 19 | 4 | 3 | 12 |
383路 | 8 | 15 | 2 | 4 | 9 |
395路 | 4 | 13 | 5 | 1 | 7 |
Tab. 6 The result for determining the fiducial groups of bus line 113表6 113路识别基准组结果 |
班次数 | 班次号 | 本研究确定的基准组数量 | 本研究确定的 基准组组号 | 正交换乘确定的 基准组数量 | 正交换乘确定的 基准组组号 |
---|---|---|---|---|---|
6 | 1 | 5 | 13、15、34、39、40 | 2 | 15、39 |
2 | 2 | 11、12 | 1 | 11 | |
3 | 3 | 11、21、25 | 0 | ||
4 | 3 | 17、26、28 | 1 | 28 | |
5 | 3 | 18、24、27 | 0 | ||
6 | 2 | 24、25 | 1 | 24 |
Fig. 4 Dividing the bus passengers into groups of six single trips图4 刷卡乘客分成6个班次 |
Fig. 5 Dividing first single trip's passengers into several groups图5 班次1乘客分组 |
Fig. 6 The results of deriving the boarding stations图6 上车站点推算结果 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
[
|
[4] |
|
[5] |
[
|
[6] |
[
|
[7] |
[
|
[8] |
[
|
[9] |
[
|
[10] |
[
|
[11] |
[
|
[12] |
[
|
[13] |
[
|
[14] |
[
|
[15] |
|
[16] |
[
|
[17] |
[
|
[18] |
[
|
/
〈 | 〉 |