为高效准确地识别旅客联程出行方式,基于旅客联程出行行为特征,引入不同运输方式场站地理位置、场站数据取样最佳半径、旅客行驶速度等关键参数,对旅客手机信令数据进行筛选、校核和计算,提出了基于手机信令数据的旅客联程出行方式识别方法,同时测算了不同运输方式场站数据取样最佳半径。以2018年国庆假期广东省内旅客出行为例进行分析,剔除了约98%的无关信令数据。分析结果显示,广东省内公铁联运出行比例最高,广州、深圳两大枢纽城市的客流集疏运效应突出,广佛城际出行联系较为密切,佛山机场的潜力较大。研究表明,识别方法大幅降低了信令数据分析量和运算成本,方法原理和技术路线清晰,分析结果准确、符合实际。
Abstract
To accurately and efficiently identify passengers′ travel modes in intermodal transportation, based on the behavior characteristics of intermodal passenger transportation, the geographical location of stations in different transportation modes, the best radius of station data sampling and passengers′ travel speed were introduced as key parameters. Through filtering, utilizing and calculating the mobile signaling data, an identification method to passengers′ travel modes in intermodal transportation was put forward taking advantage of mobile signaling data. The best radiuses of station data sampling of different transportation modes were measured simultaneously. A case study of passenger transportation in Guangdong Province during 2018 National Day holiday was conducted and about 98% irrelevant signaling data were removed. The analysis results showed that the highway-railway travel had the highest proportion within Guangdong province; Guangzhou and Shenzhen were the top two cities which had a prominent effect on passenger collection and distribution; the intercity travels between Guangzhou and Foshan were frequent and Foshan airport had great potential of passenger transportation. The identification method could greatly reduce the signaling data analysis volume and operation cost. Its principle and technical route are clear, which leads to accurate and practical analysis results.
关键词
旅客联程出行 /
手机信令数据 /
出行方式识别 /
场站位置 /
场站数据取样最佳半径
Key words
intermodal passenger transportation /
mobile signaling data /
travel mode identification /
station location /
best radius of station data sampling
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]杨朗,周丽娜,张晓明. 基于手机信令数据的广州市职住空间特征及其发展模式探究[J]. 城市观察,2019(3):87-96.
[2]张天然. 基于手机信令数据的上海市域职住空间分析[J]. 城市交通,2016,14(1):15-23.
[3]冉斌. 手机数据在交通调查和交通规划中的应用[J]. 城市交通,2013,11(1):72-81,32.
[4]胡永恺,宋璐,张健,等. 基于手机信令数据的交通OD提取方法改进[J]. 交通信息与安全,2015(5):84-90,111.
[5]扈中伟,邓小勇,郭继孚,等. 基于手机定位数据的居民出行需求特征分析[C]// 第八届中国智能交通年会优秀论文集——轨道交通. 北京:中国智能交通协会,2013:889-897.
[6]杜翠凤,蒋仕宝. 基于移动信令数据的用户出行特征研究[J]. 移动通信,2015,39(23):9-12.
[7]詹益旺. 基于手机信令的道路交通状态识别及预测研究[D]. 广州:华南理工大学,2017.
[8]李耀辉. 基于移动信令数据的用户出行行为研究[D]. 重庆:重庆邮电大学,2017.
[9]包婷,章志刚,金澈清. 基于手机大数据的城市人口流动分析系统[J]. 华东师范大学学报(自然科学版),2015(5):162-171.
[10]宋少飞,李玮峰,杨东援. 基于移动通信数据的居民居住地识别方法研究[J]. 综合运输,2015,37(12):72-76.
[11]宋现敏,刘明鑫,马林,等. 基于极限学习机的公交行程时间预测方法[J]. 交通运输系统工程与信息,2018,18(5):136-142,150.
[12]Bloch A, Erdin R, Meyer S, et al. Battery-Efficient Transportation Mode Detection on Mobile Devices[C]// 2015 16th IEEE International Conference on Mobile Data Management. Pittsburgh: IEEE, 2015: 185-190.
[13]杜亚朋,雒江涛,程克非,等. 基于手机信令和导航数据的出行方式识别方法[J]. 计算机应用研究,2018,35(8):2311-2314.
[14]Anderson I, Muller H. Practical Activity Recognition Using GSM Data[C]// Proceedings of the 5th International Semantic Web Conference (ISWC). Athens: ISWC, 2006: 1-8.
[15]李振邦,应俊杰,顾承华,等. 基于数据挖掘的手机用户出行方式识别研究[J]. 黑龙江科技信息,2014(34):55-57.
[16]Wang H Y, Calabrese F, Di Lorenzo G, et al. Transportation Mode Inference from Anonymized and Aggregated Mobile Phone Call Detail Record[C]// Proceedings of 13th International IEEE Conference on Intelligent Transportation Systems(ITSC). Piscataway: IEEE, 2010: 318-323.
[17]李祖芬,于雷,高永,等. 基于手机信令定位数据的居民出行时空分布特征提取方法[J]. 交通运输研究,2016,2(1):51-57.
基金
中央级公益性科研院所基本科研业务费项目(20196109)