基于深度学习的网约车供需缺口短时预测研究

交通运输系统工程与信息 ›› 2019, Vol. 19 ›› Issue (2): 223-230.

基于深度学习的网约车供需缺口短时预测研究

谷远利¹，李萌¹，芮小平^{* 2}，陆文琦¹，王硕¹

1. 北京交通大学综合交通运输大数据应用技术交通运输行业重点实验室, 北京 100044； 2. 河海大学地球科学与工程学院，南京 211100

收稿日期:2018-10-08 修回日期:2018-11-25 出版日期:2019-04-25 发布日期:2019-04-25
作者简介:谷远利(1973-)，男，辽宁海城人，副教授，博士.
基金资助:
国家自然科学基金/National Natural Science Foundation of China(41771478).

Short-term Forecasting of Supply-demand Gap under Online Car-hailing Services Based on Deep Learning

GU Yuan-li¹, LI Meng¹, RUI Xiao-ping², LU Wen-qi¹, WANG Shuo¹

1. Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport， Ministry of Transport，Beijing Jiaotong University, Beijing 100044, China; 2. School of Earth Sciences and Engineering, Hohai University, Nanjing 211000, China

Received:2018-10-08 Revised:2018-11-25 Online:2019-04-25 Published:2019-04-25

摘要/Abstract

摘要：

城市不同区域网约车供需缺口预测可为车辆调度策略提供支持，从而提高车辆运行效率和乘客服务水平.为实现网约车供需缺口短时预测，提出一种基于时空数据挖掘的深度学习预测模型(Spatio-Temporal Deep Learning Model, S-TDL).该模型由时空变量模型、空间属性变量模型和环境变量模型 3个子模型融合而成，可捕捉时空关联性、区域差异性和环境变化对供需缺口的影响.同时，提出特征聚类—最大信息系数两阶段特征选择方法，筛选与供需缺口相关性强的特征变量，提高训练效率，减少过拟合.滴滴出行实例分析证明，特征选择后的 STDL模型预测精度显著优于BP神经网络、长短期记忆网络和卷积神经网络.

关键词: 城市交通, 供需缺口预测, 深度学习, 网约车, 时空关联性

Abstract:

The results of supply- demand gap prediction for online car- hailing services in different areas can provide support for online car-hailing scheduling system, thereby improving efficiency and service levels. In order to realize the short-term forecast of supply-demand gap for online car-hailing services, this paper proposes a novel spatio- temporal deep learning model (S- TDL). The model is composed of three sub- models: spatiotemporal variable model, spatial attribute variable model and environment variable model. It can capture the impact of spatio- temporal correlation, regional difference and environmental change on supply- demand gap. Moreover, a feature selection method named feature clustering-maximum information coefficient two-stage feature selection is proposed to screen out the important features which are strongly correlated with the supply- demand gap, improve training efficiency. The experimental results show that the S-TDL model after feature selection achieves the better performance than the existing methods.

Key words: urban traffic, supply- demand gap forecasting, deep learning, online car- hailing, spatio- temporal correlation

中图分类号:

U491

谷远利，李萌，芮小平，陆文琦，王硕. 基于深度学习的网约车供需缺口短时预测研究[J]. 交通运输系统工程与信息, 2019, 19(2): 223-230.

GU Yuan-li, LI Meng, RUI Xiao-ping, LU Wen-qi, WANG Shuo. Short-term Forecasting of Supply-demand Gap under Online Car-hailing Services Based on Deep Learning[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(2): 223-230.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract19823.shtml

http://www.tseit.org.cn/CN/Y2019/V19/I2/223

参考文献

[1] KE J, ZHENG H, YANG H, et al. Short-term forecasting of passenger demand under on- demand ride services: A spatio- temporal deep learning approach[J]. Transportation Research Part C: Emerging Technologies, 2017(85): 591-608.

[2] JAMIL M S, AKBAR S. Taxi passenger hotspot prediction using automatic ARIMA model[C]//Science in Information Technology (ICSITech), 2017 3rd International Conference on. IEEE, 2017.

[3] MOREIRA-MATIAS L, GAMA J, FERREIRA M, et al. Predicting taxi-passenger demand using streaming data[J]. IEEE Transactions on Intelligent Transportation Systems, 2013, 14(3): 1393-1402.

[4] LIU J, CUI E, HU H, et al. Short- term forecasting of emerging on- demand ride services[C]//Transportation Information and Safety (ICTIS), 2017 4th International Conference on. IEEE, 2017.

[5] LI Y, LU J, ZHANG L, et al. Taxi booking mobile app order demand prediction based on short- term traffic forecasting[J]. Transportation Research Record: Journal of the Transportation Research Board, 2017 (2634): 57- 68.

[6] XU J, RAHMATIZADEH R, BÖLÖNI L, et al. A sequence learning model with recurrent neural networks for taxi demand prediction[C]//Local Computer Networks (LCN), 2017 IEEE 42nd Conference on. IEEE, 2017.

[7] 段宗涛, 张凯, 杨云, 等. 基于深度 CNN-LSTM-ResNet 组合模型的出租车需求预测[J]. 交通运输系统工程与信息, 2018, 18(4): 215-223. [DUAN Z T, ZHANG K, YANG Y, et al. Taxi demand prediction based on CNNLSTM- ResNet hybrid depth learning model[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18(4): 215-223.]

[8] ZHANG X, WANG X, CHEN W, et al. A Taxi gap prediction method via double ensemble gradient boosting decision tree[C]// IEEE, International Conference on Big Data Security on Cloud. IEEE, 2017.

[9] LI J, WANG Z. Online car- hailing dispatch: Deep supply- demand gap forecast on spark[C]// IEEE, International Conference on Big Data Analysis. IEEE, 2017.

[10] WANG D, CAO W, LI J, et al. DeepSD: supply-demand prediction for online car- hailing services using deep neural networks[C]//2017 IEEE 33rd International Conference on Data Engineering (ICDE). IEEE, 2017.

[11] SHI X, CHEN Z, WANG H, et al. Convolutional LSTM network: A machine learning approach for precipitation nowcasting[C]//Advances in Neural Information Processing Systems, 2015.

[12] 国家质量监督检验检疫总局. GB3095-2012 环境空气质量标准 [S]. 北京：中国环境科学出版社，2012. [General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. GB3095- 2012 Ambient Air Quality Standard [S]. Beijing: China Environmental Science Press, 2012.]

[1]	刘晓冰, 李奉孝, 田欣妹, 闫学东. 基于通勤模式的都市圈中心结构判别研究[J]. 交通运输系统工程与信息, 2022, 22(2): 17-28.
[2]	许奇, 陈越, 黄靖茹, 高顺祥, 张志健. 考虑出行费用的就业可达性分析[J]. 交通运输系统工程与信息, 2022, 22(2): 37-44.
[3]	王健, 王慧, 胡晓伟, 李园园. 高峰期考虑乘客议价的网约车定价与平台收益及社会福利优化[J]. 交通运输系统工程与信息, 2022, 22(2): 54-63.
[4]	刘春禹, 刘永红, 罗霞, 朱颖. 混合交通流环境下单交叉口自动驾驶车辆轨迹优化[J]. 交通运输系统工程与信息, 2022, 22(2): 154-162.
[5]	王景鹏, 祝蕊, 王鹏飞. 需求不确定下的网约车司机最优工资水平研究[J]. 交通运输系统工程与信息, 2022, 22(2): 178-185.
[6]	张小雨, 邵春福, 王博彬, 黄士琛. 新冠疫情影响下居民共享出行方式选择行为研究[J]. 交通运输系统工程与信息, 2022, 22(2): 186-196.
[7]	李欣, 戴章, 李怀悦, 胡笳. 基于连续近似模型的轨道交通与常规公交耦合优化设计[J]. 交通运输系统工程与信息, 2022, 22(2): 206-213.
[8]	郝妍熙, 刘戎阳, 胡华, 方勇, 刘志钢. 单向通道行人-自行车混合流建模及自组织现象研究[J]. 交通运输系统工程与信息, 2022, 22(2): 223-229.
[9]	赵传林, 贺少松, 孙淑敏, 王钰涵. 基于理性疏忽理论的交通疏散网络双层优化模型[J]. 交通运输系统工程与信息, 2022, 22(2): 239-246.
[10]	陈小红, 蓝秋雨. 不确定交通需求下交叉口信号配时区间优化模型[J]. 交通运输系统工程与信息, 2022, 22(2): 247-256.
[11]	李冰, 王正辉, 马铭炜, 杨鸿宇, 冯悦. 信号交叉口行人二次过街下右转车通行能力与延误估计模型[J]. 交通运输系统工程与信息, 2022, 22(2): 257-267.
[12]	朱彤, 秦丹, 董傲然, 杜雨萌, 魏雯. 公交驾驶员违规类型同交通事故间的关联分析[J]. 交通运输系统工程与信息, 2022, 22(2): 322-329.
[13]	毛保华, 王敏, 何天健, 陈海波. 城市公共交通服务水平研究回顾和展望[J]. 交通运输系统工程与信息, 2022, 22(1): 2-13.
[14]	王子甲, 贾慧慧, 朱亚迪, 陈峰. 基于智能卡数据的轨道与公交复合网络通勤方式选择行为研究[J]. 交通运输系统工程与信息, 2022, 22(1): 67-73.
[15]	贾斌, 朱凌, 李树凯, 刘家林. 城市轨道交通小交路计划与客流控制协同优化策略[J]. 交通运输系统工程与信息, 2022, 22(1): 124-132.