基于图卷积网络的乘客打车需求预测

doi:10.3969/j.issn.1007-7375.2022.05.012

工业工程 ›› 2022, Vol. 25 ›› Issue (5): 98-105.doi: 10.3969/j.issn.1007-7375.2022.05.012

基于图卷积网络的乘客打车需求预测

董成祥¹, 魏昕¹, 张坤鹏², 汪永超^1,3, 杨宇辉¹

1. 广东工业大学机电工程学院，广东广州，510006;
2. 河南工业大学电气工程学院，河南郑州，450001;
3. 广州番禺职业技术学院智能制造学院，广东广州，511483

收稿日期:2022-04-22 发布日期:2022-10-20
作者简介:董成祥(1989—)，男，河南省人，博士研究生，主要研究方向为智能交通系统和自动驾驶汽车
基金资助:
国家自然科学基金资助项目(62002101)

Passenger Ride-hailing Demand Prediction Based on Graph Convolutional Networks

DONG Chengxiang¹, WEI Xin¹, ZHANG Kunpeng², WANG Yongchao^1,3, YANG Yuhui¹

1. School of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China;
2. College of Electrical Engineering, Henan University of Technology, Zhengzhou 450001, China;
3. School of Intelligent Manufacturing, Guangzhou Panyu Polytechnic, Guangzhou 511483, China

Received:2022-04-22 Published:2022-10-20

摘要/Abstract

摘要： 乘客打车需求具有非线性和动态性的特点。为了提高乘客需求预测的准确性，需要充分考虑城市区域间的时空特性。针对城市中乘客的打车需求预测问题，利用图卷积网络 (graph convolutional network, GCN) 和长短期记忆单元 (long short-term memory, LSTM) 建立GCN-LSTM预测模型。在分析城市区域间时空特性的基础上，基于动态时间规整算法 (dynamic time warping, DTW)，将具有相似时空特性的区域重组并构建乘客打车需求图，利用图卷积网络提取需求图的空间特征；运用基于LSTM的编码器捕捉区域的时间特性；运用基于LSTM的解码器实现乘客打车需求的多区域同时、多时间步长预测。通过与传统模型的对比实验表明，本文提出的GCN-LSTM预测模型的均方根误差、平均绝对百分比误差和平均绝对误差最小，验证了所提出模型的预测准确性。

关键词: 图卷积网络, 长短期记忆单元, 时空特性, 乘客打车需求预测

Abstract: Predicting passenger demand is challenging due to nonlinear and dynamic demand patterns. The spatiotemporal dependencies among various zones need to be taken into consideration to improve the prediction accuracy. To address the problem of passenger demand prediction, a GCN-LSTM model is established by combining the graph convolutional network (GCN) and long short-term memory (LSTM). Based on the analysis of the spatiotemporal correlation among various urban areas, passenger demand maps are constructed with the help of the dynamic time warping (DTW) algorithm and the spatial dependencies of the demand map captured using GCN. The LSTM-based encoder is used to capture the temporal dependencies of the zones. The LSTM-based decoder is used to carry out the multi-step predictions of passenger demand for multiple zones simultaneously. The comparison experiments show that the proposed GCN-LSTM model outperforms the traditional models in terms of the root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE), which verifies the prediction accuracy of the proposed model.

Key words: graph convolutional networks, long short-term memory, spatiotemporal dependencies, passenger ride-hailing demand prediction

中图分类号:

U491

董成祥, 魏昕, 张坤鹏, 汪永超, 杨宇辉. 基于图卷积网络的乘客打车需求预测[J]. 工业工程, 2022, 25(5): 98-105.

DONG Chengxiang, WEI Xin, ZHANG Kunpeng, WANG Yongchao, YANG Yuhui. Passenger Ride-hailing Demand Prediction Based on Graph Convolutional Networks[J]. Industrial Engineering Journal, 2022, 25(5): 98-105.

参考文献

[1] 陈茜. 网络预约出租车智能调度系统的研究与设计 [D]. 北京: 北京化工大学, 2019.
CHEN Qian. Research and design of network vehicle intelligent dispatching system [D]. Beijing: Beijing University of Chemical Technology, 2019
[2] ZHANG K, FENG Z, CHEN S, et al. A framework for passengers demand prediction and recommendation[C]//Proceedings of the 2016 IEEE International Conference on Services Computing (SCC). San Fransisco: IEEE, 2016.
[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): 1393-402
[4] DAVIS N, RAINA G, JAGANNATHAN K. A multi-level clustering approach for forecasting taxi travel demand[C]//Proceedings of the 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC). Rio: IEEE, 2016.
[5] LI X, ZHANG Y, DU M, et al. The forecasting of passenger demand under hybrid ridesharing service modes: A combined model based on WT-FCBF-LSTM[J]. Sustainable Cities and Society, 2020, 62: 102419
[6] XU J, RAHMATIZADEH R, BöLöNI L, et al. Real-time prediction of taxi demand using recurrent neural networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 19(8): 2572-2581
[7] 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, 2017, 2634(1): 57-68
[8] VANICHRUJEE U, HORANONT T, PATTARA-ATIKOM W, et al. Taxi demand prediction using ensemble model based on RNNs and XGBOOST[C]//Proceedings of the 2018 International Conference on Embedded Systems and Intelligent Technology & International Conference on Information and Communication Technology for Embedded Systems (ICESIT-ICICTES).Khon Kaen: IEEE, 2018.
[9] YAN Z, LV S. Short-term forecast model of taxi demand based on time and space heterogeneity[J]. Journal of Intelligent & Fuzzy Systems, 2021, 41: 1-12
[10] AHMED M S, COOK A R. Analysis of freeway traffic time-series data by using Box-Jenkins techniques[J]. Transportation Research Record, 1979, 722: 1-9
[11] WILLIAMS R J, ZIPSER D. A learning algorithm for continually running fully recurrent neural networks[J]. Neural Computation, 1989, 1(2): 270-280
[12] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780
[13] ZHANG K, LIU Z, ZHENG L. Short-term prediction of passenger demand in multi-zone level: temporal convolutional neural network with multi-task learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(4): 1480-1490
[14] 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
[15] ZHAO L, SONG Y, ZHANG C, et al. T-gcn: A temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(9): 3848-3858
[16] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks [R/OL]. (2017-02-22). https://doi.org/10.48550/arXiv.1609.02907.
[17] TANG J, LIANG J, LIU F, et al. Multi-community passenger demand prediction at region level based on spatio-temporal graph convolutional network[J]. Transportation Research Part C:Emerging Technologies, 2021, 124: 102951
[18] BAI L, YAO L, WANG X, et al. Deep spatial temporal sequence modeling for multi-step passenger demand prediction[J]. Future Generation Computer Systems, 2021, 121: 25-34
[19] HU Q, MING L, XI R, et al. SOUP: A fleet management system for passenger demand prediction and competitive taxi supply[C/OL]//Proceedings of the 2021 IEEE 37th International Conference on Data Engineering (ICDE). (2021-04). http://dx.doi.org/10.1109/ICDE51399.2021.00297.
[20] VAPNIK V. Pattern recognition using generalized portrait method[J]. Automation and Remote Control, 1963, 24: 774-780

基于图卷积网络的乘客打车需求预测

Passenger Ride-hailing Demand Prediction Based on Graph Convolutional Networks

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	曹涛涛, 蒋阳升, 赵斌, 姚志洪, 谭宇. 考虑多交叉口相互影响的干道相位差仿真优化研究[J]. 工业工程, 2018, 21(6): 40-45.
[2]	傅惠, 陈恺宇. 基于工作流网的应急资源配置与路径规划集成优化[J]. 工业工程, 2018, 21(5): 1-8.
[3]	王逸, 姚志洪, 蒋阳升, 赵斌, 谭宇. 基于自适应遗传算法的双环信号配时优化模型[J]. 工业工程, 2018, 21(5): 72-80.
[4]	罗孝羚, 蒋阳升, 吴奇, 姚志洪, 李思怡. 基于IC卡识别通勤客流的公交线路多模式调度[J]. 工业工程, 2018, 21(5): 81-86.
[5]	李彦瑾, 罗霞, 王莹. 考虑多节点拥堵的城市道路网级联失效仿真[J]. 工业工程, 2018, 21(4): 1-7.
[6]	姚志洪, 蒋阳升, 王逸. 车路协同环境下自适应信号配时优化模型[J]. 工业工程, 2018, 21(4): 8-14,33.
[7]	罗孝羚, 蒋阳升. 中小城市公交线网及发车频率同步优化[J]. 工业工程, 2017, 20(4): 25-30.
[8]	傅惠, 王叶飞, 陈赛飞. 面向宏观基本图的多模式交通路网分区算法[J]. 工业工程, 2020, 23(1): 1-9.
[9]	李婷婷. 有能力限制的轴辐网络关键枢纽识别模型[J]. 工业工程, 2020, 23(2): 100-104,123.
[10]	李妍峰, 龚桂琳. 基于用户异质性的旅游巴士定价研究[J]. 工业工程, 2020, 23(4): 69-74,130.
[11]	廖晔, 王顺意. 基于改进网络最大流的道路通行能力优化研究[J]. 工业工程, 2020, 23(5): 96-102.
[12]	曹振宇, 蒋阳升, 罗孝羚, 杨洁, 张姚. 中小城市直达公交网络优化[J]. 工业工程, 2020, 23(6): 117-123.
[13]	王宁, 张玉萍, 赵姣, 金子阳. 考虑碳排放及时间窗的LIRP模型与算法研究[J]. 工业工程, 2021, 24(2): 34-42.
[14]	陈红梅, 张远航, 张春玲. 动态视角下城市通勤路径优化研究[J]. 工业工程, 2022, 25(1): 64-74.
[15]	徐韬, 祝烨, 谢晓忠, 晏秋萍, 程龙春. 基于全过程的偶发性拥堵消散时间预测模型[J]. 工业工程, 2022, 25(3): 157-163.