A spatio-temporal deep learning model for short-term bike-sharing demand prediction

Ruo Jia; Richard Chamoun; Alexander Wallenbring; Masoomeh Advand; Shanchuan Yu; Yang Liu; Kun Gao; Ruo Jia; Richard Chamoun; Alexander Wallenbring; Masoomeh Advand; Shanchuan Yu; Yang Liu; Kun Gao

doi:10.3934/era.2023051

Electronic Research Archive

2023, Volume 31, Issue 2: 1031-1047. doi: 10.3934/era.2023051

Previous Article Next Article

Research article Special Issues

A spatio-temporal deep learning model for short-term bike-sharing demand prediction

1.
Department of Architecture and Civil Engineering, Chalmers University of Technology, Gothenburg, Sweden
2.
Faculty of Electrical, Computer and IT Engineering, Qazvin Islamic Azad University, Qazvin, Iran
3.
Research and Development Center of Transport Industry of Self-driving Technology, China Merchants Chongqing Communications Research & Design Institute Co. Ltd., Chongqing, China

Received: 10 September 2022 Revised: 23 November 2022 Accepted: 27 November 2022 Published: 12 December 2022

Bike-sharing systems are widely operated in many cities as green transportation means to solve the last mile problem and reduce traffic congestion. One of the critical challenges in operating high-quality bike-sharing systems is rebalancing bike stations from being full or empty. However, the complex characteristics of spatiotemporal dependency on usage demand may lead to difficulties for traditional statistical models in dealing with this complex relationship. To address this issue, we propose a graph-based neural network model to learn the representation of bike-sharing demand spatial-temporal graph. The model has the ability to use graph-structured data and takes both spatial- and temporal aspects into consideration. A case study about bike-sharing systems in Nanjing, a large city in China, is conducted based on the proposed method. The results show that the algorithm can predict short-term bike demand with relatively high accuracy and low computing time. The predicted errors for the hourly station level usage demand prediction are often within 20 bikes. The results provide helpful tools for short-term usage demand prediction of bike-sharing systems and other similar shared mobility systems.
- demand forecast,
- artificial intelligence,
- deep learning,
- graph neural network
Citation: Ruo Jia, Richard Chamoun, Alexander Wallenbring, Masoomeh Advand, Shanchuan Yu, Yang Liu, Kun Gao. A spatio-temporal deep learning model for short-term bike-sharing demand prediction[J]. Electronic Research Archive, 2023, 31(2): 1031-1047. doi: 10.3934/era.2023051

Related Papers:

Abstract

Bike-sharing systems are widely operated in many cities as green transportation means to solve the last mile problem and reduce traffic congestion. One of the critical challenges in operating high-quality bike-sharing systems is rebalancing bike stations from being full or empty. However, the complex characteristics of spatiotemporal dependency on usage demand may lead to difficulties for traditional statistical models in dealing with this complex relationship. To address this issue, we propose a graph-based neural network model to learn the representation of bike-sharing demand spatial-temporal graph. The model has the ability to use graph-structured data and takes both spatial- and temporal aspects into consideration. A case study about bike-sharing systems in Nanjing, a large city in China, is conducted based on the proposed method. The results show that the algorithm can predict short-term bike demand with relatively high accuracy and low computing time. The predicted errors for the hourly station level usage demand prediction are often within 20 bikes. The results provide helpful tools for short-term usage demand prediction of bike-sharing systems and other similar shared mobility systems.

References

[1]	K. Gao, Y. Yang, X. Qu, Diverging effects of subjective prospect values of uncertain time and money, Commun. Transp. Res., 1 (2021), 100007. https://doi.org/10.1016/j.commtr.2021.100007 doi: 10.1016/j.commtr.2021.100007
[2]	J. Ke, X. Qin, H. Yang, Z. Zheng, Z. Zhu, J. Ye, Predicting origin-destination ride-sourcing demand with a spatio-temporal encoder-decoder residual multi-graph convolutional network, Transp. Res. Part C Emerging Technol., 122 (2021), 102858. https://doi.org/10.1016/j.trc.2020.102858 doi: 10.1016/j.trc.2020.102858
[3]	J. G. Jin, H. Nieto, L. Lu, Robust bike-sharing stations allocation and path network design: a two-stage stochastic programming model, Transp. Lett., 12 (2020), 682–691. https://doi.org/10.1080/19427867.2019.1691299 doi: 10.1080/19427867.2019.1691299
[4]	H. I. Ashqar, M. Elhenawy, H. A. Rakha, M. Almannaa, L. House, Network and station-level bike-sharing system prediction: a San Francisco bay area case study, J. Intell. Transp. Syst., 26 (2022), 602–612. https://doi.org/10.1080/15472450.2021.1948412 doi: 10.1080/15472450.2021.1948412
[5]	G. E. Box, D. A. Pierce, Distribution of residual autocorrelations in autoregressive-integrated moving average time series models, J. Am. Stat. Assoc., 65 (1970), 1509–1526. https://doi.org/10.1080/01621459.1970.10481180 doi: 10.1080/01621459.1970.10481180
[6]	H. Drucker, C. J. Burges, L. Kaufman, A. Smola, V. Vapnik, Support vector regression machines, Adv. Neural Inf. Process. Syst., 9 (1996), 155–161.
[7]	K. Davoian, W. M. Lippe, Including phenotype information in mutation to evolve artificial neural networks, in International Joint Conference on Neural Networks, 2007, 2782–2787. https://doi.org/10.1109/IJCNN.2007.4371400
[8]	Y. Wang, B. Chaib-Draa, A KNN based Kalman filter Gaussian process regression, in Twenty-Third International Joint Conference on Artificial Intelligence, 2013, 1771–1777. https://dl.acm.org/doi/10.5555/2540128.2540382
[9]	U. Johansson, H. Boström, T. Löfström, H. Linusson, Regression conformal prediction with random forests, Mach. Learn., 97 (2014), 155–176. https://doi.org/10.1007/s10994-014-5453-0 doi: 10.1007/s10994-014-5453-0
[10]	X. Li, R. Bai, Freight vehicle travel time prediction using gradient boosting regression tree, in 2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA), 2016, 1010–1015. https://doi.org/10.1109/ICMLA.2016.0182
[11]	D. E. Rumelhart, G. E. Hinton, R. J. Williams, Learning representations by back-propagating errors, Nature, 323 (1986), 533–536. https://doi.org/10.1038/323533a0 doi: 10.1038/323533a0
[12]	K. Cho, B. Van Merriënboer, D. Bahdanau, Y. Bengio, On the properties of neural machine translation: Encoder-decoder approaches, arXiv preprint, 2014, arXiv: 1409.1259. https://doi.org/10.48550/arXiv.1409.1259
[13]	Y. Liang, S. Ke, J. Zhang, X. Yi, Y. Zheng, Geoman: Multi-level attention networks for geo-sensory time series prediction, in Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18), 1 (2018), 3428–3434. https://www.ijcai.org/proceedings/2018/0476.pdf.
[14]	Y. Li, Z. Zhu, D. Kong, M. Xu, Y. Zhao, Learning heterogeneous spatial-temporal representation for bike-sharing demand prediction, in Proceedings of the AAAI Conference on Artificial Intelligence, 33 (2019), 1004–1011. https://doi.org/10.1609/aaai.v33i01.33011004
[15]	C. Guo, F. Berkhahn, Entity embeddings of categorical variables, arXiv preprint, 2022, arXiv: 1604.06737.
[16]	Y. Yan, Y. Tao, J. Xu, S. Ren, H. Lin, Visual analytics of bike-sharing data based on tensor factorization, J. Visualization, 21 (2018), 495–509. https://doi.org/10.1007/s12650-017-0463-1 doi: 10.1007/s12650-017-0463-1
[17]	Y. Guo, J. A. Kelly, J. P. Clinch, Variability in total cost of vehicle ownership across vehicle and user profiles, Commun. Transp. Res., 2 (2022), 100071. https://doi.org/10.1016/j.commtr.2022.100071 doi: 10.1016/j.commtr.2022.100071
[18]	K. Gao, H. Wang, S. Wang, X. Qu, Data and code disclosure and sharing policy of communications in transportation research, Commun. Transp. Res., 2 (2022), 100055. https://doi.org/10.1016/j.commtr.2022.100055 doi: 10.1016/j.commtr.2022.100055
[19]	L. Lin, Z. He, S. Peeta, Predicting station-level hourly demand in a large-scale bike-sharing network: A graph convolutional neural network approach, Transp. Res. Part C Emerging Technol., 97 (2018), 258–276. https://doi.org/10.1016/j.trc.2018.10.011 doi: 10.1016/j.trc.2018.10.011
[20]	P. Goyal, N. Kamra, X. He, Y. Liu, Dyngem: Deep embedding method for dynamic graphs, arXiv preprint, 2018, arXiv: 1805.11273. https://doi.org/10.48550/arXiv.1805.11273
[21]	B. Yu, H. Yin, Z. Zhu, Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting, arXiv preprint, 2018, arXiv: 1709.04875. https://doi.org/10.48550/arXiv.1709.04875
[22]	K. Gao, Y. Yang, X. Qu, Examining nonlinear and interaction effects of multiple determinants on airline travel satisfaction, Transp. Res. Part D Transp. Environ., 97 (2021), 102957. https://doi.org/10.1016/j.trd.2021.102957 doi: 10.1016/j.trd.2021.102957
[23]	K. Gao, Y. Yang, A. Li, X. Qu, Spatial heterogeneity in distance decay of using bike sharing: An empirical large-scale analysis in Shanghai, Transp. Res. Part D Transp. Environ., 94 (2021), 102814. https://doi.org/10.1016/j.trd.2021.102814 doi: 10.1016/j.trd.2021.102814

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)