Skip to main content
SpringerLink
Log in

Region-based demand forecasting in bike-sharing systems using a multiple spatiotemporal fusion neural network

  • Data analytics and machine learning
  • Published:
Soft Computing Aims and scope Submit manuscript

Abstract

Bike-sharing systems (BSSs) have become increasingly popular around the globe and have attracted a wide range of research interests. In this paper, we study the region-based demand forecasting problem in BSSs. State-of-the-art methods usually employ branched residual 2D or 3D convolutional neural networks, in which each branch extracts one spatiotemporal dependence in three fragments: closeness, period, and trend. However, these methods ignore the correlations among the fragments. To address the challenge and extract the correlations, we propose a multiple spatiotemporal fusion network named MSTF-Net. It consists of multiple spatiotemporal layers: shared 3D convolutional network (3D-CNN) layers, eidetic 3D convolutional long short-term memory network (E3D-LSTM) layers, and fully connected (FC) layers. Specifically, the shared 3D-CNN layers highlight extracting low-level and short-term spatiotemporal dependence in each fragment. The E3D-LSTM layers extract long-term spatiotemporal dependence and correlations among the fragments. The FC layers extract nonlinear correlations of external factors such as weather, day-of-week, and time-of-day. MSTF-Net outperforms seven baseline models, including the branched residual 2D and 3D convolutional neural networks on station-free and station-based bike-sharing datasets. Code is available at https://github.com/yanxiao1930/MSTF-Net.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

Data Availability

The datasets analyzed during the current study are available in https://www.wunderground.com/, https://sodachallenges.com/datasets/mobike-shanghai/, and https://ride.citibikenyc.com/system-data.

Notes

  1. https://www.wunderground.com/.

  2. https://sodachallenges.com/datasets/mobike-shanghai.

  3. https://www.citibikenyc.com/system-data.

References

  • Ai Y, Li Z, Gan M et al (2019) A deep learning approach on short-term spatiotemporal distribution forecasting of dockless bike-sharing system. Neural Comput Appl 31(5):1665–1677

    Article  Google Scholar 

  • Ali A, Zhu Y, Zakarya M (2021) Exploiting dynamic spatio-temporal correlations for citywide traffic flow prediction using attention based neural networks. Inf Sci 577:852–870

    Article  MathSciNet  Google Scholar 

  • Alvarez-Valdes R, Belenguer JM, Benavent E et al (2016) Optimizing the level of service quality of a bike-sharing system. Omega 62:163–175

    Article  Google Scholar 

  • Ashqar HI, Elhenawy M, Almannaa MH, et al (2017) Modeling bike availability in a bike-sharing system using machine learning. In: 2017 5th IEEE international conference on models and technologies for intelligent transportation systems (MT-ITS), IEEE, pp 374–378

  • Caggiani L, Camporeale R, Ottomanelli M et al (2018) A modeling framework for the dynamic management of free-floating bike-sharing systems. Transp Res C: Emerg Technol 87:159–182

    Article  Google Scholar 

  • Chang PC, Wu JL, Xu Y et al (2019) Bike sharing demand prediction using artificial immune system and artificial neural network. Soft Comput 23(2):613–626

    Article  Google Scholar 

  • Chen C, Li K, Teo SG, et al (2018) Exploiting spatio-temporal correlations with multiple 3D convolutional neural networks for citywide vehicle flow prediction. In: Proceedings - IEEE international conference on data mining, ICDM 2018 (61625202) pp 893–898

  • Chen T, Guestrin C (2016) Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp 785–794

  • Chen Y, Zou X, Li K et al (2021) Multiple local 3d cNNs for region-based prediction in smart cities. Inf Sci 542:476–491

    Article  Google Scholar 

  • Dell’Amico M, Hadjicostantinou E, Iori M et al (2014) The bike sharing rebalancing problem: Mathematical formulations and benchmark instances. Omega 45:7–19

    Article  Google Scholar 

  • El-Assi W, Mahmoud MS, Habib KN (2017) Effects of built environment and weather on bike sharing demand: a station level analysis of commercial bike sharing in Toronto. Transportation 44(3):589–613

    Article  Google Scholar 

  • Giot R, Cherrier R (2014) Predicting bikeshare system usage up to one day ahead. In: 2014 IEEE symposium on computational intelligence in vehicles and transportation systems (civts), IEEE, pp 22–29

  • He P, Jiang G, Lam SK et al (2020) Learning heterogeneous traffic patterns for travel time prediction of bus journeys. Inf Sci 512:1394–1406

    Article  Google Scholar 

  • Jabbarpour MR, Zarrabi H, Khokhar RH et al (2018) Applications of computational intelligence in vehicle traffic congestion problem: a survey. Soft Comput 22(7):2299–2320

    Article  Google Scholar 

  • Ke J, Zheng H, Yang H et al (2017) Short-term forecasting of passenger demand under on-demand ride services: a spatio-temporal deep learning approach. Transp Res C: Emerg Technol 85(June):591–608

    Article  Google Scholar 

  • Li Y, Shuai B (2020) Origin and destination forecasting on dockless shared bicycle in a hybrid deep-learning algorithms. Multimed Tools Appl 79(7):5269–5280

    Article  Google Scholar 

  • Li Y, Zheng Y, Yang Q (2018) Dynamic bike reposition: a spatio-temporal reinforcement learning approach. Kdd. pp 1724–1733

  • Lloret I, Troyano JA, Enríquez F et al (2021) Two deep learning approaches to forecasting disaggregated freight flows: convolutional and encoder-decoder recurrent. Soft Comput 25(12):7769–7784

    Article  Google Scholar 

  • Loshchilov I, Hutter F (2017) Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101

  • Maneejuk P, Srichaikul W (2021) Forecasting foreign exchange markets: further evidence using machine learning models. Soft Comput 25(12):7887–7898

    Article  Google Scholar 

  • Masikos M, Demestichas K, Adamopoulou E et al (2015) Mesoscopic forecasting of vehicular consumption using neural networks. Soft Comput 19(1):145–156

    Article  Google Scholar 

  • Negahban A (2019) Simulation-based estimation of the real demand in bike-sharing systems in the presence of censoring. Eur J Oper Res 277(1):317–332

    Article  Google Scholar 

  • Rudloff C, Lackner B (2014) Modeling demand for bikesharing systems: neighboring stations as source for demand and reason for structural breaks. Transp Res Rec 2430(1):1–11

    Article  Google Scholar 

  • Schuijbroek J, Hampshire RC, Van Hoeve WJ (2017) Inventory rebalancing and vehicle routing in bike sharing systems. Eur J Oper Res 257(3):992–1004

    Article  MathSciNet  MATH  Google Scholar 

  • Singhvi D, Singhvi S, Frazier PI, et al (2015) Predicting bike usage for New York city’s bike sharing system. In: Workshops at the twenty-ninth AAAI conference on artificial intelligence

  • Tran D, Bourdev L, Fergus R, et al (2015) Learning spatiotemporal features with 3D convolutional networks. In: Proceedings of the IEEE international conference on computer vision 2015 Inter: 4489–4497

  • Wang Y, Jiang L, Yang Mh et al (2019) EIDETIC 3D LSTM: A MODEL FOR VIDEO PREDICTION AND BEYOND Yunbo. Iclr 19:1–14

    Google Scholar 

  • Xingjian S, Chen Z, Wang H, et al (2015) Convolutional LSTM network: a machine learning approach for precipitation nowcasting. In: Advances in neural information processing systems, pp 802–810

  • Xu C, Ji J, Liu P (2018) The station-free sharing bike demand forecasting with a deep learning approach and large-scale datasets. Transp Res C: Emerg Technol 95:47–60

    Article  Google Scholar 

  • Zhang J, Zheng Y, Qi D, et al (2016) DNN-based prediction model for spatio-temporal data. SIGSPATIAL’16 (In: International conference on advances in geographic information systems)

  • Zhang J, Zheng Y, Qi D (2017) Deep spatio-temporal residual networks for citywide crowd flows prediction. In: Thirty-First AAAI conference on artificial intelligence

  • Zhang S, Zhou L, Chen X et al (2020) Network-wide traffic speed forecasting: 3d convolutional neural network with ensemble empirical mode decomposition. Comput-Aided Civ Infrastruct Eng 35(10):1132–1147

    Article  Google Scholar 

  • Zhou Y, Li J, Chen H et al (2020) A spatiotemporal attention mechanism-based model for multi-step citywide passenger demand prediction. Inf Sci 513:372–385

    Article  Google Scholar 

  • Zhu J, Huang C, Yang M et al (2019) Context-based prediction for road traffic state using trajectory pattern mining and recurrent convolutional neural networks. Inf Sci 473:190–201

    Article  Google Scholar 

Download references

Funding

Gang Kou’s research is has been partially supported by grants from the National Natural Science Foundation of China (U1811462, 71725001 and 71910107002), the State Key R & D Program of China (2020YFC0832702), and the Major Project of the National Social Science Foundation of China (19ZDA092). Feng Xiao’s research has been partially supported by grants from the National Science Fund for Distinguished Young Scholars (72025104) and the National Natural Science Foundation of China (71861167001). Xianghua Gan’s research is supported by the National Natural Science Foundation of China (71771189 and 7217010132).

Author information

Authors and Affiliations

  1. School of Business Administration, Faculty of Business Administration, Southwestern University of Finance and Economics, Chengdu, China

    Xiao Yan, Gang Kou, Feng Xiao, Dapeng Zhang & Xianghua Gan

Authors
  1. Xiao Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gang Kou
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Feng Xiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Dapeng Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Xianghua Gan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Xianghua Gan.

Ethics declarations

Conflict of interest

The authors declare that they do not have any conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yan, X., Kou, G., Xiao, F. et al. Region-based demand forecasting in bike-sharing systems using a multiple spatiotemporal fusion neural network. Soft Comput 27, 4579–4592 (2023). https://doi.org/10.1007/s00500-022-07691-8

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00500-022-07691-8

Keywords

Search

Navigation