Prediction of Demand for Shared Bicycles Based on Machine Learning

Wenyu Jiang

doi:10.54097/3ctc2b08

Authors

Wenyu Jiang

DOI:

https://doi.org/10.54097/3ctc2b08

Keywords:

shared bicycles; demand forecasting; machine learning; feature engineering; XGBoost; LightGBM; time series analysis.

Abstract

With the acceleration of urbanization, shared bicycles have become an indispensable component of modern urban transportation systems, playing an important role in improving social resource utilization, alleviating traffic pressure, and promoting green travel. However, a common problem of supply-demand imbalance exists in the spatial and temporal distribution of vehicles, which not only affects user experience but also increases operating costs. In this context, accurate prediction of shared bicycle rental demand is crucial for achieving refined operations. This paper presents a high-precision prediction model constructed using machine learning technology. The proposed methodology, structured into three distinct modules, begins with detailed exploratory data analysis and comprehensive feature engineering. This includes logarithmic transformation of target variables, encoding of categorical features, and critically, the construction of key interaction features to capture the complex relationship between variables like temperature and hour. On this basis, the XGBoost model is employed to evaluate feature importance and select an optimal feature subset. Subsequently, through a series of comparative experiments, a range of machine learning regression models, including linear, tree-based, and gradient boosting models, are systematically compared. Using a time-series cross-validation method, each model is trained and evaluated with Root Mean Square Logarithmic Error (RMSLE) and the coefficient of determination (R²) as the main evaluation indicators. Finally, hyperparameters for the top-performing models are optimized to further improve prediction accuracy and generalization ability. The results indicate that the tree-based ensemble model LightGBM exhibits superior performance in the task of predicting shared bicycle demand, with its accuracy further improved after hyperparameter optimization. This study not only provides an effective demand forecasting solution for shared bicycle operators, but also offers a valuable reference for similar time series forecasting problems.

Downloads

Download data is not yet available.

References

[1] S. H. Choi and M. K. Han, “The empirical evaluation of models predicting bike sharing demand,” in 2020 International Conference on Information and Communication Technology Convergence (ICTC), 2020, pp. 1560–1562.

[2] M. M. Isalm, M. E. Biswas, M. Shahzamal, M. D. Haque, and M. S. Hos- sain, “An effective data driven approach to predict bike rental demand,” in 2023 5th International Conference on Sustainable Technologies for Industry 5.0 (STI), 2023, pp. 1–5.

[3] M. J. A. Shanto, R. Akter, D. S. Kim, and T. Jun, “Predicting bike- sharing demand: A machine learning approach for urban mobility analysis,” in 2023 14th International Conference on Information and Communication Technology Convergence (ICTC), 2023, pp. 1079–1081.

[4] A. Kealy and J. Wu, “Safety challenges and solutions in bike-sharing systems,” in 2021 IEEE 18th International Conference on Mobile Ad Hoc and Smart Systems (MASS), 2021, pp. 651–656.

[5] A. S. Patel, M. Ojha, M. Rani, A. Khare, O. P. Vyas, and R. Vyas, “Ontology-based multi-agent smart bike sharing system (sbss),” in 2018 IEEE International Conference on Smart Computing (SMARTCOMP), 2018, pp. 417–422.

[6] A. Jaber, B. Csonka, and J. Juha´sz, “Long term time series prediction of bike sharing trips: A cast study of budapest city,” in 2022 Smart City Symposium Prague (SCSP), 2022, pp. 1–5.

[7] X. Han, P. Wang, J. Gao, M. Shah, R. K. M. Ambulgekar, A. Jarandikar, and S. Dhar, “Bike sharing data analytics for silicon valley in usa,” in 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced Trusted Computed, Scalable Computing Communications, Cloud Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 2017, pp. 1–9.

[8] Z. Jiao, H. Jin, B. Liu, and Z. Zhang, “Short-term demand forecasting of shared bicycles driven by big data: A comparative analysis of machine learning models,” Journal of Business Economics, vol. 322, no. 8, pp. 16–25, 2018, [In Chinese].

[9] H. Yang, X. Zhang, L. Zhong, S. Li, X. Zhang, and J. Hu, “Short-term demand forecasting for bike sharing system based on machine learning,” in 2019 5th International Conference on Transportation Information and Safety (ICTIS), 2019, pp. 1295–1300.

[10] M. H. Almannaa, M. Elhenawy, F. Guo, and H. A. Rakha, “Incremental learning models of bike counts at bike sharing systems,” in 2018 21st International Conference on Intelligent Transportation Systems (ITSC), 2018, pp. 3712–3717.

[11] D. Cao, S. Fan, Y. Zhang, and K. Xia, “Short-term demand forecasting of shared bicycles based on long short-term memory neural network model,” Science Technology and Engineering, vol. 20, no. 20, pp. 8344–8349, 2020, [In Chinese].

[12] X. Xing, Z. Yin, and L. Fang, “Shared bicycle demand prediction incorporating multivariate meteorological factors,” Intelligent Computer and Applications, vol. 15, no. 1, pp. 178–186, 2025, [In Chinese].

[13] F. Huang, S. Qiao, J. Peng, and B. Guo, “A bimodal gaussian inhomogeneous poisson algorithm for bike number prediction in a bike- sharing system,” IEEE Transactions on Intelligent Transportation Sys- tems, vol. 20, no. 8, pp. 2848–2857, 2019.

[14] Y. Li and Y. Zheng, “Citywide bike usage prediction in a bike sharing system,” IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 6, pp. 1079–1091, 2020.

[15] S. Feng, H. Chen, C. Du, J. Li, and N. Jing, “A hierarchical demand prediction method with station clustering for bike sharing system,” in 2018 IEEE Third International Conference on Data Science in Cyberspace (DSC), 2018, pp. 829–836.

[16] M. He, X. Xue, X. Zhang, and C. Zhou, “A bike-sharing demand predicting model with integrating temporal convolutional network and self-attention,” in 2021 International Conference on Electronic Information Engineering and Computer Science (EIECS), 2021, pp. 278–281.

[17] H. Yang, S. M. Raza, D. T. Le, D. S. Kim, and H. Choo, “Dual- branch neural networks for predicting shared bikes,” in 2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM), 2023, pp. 1–4.

[18] J. Gu, Q. Zhou, J. Yang, Y. Liu, F. Zhuang, Y. Zhao, and H. Xiong, “Exploiting interpretable patterns for flow prediction in dockless bike sharing systems,” IEEE Transactions on Knowledge and Data Engineering, vol. 34, no. 2, pp. 640–652, 2022.

[19] J. Huang, X. Wang, and H. Sun, “Central station based demand prediction in a bike sharing system,” in 2019 20th IEEE International Conference on Mobile Data Management (MDM), 2019, pp. 346–348.

[20] W. Zheng, H. Deng, and F. Han, “Gst-net: A gis-based hybrid prediction model for shared bike traffic flow,” in 2021 IEEE 21st International Conference on Software Quality, Reliability and Security Companion (QRS-C), 2021, pp. 941–946.

[21] E. Collini, P. Nesi, and G. Pantaleo, “Deep learning for short-term prediction of available bikes on bike-sharing stations,” IEEE Access, vol. 9, pp. 124337–124347, 2021.

[22] Y. Du, B. Xiao, W. Xu, D. Cui, Q. Xu, and L. Yan, “Destination prediction for sharing-bikes’ trips,” in 2018 International Conference on Network Infrastructure and Digital Content (IC-NIDC), 2018, pp. 198–202.

[23] A. K. Das, A. M. Joshi, and S. Dhal, “A machine learning based bike recommendation system catering to user’s travel needs,” in 2020 IEEE 17th India Council International Conference (INDICON), 2020, pp. 1–6.

[24] M. Jiang, C. Li, K. Li, Z. Yang, and H. Liu, “Interblock flow prediction with relation graph network for cold start on bike-sharing system,” IEEE Internet of Things Journal, vol. 9, no. 15, pp. 13390–13404, 2022.

[25] R. Guo, Z. Jiang, J. Huang, J. Tao, C. Wang, J. Li, and L. Chen, “Bikenet: Accurate bike demand prediction using graph neural net- works for station rebalancing,” in 2019 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Comput- ing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (Smart- World/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 2019, pp. 686–693.

[26] R. Qin, L. Kong, M. Guo, B. Yao, and M. Guizani, “Rebalance modern bike sharing system: Spatio-temporal data prediction and path planning for multiple carriers,” in 2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS), 2018, pp. 1081–1086.

[27] J. Chai, J. Song, H. Fan, Y. Xu, L. Zhang, B. Guo, and Y. Xu, “St-bikes: Predicting travel-behaviors of sharing-bikes exploiting urban big data,” IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 7, pp. 7676–7686, 2023.

[28] Y. Liang, G. Huang, and Z. Zhao, “Cross-mode knowledge adaptation for bike sharing demand prediction using domain-adversarial graph neural networks,” IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 5, pp. 3642–3653, 2024.

[29] M. Tabandeh, C. Antoniou, and G. Cantelmo, “Long-term & short-term bike sharing demand predictions using contextual data,” in 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), 2023, pp. 1–6.

[30] Y. Zhou and Y. Huang, “Place representation based bike demand prediction,” in 2019 IEEE International Conference on Big Data (Big Data), 2019, pp. 1577–1586.

[31] S. S. Chawathe, “Mining bike-share data,” in 2020 IEEE International Smart Cities Conference (ISC2), 2020, pp. 1–8.

[32] M. Bencekri, A. Founoun, A. Haqiq, and A. Hayar, “Investigation of shared-bike demand using data analytics,” in 2022 IEEE International Smart Cities Conference (ISC2), 2022, pp. 1–4.

[33] A. A. Ramesh, S. P. Nagisetti, N. Sridhar, K. Avery, and D. Bein, “Station-level demand prediction for bike-sharing system,” in 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC), 2021, pp. 0916–0921.