Research on Bus Passenger Flow Dynamic Prediction and Route Optimization Strategy Based on Spatiotemporal Clustering and ARIMA Model
DOI:
https://doi.org/10.53469/wjimt.2025.08(04).02Keywords:
Spatio-temporal clustering, ARIMA model, Bus passenger flow forecast, Route optimization strategyAbstract
With the acceleration of urbanization, urban public transportation system is an indispensable part of urban transportation, and its operation efficiency and service quality are directly related to citizens' daily commuting experience. Accurate prediction of bus passenger flow is the premise of making public transportation planning, implementing operation management and optimizing route layout, and it is crucial to improve the overall efficiency and service quality of public transportation system. However, due to the complex characteristics of bus passenger flow data in time and space, it is often difficult for traditional forecasting methods to accurately grasp its dynamic change trend. In view of this, this paper designs a dynamic prediction method of bus passenger flow that integrates spatiotemporal clustering technology and ARIMA model, and combines route optimization strategy to improve the accuracy of bus passenger flow prediction and the scientific purpose of route optimization decision.
References
Zhang, Y., Cheng, X., & Liu, Y. (2021). A spatiotemporal graph convolutional network for citywide crowd flows prediction in transportation systems. Transportation Research Part C: Emerging Technologies, 132, 103410.
Liu, L., Chen, J., Wu, H., Zhen, J., Li, G., & Lin, L. (2020). Physical-virtual collaboration modeling for urban crowd flow prediction. IEEE Transactions on Intelligent Transportation Systems, 22(6), 3521-3534.
Chen, D., Zhang, J., Jiang, S., & Wang, Y. (2022). Hybrid deep learning model with spatial-temporal attention for metro passenger flow prediction. Sustainable Cities and Society, 76, 103445.
Zhou, Y., Chen, X., & Zhang, Q. (2020). A multi-objective optimization model for bus route planning considering temporal-spatial equity. Transportation Research Part A: Policy and Practice, 142, 1-18.
Chen, L., Li, T., & Wang, Z. (2021). Dynamic bus scheduling using deep reinforcement learning with passenger flow prediction. IEEE Transactions on Intelligent Transportation Systems, 23(5), 4872-4884.
Liu, X., Zhang, H., & Wang, Y. (2022). Spatiotemporal patterns of metro passenger flow: A case study of Beijing subway system. Cities, 131, 103998.
Wang, F., Li, D., & Liu, J. (2021). A graph neural network approach for spatial dependency modeling in public transport demand prediction. Sustainable Cities and Society, 75, 103326.
Zheng, Y., Capra, L., & Wolfson, O. (2020). Urban computing with spatiotemporal clustering: A case study of bike sharing systems. ACM Transactions on Intelligent Systems and Technology, 11(3), 1-25.
Li, M., Chen, Y., & Zhang, W. (2023). Hybrid clustering algorithm for temporal public transport data: Integrating DTW and spectral clustering. Expert Systems with Applications, 213, 119207.
Xu, J., Rahimi, E., & Chow, J. Y. (2022). Evaluation metrics for spatiotemporal clustering in transportation systems: A comparative analysis. Transportmetrica B: Transport Dynamics, 10(1), 1-22.