Adaptive Feature Fusion Networks for Origin Destination Passenger Flow Prediction in Metro Systems
DOI:
https://doi.org/10.69996/fmep.2025031Keywords:
Origin-Destination prediction, Metro systems, Passenger flow forecasting, Adaptive feature fusion networks (AFFN), Spatial-Temporal ModellingAbstract
Accurately predicting Origin-Destination (OD) passenger flow can help metro service quality and efficiency. Existing works have focused on predicting incoming and outgoing flows for individual stations, while little attention was paid to OD prediction in metro systems. The challenges are that OD flows 1) have high temporal dynamics and complex spatial correlations, 2) are affected by external factors, and 3) have sparse and incomplete data slices. In this paper, we propose an Adaptive Feature Fusion Network (AFFN) to a) adaptively fuse spatial dependencies from multiple knowledge-based graphs and even hidden correlations between stations and b) accurately capture the periodic patterns of passenger flows based on the auto-learned impact from external factors. To deal with the incompleteness and sparsity of OD matrices, we extend AFFN to multi-task AFFN to predict the inflow and outflow of each station as a side-task to further improve OD prediction accuracy. We conducted extensive experiments on two real-world metro trip datasets collected in Nanjing and Xi’an, China. Evaluation results show that our AFFN and multi-task AFFN outperform the state-of-the art baseline techniques and AFFN variants in various accuracy metrics, demonstrating the effectiveness of AFFN and each of its key components in OD prediction.
References
[1] Y. Xu, Y. Lyu, G. Xiong, S. Wang, W. Wu, H. Cui and J. Luo, “Adaptive feature fusion networks for origin-destination passenger flow prediction in metro systems,” IEEE Transactions on Intelligent Transportation Systems, vol.24, no.5, pp.5296-5312, 2023.
[2] T. Wang, J. Song, J. Zhang, J. Tian, J. Wu and J. Zheng, “Short-term metro passenger flow prediction based on hybrid spatiotemporal extraction and multi-feature fusion,” Tunnelling and Underground Space Technology, vol.159, no.106491, 2025.
[3] B. Shi, Y. Liu, S. Lu and Z. W. Gao, “A new adaptive feature fusion and selection network for intelligent transportation systems,” Control Engineering Practice, vol.146, no.105885, 2024.
[4] L. Wan, W. Cheng and J. Yang, “Optimizing Metro Passenger Flow Prediction: Integrating Machine Learning and Time‐Series Analysis with Multimodal Data Fusion,” IET Circuits, Devices & Systems, vol.2024, no.1, pp.5259452, 2024.
[5] L. Xiong, X. Yuan, Z. Hu, X. Huang and P. Huang, “Gated fusion adaptive graph neural network for urban road traffic flow prediction,” Neural Processing Letters, vol.56, no.1, pp.9, 2024.
[6] B. Wang, Y. Leng, G. Wang and Y. Wang, “Fusiontransnet for smart urban mobility: Spatiotemporal traffic forecasting through multimodal network integration,” arXiv preprint arXiv:2405.05786, 2024.
[7] J. Zhao, R. Zhang, Q. Sun, J. Shi, F. Zhuo and Q. Li, “Adaptive graph convolutional network-based short-term passenger flow prediction for metro,” Journal of Intelligent Transportation Systems, vol.28, no.6, pp.806-815, 2024.
[8] J. Yan, H. Li, D. Zhang, Y. Bai, Y. Xu and C. Han, “A multi-feature spatial–temporal fusion network for traffic flow prediction,” Scientific reports, vol.14, no.1, pp.14264, 2024.
[9] J. Sun, X. Ye, X. Yan, T. Wang and J. Chen, “Multi-Step Peak Passenger Flow Prediction of Urban Rail Transit Based on Multi-Station Spatio-Temporal Feature Fusion Model,” Systems, vol.13, no.2, pp.96, 2025.
[10] S. Zhang, J. Zhang, L. Yang, J. Yin and Z. Gao, “Spatiotemporal attention fusion network for short-term passenger flow prediction on new year’s day holiday in urban rail transit system,” IEEE Intelligent Transportation Systems Magazine, vol.15, no.5, pp.59-77, 2023.
[11] Y. Zhou, Y. Liu, N. Ning, L. Wang, Z. Zhang, X. Gao and N. Lu, “Integrating knowledge representation into traffic prediction: a spatial–temporal graph neural network with adaptive fusion features,” Complex & Intelligent Systems, vol.10, no.2, pp.2883-2900, 2024.
[12] Biswanath Saha, “Enhancing Deep Learning Models for Real-Time Decision Making in Autonomous Systems,” International Journal of Advance Research and Innovation, vol.13, no.2, pp.1-11, 2025.
[13] Y. Zhang, Y. Chen, Z. Wang and D. Xin, “TMFO-AGGRU: A graph convolutional gated recurrent network for metro passenger flow forecasting,” IEEE Transactions on Intelligent Transportation Systems, vol.25, no.3, pp.2893-2907, 2023.
[14] C. Xiu, S. Zhan, J. Pan, Q. Peng, Z. Lin and S. C. Wong, “Correlation-based feature selection and parallel spatiotemporal networks for efficient passenger flow forecasting in metro systems,” Transportmetrica A: Transport Science, pp.1-37, 2024.
[15] L. Liu, Y. Liu and X. Ye, “Multi-sequence spatio-temporal feature fusion network for peak-hour passenger flow prediction in urban rail transit,” Transportation Letters, vol.17, no.1, pp.86-102, 2025.
[16] Sreedhar Bhukya and Shaik Khasim Saheb, “A Novel Approach for Analyzing Temporal Influence Dynamics in Social Networks,” Journal of Computer Allied Intelligence, vol.2, no.2, pp.13-21, 2024.
[17] H. Huang, J. Mao, R. Liu, W. Lu, T. Tang and L. Liu, “MTLMetro: A deep multi-task learning model for metro passenger demands prediction,” IEEE Transactions on Intelligent Transportation Systems, vol.25, no.9, pp.11805-11820, 2024.
[18] J. Kong, X. Fan, M. Zuo, M. Deveci, X. Jin and K. Zhong, “ADCT-Net: Adaptive traffic forecasting neural network via dual-graphic cross-fused transformer,” Information Fusion, vol.103, no.102122, 2024.
[19] N. Ramana and E. Hari Krishna, “Intrusion Detection System Fog Security Model for the Smart Cities and Urban Sensing,” Journal of Sensors, IoT & Health Sciences, vol.1, no.1, pp.51-63, 2023.
[20] Y. Wang, T. Liu, Y. Lu, H. Wan, P. Huang and F. Deng, “Traffic Accident Risk Prediction of Tunnel Based on Multi-Source Heterogeneous Data Fusion,” IEEE Access, vol.12, pp.18694-18702, 2024.
[21] P. Chen, J. Lin, Q. Zhao, L. Zhou, T. Yang, X. Huang and J. Wu, “ADF-Net: An attentionguided dual-branch fusion network for building change detection near the Shanghai metro line using sequences of TerraSAR-X images,” Remote Sensing, vol.16, no.6, pp.1070, 2024.
[22] Z. Ul Abideen, X. Sun and C. Sun, “Traffic flow prediction: A 3D adaptive multi‐module joint modeling approach integrating spatial‐temporal patterns to capture global features,” Journal of Forecasting, vol.43, no.7, pp.2766-2791, 2024.
[23] Y. Zhu, Y. Xu, S. Cao, M. Zhang, J. Wang, T. Xie and H. Cai, “Prediction of remaining useful life of metro traction motor bearings based on DCCNN-GRU and multi-information fusion,” Journal of Mechanical Science and Technology, vol.38, no.5, pp.2247-2264, 2024.
