A Multi-Level Dynamic GCN-Transformer Framework with Spatio-Temporal Interaction for Traffic Flow Prediction
Downloads
As a core task of intelligent transportation systems, traffic flow prediction is characterised by high spatio-temporal complexity. Due to the limitations of existing methods in modelling complex spatio-temporal dependencies, particularly regarding medium- to long-term prediction accuracy and generalisation capabilities, this paper proposes a combination prediction model based on a multi-level dynamic GCN-Transformer framework (DGTFormer), to enhance the accuracy of short- and long-term traffic flow predictions. DGTFormer adopts a dual-stream architecture to achieve spatio-temporal decoupling modelling. The spatio-temporal dynamic graph convolutional network processes dynamic changes in the road network structure, and the temporal transformer encoder processes temporal information related to traffic flow. A spatio-temporal gated fusion mechanism is introduced to deeply couple spatial and temporal information. Experimental results on three real-world traffic datasets (PeMSD4, PeMSD7 and PeMSD8) demonstrate that DGTFormer significantly outperforms mainstream baseline models on multiple key evaluation metrics. Compared with advanced baseline methods, DGTFormer achieves performance improvements of up to 8.61% and 9.15% in RMSE and MAE, respectively. Furthermore, the coefficient of determination, R², remains stable at an excellent level above 0.9 across different prediction time steps, which fully validates that the DGTFormer model possesses superior predictive performance and generalisation capabilities.
Downloads
Zhang K, et al. A deep learning based multitask model for network-wide traffic speed prediction. Neurocomputing. 2020;396:438-450. DOI: 10.1016/j.neucom.2018.10.097.
Han SY, et al. Adaptation scheduling for urban traffic lights via FNT-based prediction of traffic flow. Electronics. 2022;11(4):658. DOI: 10.3390/electronics11040658.
Chen W, et al. Dynamic Path Planning Based on Traffic Flow Prediction and Traffic Light Status. International Conference on Algorithms and Architectures for Parallel Processing 2023 Oct 20. Singapore: Springer Nature Singapore. 2023. p. 419-438. DOI: 10.1007/978-981-97-0834-5_24.
Siroos S, et al. Ensemble of ARIMA: combining parametric and bootstrapping technique for traffic flow prediction. Transport metrica A Transport Science. 2020;16(3):1552-1573. DOI: 10.1080/23249935.2020.1764662.
Pei-Lin N, Zhi YU, Zhao-Cheng. Constrained kalman filter combined predictor for short-term traffic flow. Journal of Traffic and Transportation Engineering. 2008;8(5):86-90.
Guo HF, Fang LJ, Yu L. A short-term traffic flow prediction model based on fuzzy Kalman filtering. Journal of Zhejiang University of Technology. 2013;41(2):218-221.
Wang Z, et al. A novel hybrid method for achieving accurate and timeliness vehicular traffic flow prediction in road networks. Computer Communications. 2023;209:378-386. DOI: 10.1016/j.comcom.2023.07.019.
Mrudula ST, et al. Internet of things and optimized knn based intelligent transportation system for traffic flow prediction in smart cities. Measurement: Sensors. 2024;35:101297. DOI: 10.1016/j.measen.2024.101297.
CHEN Z, et al. Short-term traffic state prediction approach based on FCM and random forest. Journal of Electronics and Information Technology. 2018;40 (8):1879-1886. DOI: 10.11999/JEIT171090.
Xia Y, Chen J. Traffic flow forecasting method based on gradient boosting decision tree. 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017). Atlantis Press. 2017. p. 413-416. DOI: 10.2991/fmsmt-17.2017.87.
CUI Jian-xun, et al. Review on short-term traffic flow prediction methods based on deep learning. Journal of Traffic and Transportation Engineering. 2024;24(2):50-64. DOI: 10.19818/j.cnki.1671-1637.2024.02.003.
Y. Lv, et al. Traffic Flow Prediction With Big Data: A Deep Learning Approach. IEEE Transactions on Intelligent Transportation Systems. 2015;16(2):866-875. DOI: 10.1109/TITS.2014.2345663.
He Y, et al. In-depth insights into the application of Recurrent Neural Networks (RNNs) in traffic prediction: A comprehensive review. Algorithms. 2024;17(9):398. DOI: 10.3390/a17090398.
Song Ruirong, Wang Binjun, Tong Xin. Traffic flow forecasting model based on multi dimensional spatial and temporal characteristics. Science Technology and Engineering. 2021;21(31):13439-13446.
Shao HX, Soong BH. Traffic flow prediction with long short-term memory networks(lstms). Proceedings of 2016 IEEE Region 10 Conference. Singapore: IEEE. 2016:2986-2989. DOI: 10.1109/TENCON.2016.7848593.
Luo W, Dong B, Wang Z. Short-term traffic flow prediction based on CNN-SVR hybrid deep learning model. Journal of Transportation Systems Engineering and Information Technology. 2017;17(5):68-74.
Li Y, et al. Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting. In International Conference on Learning Representations (ICLR), 2018. DOI: 10.48550/arXiv.1707.01926.
Wang X, Xu L. Short-term traffic flow prediction based on deep learning. Journal of Transportation Systems Engineering and Information Technology. 2018;18(1):81-88. DOI: 10.16097/j.cnki.1009-6744.2018.01.012.
Zhang J, et al. Short-term origin-destination demand prediction in urban rail transit systems: A channel-wise attentive split-convolutional neural network method. Transportation Research Part C: Emerging Technologies. 2021;124:102928. DOI: 10.1016/j.trc.2020.102928.
Pareja A, et al. EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs. Proceedings of the AAAI Conference on Artificial Intelligence. 2020;34(04):4542-4549. DOI: 10.1609/aaai.v34i04.5984.
Vaswani A, et al. Attention is all you need. Advances in Neural Information Processing Systems 30 (NIPS 2017). Long Beach, CA, USA: Curran Associates, Inc.; 2017. p. 5998-6008. DOI: 10.48550/arXiv.1706.03762.
Yu B, Yin H, Zhu Z. Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting. Proceedings of the 27th International Joint Conference on Artificial Intelligence; 2018. DOI: 10.24963/ijcai.2018/505.
Zhao L, et al. T-GCN: A temporal graph convolutional network for traffic prediction. IEEE transactions on intelligent transportation systems. 2019;21(9):3848-3858. DOI: 10.1109/TITS.2019.2935152.
Zhang A. Dynamic graph convolutional networks with Temporal representation learning for traffic flow prediction. Scientific Reports. 2025;15(1):1-17. DOI: 10.1038/s41598-025-01696-7.
Chen, Hui, et al. Multi-scale spatio-temporal graph neural network for urban traffic flow prediction. Scientific Reports. 2025;15(1):26732. DOI: 10.1038/s41598-025-11072-0.
Deng H. Traffic-Forecasting Model with Spatio-Temporal Kernel. Electronics. 2025;14(7): 1410. DOI: 10.3390/electronics14071410.
Reza S, Marta C, José J M. A multi-head attention-based transformer model for traffic flow forecasting with a comparative analysis to recurrent neural networks. Expert Systems with Applications. 2022;202:117275. DOI: 10.1016/j.eswa.2022.117275.
Ma J, Zhao J, Hou Y. Spatial–Temporal Transformer Networks for Traffic Flow Forecasting Using a Pre-Trained Language Model. Sensors. 2024;24(17):5502. DOI: 10.3390/s24175502.
Zhou CH, Lin PQ. Traffic volume prediction method based on multi-channel transformer. Comput. Appl. Res. 2023;40:435-439. DOI: 10.19734/j.issn.1001-3695.2022.06.0306.
Liu, S., Wang, X. An improved transformer based traffic flow prediction model. Scientific Reports. 2025;15(1):8284. DOI: 10.1038/s41598-025-92425-7.
Xing Z, et al. Spatial linear transformer and temporal convolution network for traffic flow prediction. Scientific Reports. 2024;14(1):4040. DOI: 10.1038/s41598-024-54114-9.
Xu M, et al. Spatial-temporal transformer networks for traffic flow forecasting. arxiv preprint arxiv:2001.02908; 2020. DOI: 10.48550/arXiv.2001.02908.
Xiong L, et al. Generalized spatial–temporal regression graph convolutional transformer for traffic forecasting. Complex Intell. Syst. 2024;10:7943–7964. DOI: 10.1007/s40747-024-01578-x.
Hu X, et al. GCN-Transformer-Based Spatio-Temporal Load Forecasting for EV Battery Swap Stations under Differential Couplings. Electronics. 2024;13(17): 3401. DOI: 10.3390/electronics13173401.
Zhao Y, Lu J. Spatiotemporal Sequence Prediction Based on Spatiotemporal Self-Attention Mechanism. International Journal of Computers Communications & Control (December). 2024;19(6). DOI: 10.15837/ijccc.2024.6.6771.
Remmouche B, et al. Long-Term Spatio-Temporal Graph Attention Network for Traffic Forecasting. Expert Systems with Applications. 2025;128244. DOI: 10.1016/j.eswa.2025.128244.
Zhang J, et al. Spatio-temporal transformer and graph convolutional networks based traffic flow prediction. Scientific Reports. 2025;15(1):24299. DOI: 10.1038/s41598-025-10287-5.
He H, et al. Spatio-temporal transformer traffic prediction network based on multi-level causal attention. PLoS One. 2025;20(9):e0331139. DOI: 10.1371/journal.pone.0331139.
Chen D, et al. Two-stream spatio-temporal GCN-transformer networks for skeleton-based action recognition. Scientific Reports. 2025;15(1):4982. DOI: 10.1038/s41598-025-87752-8.
Lv Z, et al. ST-TDCN: A two-channel tree-structure spatial–temporal convolutional network model for traffic velocity prediction. Expert Systems with Applications. 2024;257: 125053. DOI: 10.1016/j.eswa.2024.125053.
Lu H, et al. A temporal-aware LSTM enhanced by loss-switch mechanism for traffic flow forecasting. Neurocomputing. 2021;427:169-178. DOI: 10.1016/jneucom.2020.11.026.
Li S, Yang L, Zhao X. Prediction Algorithm of Short-term Traffic Congestion in Urban Area based on Deep Learning. Science Technology and Engineering. 2023;23(25):10866-10878.
Copyright (c) 2026 Guan Lian, Caihua Huang, Qi Sun, Wenyong Li, Yingzi Wu
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
