Traffic flow prediction is an important part of intelligent transportation system, which can provide important support for traffic management optimization, resource scheduling and urban planning. At its core, it uses historical and real-time data to predict future traffic conditions. The main challenge of traffic flow forecasting is how to effectively model complex spatio-temporal dependencies in traffic data. However, with the continuous development of traffic forecasting technology, traffic forecasting models still have obvious shortcomings in dynamic spatial dependence and periodicity modeling, which limits their forecasting effect. Therefore, this paper proposes an improved traffic flow prediction model DASTFormer (Dynamic Adaptive Spatio-Temporal Transformer). Firstly, an adaptive adjacency matrix is added to the embedding layer to enable the model to dynamically adjust the relationship weights between nodes, thereby enhancing the adaptability of the model to the spatial dependence of real-time changes. Secondly, an Adaptive Weighted Adapter is designed and combined with semantic spatial attention to further optimize the model’s semantic dependency modeling in complex traffic scenarios. Finally, adding the Auto-Correlation and then combining the self-attention mechanism into the time prediction module can effectively capture the periodic features in the data, and also improve the modeling ability of the model for complex temporal relationships, thus further enhancing the accuracy and stability of the prediction. Experimental results show that the proposed method achieves better prediction accuracy and robustness on multiple traffic datasets.
QureshiKNAbdullahAH.A survey on intelligent transportation systems. Middle East J Sci Res2013; 15(5): 629–642.
2.
SinghBGuptaA.Recent trends in intelligent transportation systems: a review. J Transp Lit2015; 9(2): 30–34.
3.
GomesBCoelhoJAidosH.A survey on traffic flow prediction and classification. Intell Syst Appl2023; 20: 200268.
4.
GeHXDaiSQDongLY, et al. Stabilization effect of traffic flow in an extended car-following model based on an intelligent transportation system application. Phys Rev E2004; 70(6): 066134.
KarimpourMKarimpourAKompanyK, et al. Online traffic prediction using time series: a case study. In: ConstandaCDalla RivaMLambertiP, et al. (eds) Integral methods in science and engineering, volume 2: practical applications. Birkhäuser, 2017, pp. 147–156.
7.
JiangWLuoJ.Graph neural network for traffic forecasting: a survey. Expert Syst Appl2022; 207: 117921.
8.
RiediRHWillingerW.Toward an improved understanding of network traffic dynamics. In: ParkKWillingerW (eds) Self-similar network traffic and performance evaluation. John Wiley & Sons, Inc., 2000, pp. 507–530.
9.
CaiLJanowiczKMaiG, et al. Traffic transformer: capturing the continuity and periodicity of time series for traffic forecasting. Trans GIS2020; 24(3): 736–755.
10.
ZhangYLiuSZhangP, et al. GRU-and transformer-based periodicity fusion network for traffic forecasting. Electronics2023; 12(24): 4988.
11.
TangJWangYWangH, et al. Complex network approach for the complexity and periodicity in traffic time series. Procedia Soc Behav Sci2013; 96: 2602–2610.
12.
YuanHLiGBaoZ.Route travel time estimation on a road network revisited: heterogeneity, proximity, periodicity and dynamicity. Proceedings VLDB Endowment2022; 16(3): 393–405.
13.
TangJWangYWangH, et al. Dynamic analysis of traffic time series at different temporal scales: a complex networks approach. Physica A2014; 405: 303–315.
14.
ChenMYuXLiuY.PCNN: deep convolutional networks for short-term traffic congestion prediction. IEEE Trans Intell Transp Syst2018; 19(11): 3550–3559.
15.
WilliamsBMHoelLA.Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: theoretical basis and empirical results. J Transp Eng2003; 129(6): 664–672.
16.
KumarSVVanajakshiL.Short-term traffic flow prediction using seasonal ARIMA model with limited input data. Eur Transp Res Rev2015; 7: 1–9.
17.
ShekharS.Recursive methods for forecasting short-term traffic flow using seasonal ARIMA time series model. Master Thesis, North Carolina State University, USA, 2004.
18.
SherstinskyA.Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Physica D2020; 404: 132306.
AndreolettiDTroiaSMusumeciF, et al. Network traffic prediction based on diffusion convolutional recurrent neural networks. In: IEEE INFOCOM 2019-IEEE conference on computer communications workshops (INFOCOM WKSHPS), Paris, France, 29 April–2 May 2019, pp. 246–251. New York: IEEE.
26.
YuBYinHZhuZ.Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. arXiv preprint arXiv:1709.04875, 2017.
27.
HeSLuoQDuR, et al. STGC-GNNs: a GNN-based traffic prediction framework with a spatial–temporal Granger causality graph. Physica A2023; 623: 128913.
28.
ZhuJWangQTaoC, et al. AST-GCN: attribute-augmented spatiotemporal graph convolutional network for traffic forecasting. IEEE Access2021; 9: 35973–35983.
29.
GuoSLinYFengN, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. Proc AAAI Conf Artif Intell2019; 33(1): 922–929.
30.
WuZPanSLongG, et al. Graph WaveNet for deep spatial-temporal graph modeling. arXiv preprint arXiv:1906.00121, 2019.
31.
TikunovDNishimuraT. Traffic prediction for mobile network using Holt-Winter’s exponential smoothing. In: 2007 15th international conference on software, telecommunications and computer networks, Split, Croatia, 27–29 September 2007, pp. 1–5. New York: IEEE.
32.
ZivotEWangJ.Vector autoregressive models for multivariate time series. In: ZivotEWangJ (eds) Modeling financial time series with S-PLUS®. Springer, 2006, pp. 385–429.
33.
FuRZhangZLiL. Using LSTM and GRU neural network methods for traffic flow prediction. In: 2016 31st youth academic annual conference of Chinese Association of Automation (YAC), Wuhan, China, 11–13 November 2016, pp. 324–328. New York: IEEE.
34.
ChenJZhengLHuY, et al. Traffic flow matrix-based graph neural network with attention mechanism for traffic flow prediction. Inf Fusion2024; 104: 102146.
35.
LiuYWuH. Prediction of road traffic congestion based on random forest. In: 2017 10th international symposium on computational intelligence and design (ISCID), Hangzhou, China, 9–10 December 2017, vol. 2, pp. 361–364. New York: IEEE.
36.
HanKXiaoAWuE, et al. Transformer in transformer. Adv Neural Inf Process Syst2021; 34: 15908–15919.
37.
XuSZengB.Network traffic prediction model based on auto-regressive moving average. J Netw2014; 9(3): 653.
38.
KrithikaivasanBZengYDekaK, et al. ARCH-based traffic forecasting and dynamic bandwidth provisioning for periodically measured nonstationary traffic. IEEE ACM Trans Netw2007; 15(3): 683–696.
39.
AnandNCScoglioCNatarajanB. GARCH—non-linear time series model for traffic modeling and prediction. In: NOMS 2008-2008 IEEE network operations and management symposium, Salvador, Brazil, 7–11 April 2008, pp. 694–697. New York: IEEE.
40.
GopinathRA. Maximum likelihood modeling with Gaussian distributions for classification. In: Proceedings of the 1998 IEEE international conference on acoustics, speech and signal processing, ICASSP’98 (Cat. No. 98CH36181), Seattle, WA, USA, 15 May 1998, vol. 2, pp. 661–664. New York: IEEE.
41.
JiangJHanCZhaoWX, et al. PDFormer: propagation delay-aware dynamic long-range transformer for traffic flow prediction. Proc AAAI Conf Artif Intell2023; 37(4): 4365–4373.
42.
WuHXuJWangJ, et al. Autoformer: decomposition transformers with auto-correlation for long-term series forecasting. Adv Neural Inf Process Syst2021; 34: 22419–22430.
43.
WuZPanSLongG, et al. Connecting the dots: multivariate time series forecasting with graph neural networks. In: Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, Virtual Event, 6–10 July 2020, pp. 753–763. New York: ACM.
44.
SongCLinYGuoS, et al. Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting. Proc AAAI Conf Artif Intell2020; 34(1): 914–921.
FangZLongQSongG, et al. Spatial-temporal graph ode networks for traffic flow forecasting. In: Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, Virtual Event, 14–18 August 2021, pp. 364–373. New York: ACM.
47.
ChoiJChoiHHwangJ, et al. Graph neural controlled differential equations for traffic forecasting. Proc AAAI Conf Artif Intell2022; 36(6): 6367–6374.
48.
XuMDaiWLiuC, et al. Spatial-temporal transformer networks for traffic flow forecasting. arXiv preprint arXiv:2001.02908, 2020.
49.
ZhengCFanXWangC, et al. GMAN: a graph multi-attention network for traffic prediction. Proc AAAI Conf Artif Intell2020; 34(1): 1234–1241.
50.
YanHMaXPuZ.Learning dynamic and hierarchical traffic spatiotemporal features with transformer. IEEE Trans Intell Transp Syst2021; 23(11): 22386–22399.
51.
JiangRWangZYongJ, et al. MegaCRN: meta-graph convolutional recurrent network for spatio-temporal modeling. arXiv preprint arXiv:2212.05989, 2022.
52.
JinDHuoCShiJ, et al. LLGformer: learnable long-range graph transformer for traffic flow prediction. In: Proceedings of the ACM on web conference 2025, Sydney, NSW, Australia, 28 April–2 May 2025, pp. 2860–2871. New York: ACM.
53.
GuoQTanQTangJ, et al. Unifying spatiotemporal and frequential attention for traffic prediction. Sci Rep2025; 15(1): 953.
54.
WangXYangSR.Lightweight spatio-temporal attention network with graph embedding and rotational position encoding for traffic forecasting. arXiv preprint arXiv:2505.12136, 2025.
55.
WengWFanJWuH, et al. A decomposition dynamic graph convolutional recurrent network for traffic forecasting. Pattern Recognit2023; 142: 109670.
