The Traffic Flow Prediction Using Bayesian and Neural Networks

Part of the Studies in Systems, Decision and Control book series (SSDC, volume 32)


The article presents two short-term forecasting models for determining the traffic flow volumes. The road traffic characteristics are essential for identification the trends in the distribution of the road traffic in the network, determination the capacity of the roads and the traffic variability over the time. The presented model is based on the historical, detailed data concerning the road traffic. The aim of the study was to compare the short-term forecasting models based on Bayesian networks (BN) and artificial neural networks (NN), which can be used in traffic control systems especially incorporated into modules of Intelligent Transportation Systems (ITS). Additionally the comparison with forecasts provided by the Bayesian Dynamic Linear Model (DLM) was performed. The results of the research shows that artificial intelligence methods can be successfully used in traffic management systems.


Root Mean Square Error Traffic Flow Road Traffic Mean Absolute Percentage Error Error Indicator 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Chrobok R, Kaumann O, Wahle J, Schreckenberg M (2004) Different methods of traffic forecast based on real data. Eur J Oper Res 155(3):558–568zbMATHMathSciNetCrossRefGoogle Scholar
  2. 2.
    Chen H, Grant-Muller S, Mussone L, Montgomery F (2001) A study of hybrid neural network approaches and the effects of missing data on traffic forecasting. Neural Comput Appl 10:277–286zbMATHCrossRefGoogle Scholar
  3. 3.
    Tan MC, Wong SC, Xu JM, Guan ZR, Zhang P (2009) An aggregation approach to short-term traffic flow prediction. IEEE Trans Intell Trans Syst 10:60–69Google Scholar
  4. 4.
    Vlahogianni EI, Karlaftis MG, Golias JC (2005) Optimized and meta-optimized neural networks for short-term traffic flow prediction: a genetic approach. Transp Res Part C 13:211–234CrossRefGoogle Scholar
  5. 5.
    Pamuła T (2012) Traffic flow analysis based on the real data using neural networks. In: Mikulski J (ed) Telematics in the transport environment. Selected papers. Springer, Berlin, pp 364–371Google Scholar
  6. 6.
    Bolstad WM (2004) Introduction to Bayesian statistics. Wiley-Interscience, HobokenGoogle Scholar
  7. 7.
    Srinivasan D, Choy MC, Cheu RL (2006) Neural networks for real-time traffic signal control. IEEE Trans Intell Trans Syst 7(3):261–271CrossRefGoogle Scholar
  8. 8.
    Skrobisz C (2010) Bayesian prediction for non-full information on the example of electricity. Folia Pomer Univ Technol Stetin Oeconomica 280(59):99–108Google Scholar
  9. 9.
    Pamuła T (2012) Classification and prediction of traffic flow based on real data using neural networks, Arch Transp 24(4):519–530Google Scholar
  10. 10.
    GeNie package (2014).
  11. 11.
    Cheng J, Druzdzel MJ (2000) An adaptive importance sampling algorithm for evidential reasoning in large Bayesian networks. J Artif Intell Res 13:155–188Google Scholar
  12. 12.
    Heckerman D, Geiger D, Chickering DM (1995) Learning Bayesian networks: the combination of knowledge and statistical data. Mach Learn 20(3):197–243Google Scholar
  13. 13.
    Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J Roy Stat Soc B 39(1):1–38zbMATHMathSciNetGoogle Scholar
  14. 14.
    West M, Harrison J (1997) Bayesian forecasting and dynamic models, 2nd edn. Springer, New York, p 34Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Faculty of TransportSilesian University of TechnologyKatowicePoland

Personalised recommendations