Forecasting Road Condition after Maintenance Works by Linear Methods and Radial Basis Function Networks

  • Konsta Sirvio
  • Jaakko Hollmén
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6792)


Forecasting road condition after maintenance can help in better road maintenance planning. As road administrations annually collect and store road-related data, data-driven methods can be used in determining forecasting models that result in improved accuracy. In this paper, we compare the prediction models identified by experts and currently used in road administration with simple data-driven prediction models, and parsimonious models based on a input selection algorithm. Furthermore, non-linear prediction using radial basis function networks is performed. We estimate and validate the prediction models with a database containing data of over two million road segments.


Forecasting variable selection road maintenance planning road condition rutting International Roughness Index Radial Basis Function networks 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford (1996)zbMATHGoogle Scholar
  2. 2.
    Finnish Road Administration: PMSPro:n kuntoennustemallit 2004. Tiehallinnon selvityksiä 9/2005. Finnish Road Administration (2005) (in Finnish)Google Scholar
  3. 3.
    Finnish Road Administration: Tien päällysteen epätasaisuuden vaikutus ajoneuvojen vierintävastukseen ja ajoneuvokustannuksiin. Tiehallinnon selvityksiä 27/2005. Finnish Road Administration (2005) (in Finnish) Google Scholar
  4. 4.
    Finnish Transport Agency: Finnish Road Statistics 2009. Statistics from the Finnish Transport Agency 2/2010. Finnish Transport Agency (2010)Google Scholar
  5. 5.
    Finnish Transport Agency: Päällysteiden pintakarkeuden vaikutus tienkäyttäjiin ja tienpitoon. Liikenneviraston tutkimuksia ja selityksiä 1/2010. Finnish Transport Agency (2010) (in Finnish)Google Scholar
  6. 6.
    Guyon, I., Elisseeff, A.: An Introduction to Variable and Feature Selection. Journal of Machine Learning Research 3, 1157–1182 (2003)zbMATHGoogle Scholar
  7. 7.
    Kerali, H., Odoki, J.B., Stannard, E.: Overview of HDM-4. The Highway Development and Management Series, vol. 1. The World Road Association (2006)Google Scholar
  8. 8.
    Ruotoistenmäki, A.: Kuntotiedon käyttö tie- ja katuverkon ylläpidon päätöksenteossa. Tiehallinnon selvityksiä 7/2005. Finnish Road Administration (2005)Google Scholar
  9. 9.
    Sirvio, K., Hollmén, J.: Spatio-temporal road condition forecasting with markov chains and artificial neural networks. In: Corchado, E., Abraham, A., Pedrycz, W. (eds.) HAIS 2008. LNCS (LNAI), vol. 5271, pp. 204–211. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  10. 10.
    Sirvio, K., Hollmén, J.: Multi-year network level road maintenance programming by genetic algorithms and variable neighbourhood search. In: Proceedings of the 13th International IEEE Conference on Intelligent Transportation Systems, pp. 581–586 (2010)Google Scholar
  11. 11.
    Sirvio, K., Huda, K.: Implementation of the Road Assets Management System in Sindh Province of Pakistan, G1-01014. In: Viegas, J.M., Macário, R. (eds.) General Proceedings of the 12th World Conference on Transport Research Society (2010)Google Scholar
  12. 12.
    Tikka, J., Hollmén, J.: Sequential input selection algorithm for long-term prediction of time series. Neurocomputing 71, 2604–2615 (2008)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Konsta Sirvio
    • 1
  • Jaakko Hollmén
    • 1
  1. 1.Department of Information and Computer ScienceAalto University School of ScienceAaltoFinland

Personalised recommendations