Quantile Regression Model for Impact Toughness Estimation

  • Satu Tamminen
  • Ilmari Juutilainen
  • Juha Röning
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6171)


The purpose of this study was to develop a product design model for estimating the impact toughness of low-alloy steel plates. The rejection probability in a Charpy-V test (CVT) is predicted with process variables and chemical composition. The proposed method is suitable for the whole production line of a steel plate mill, including all grades of steel in production. The quantile regression model was compared to the joint model of mean and dispersion and the constant variance model. The quantile regression model proved out to be the most effective method for modelling a highly complicated property at this extent.

Next, the developed model will be implemented into a graphical simulation tool that is in daily use in the product planning department and already contains some other mechanical property models. The model will guide designers in predicting the related risk of rejection and in producing desired properties in the product at lower cost.


MLP quantile regression Charpy-V test product design 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Tóth, L., Rossmanith, H.P., Siewert, T.: Historical background and development of the charpy test. In: François, D., Pineau, A. (eds.) From Charpy to Present Impact Testing. Elsevier Science Ltd., UK (2002)Google Scholar
  2. 2.
    Wallin, K., Nevasmaa, P., Planman, T., Valo, M.: Evaluation of the charpy-v test from a quality control test to a materials evaluation tool for structural integrity assessment. In: François, D., Pineau, A. (eds.) From Charpy to Present Impact Testing. Elsevier Science Ltd., UK (2002)Google Scholar
  3. 3.
    Lindroos, Sulonen, Veistinen: Uudistettu Miekk-ojan metallioppi. Otava, Finland (1986) (In Finnish)Google Scholar
  4. 4.
    Cawley, G., Janacek, G., Haylock, M., Dorling, S.: Predictive uncertainty in environmental modelling. Neural Networks 20, 537–549 (2007)zbMATHCrossRefGoogle Scholar
  5. 5.
    Haušild, P.: The influence of ductile tearing on fracture energy in the ductile-to-brittle transition temperature range. Materials Science and Engineering A335, 164–174 (2002)CrossRefGoogle Scholar
  6. 6.
    Todinov, M.: Uncertainty and risk associated with the Charpy impact energy of multi-run welds. Nuclear Engineering and Design 231, 27–38 (2004)CrossRefGoogle Scholar
  7. 7.
    Bhadeshia, H.: Neural networks in materials science. ISIJ International 39(10), 966–979 (1999)CrossRefGoogle Scholar
  8. 8.
    Malinov, S., Sha, W., McKeown, J.: Modelling the correlation between processing parameters and properties in titaniumalloys using artificial neural network. ComputationalMaterials Science 21, 375–394 (2001)Google Scholar
  9. 9.
    Golodnikov, A., Macheret, Y., Trindade, A., Uryasev, S., Zrazhevsky, G.: Statistical modeling of composition and processing parameters for alloy development. Modelling and Simulation in Materials Science and Engineering 13(4), 633–644 (2005)CrossRefGoogle Scholar
  10. 10.
    Koenker, R.: Quantile Regression. Cambridge University Press, USA (2005)zbMATHGoogle Scholar
  11. 11.
    Yu, K., Lu, Z., Stander, J.: Quantile regression: Applications and current research areas. The Statistician 52(3), 331–350 (2003)MathSciNetGoogle Scholar
  12. 12.
    Engel, J.: Modelling variation in industrial experiments. Applied Statistics 41(3), 579–593 (1992)zbMATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Carroll, R., Ruppert, D.: Transformation and Weighting in Regression. Chapman and Hall, USA (1988)zbMATHGoogle Scholar
  14. 14.
    Smyth, G., Huele, A., Verbyla, A.: Exact and approximate REML for heteroscedastic regression. Statistical Modelling 1(3), 161–175 (2001)zbMATHCrossRefGoogle Scholar
  15. 15.
    Juutilainen, I.: Modelling of Conditional Variance and Uncertainty Using Industrial Process Data. PhD thesis, University of Oulu, Finland (2006)Google Scholar
  16. 16.
    Juutilainen, I., Röning, J.: Modelling the probability of rejection in a qualification test based on process data. In: Proc. 16th Symposium of IASC (COMPSTAT 2004), Prague, Czech Republic, August 23-27, pp. 1271–1278 (2004)Google Scholar
  17. 17.
    Koenker, R., Hallock, K.: Quantile regression. Journal of Economic Perspectives 15(4), 143–156 (2001)CrossRefGoogle Scholar
  18. 18.
    Bishop, C.: Neural Networks for Pattern Recognition. Oxford University Press, Inc., USA (1995)Google Scholar
  19. 19.
    Svozil, D., Kvasnička, V., Pospichal, J.: Introduction to multi-layer feed-forward neural networks. Chemometrics and Intelligent Laboratory Systems 39, 43–62 (1997)CrossRefGoogle Scholar
  20. 20.
    Harvey, A.: Regression models with multiplicative heteroscedasticity. Econometrica 44(3), 461–465 (1976)zbMATHCrossRefMathSciNetGoogle Scholar
  21. 21.
    Saerens, M.: Building cost functions minimizing to some summary statistics. IEEE Transactions on Neural Networks 11(6), 1263–1271 (2000)CrossRefGoogle Scholar
  22. 22.
    Dorling, S., Foxall, R., Mandic, D., Cawley, G.: Maximum likelihood cost functions for neural network models of air quality data. Atmospheric Environment 37, 3435–3443 (2003)CrossRefGoogle Scholar
  23. 23.
    Laurinen, P., Tuovinen, L., Röning, J.: Smart archive: A component-based data mining application framework. In: Proc. 5th International Conference on Intelligent Systems Design and Applications, Wroclaw, Poland, September 8-10, pp. 20–26 (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Satu Tamminen
    • 1
  • Ilmari Juutilainen
    • 1
  • Juha Röning
    • 1
  1. 1.Intelligent Systems GroupUniversity of OuluFinland

Personalised recommendations