Probabilistic Approach for Development of Track Geometry Defects as a Function of Ground Penetrating Radar Measurements

  • Denis Yurlov
  • Allan M. ZarembskiEmail author
  • Nii Attoh-Okine
  • Joseph W. Palese
  • Hugh Thompson
Technical Paper


Recent research has shown a relationship between track geometry defects and track subsurface conditions as measured by Ground Penetrating Radar (GPR). This paper presents the results of a comprehensive study looking at the development of a probabilitic model for the prediction of track geometry defects as a function of key subgrade parameters as measured by GPR. Specifically, several Logistical Regression (LR) analyses were performed, to include conventional LR modeling as well as a hybrid LR modeling approach based on Hierarchical Clustering Analysis with Histogram Data. The result was a higher order polynomial Logistic Regression model for determination of the probability of a track geometry surface defect occurring at locations with measured ballast fouling and measured ballast thickness. The results showed that there was a statistically significant relationship between high rates of geometry degradation and poor subsurface condition as defined by the GPR parameters: Ballast Fouling Index (BFI) and Ballast Layer Thickness (BLT). Furthermore, a predictive model was developed to determine the probability of a high rate of geometry degradation as a function of these key GPR parameters.


Railways Railway track Track geometry Track deterioration Ground penetrating radar Hierarchical clustering analysis Logistic regression 



The authors would like to acknowledge and thank the Federal Railroad Administration, U.S. Department of Transportation for their support of this research. The authors would also like to thank Gary Carr of the FRA and Ted Sussman of the Volpe National Transportation Systems Center for their support of this project. The authors would also like to acknowledge and thank Mike Trosino and Amanda Kessler of Amtrak for providing the geometry and GPR data used in this analysis.


  1. 1.
    Zarembski, A.M., Grissom, G.T., Euston, T.L.: On the use of ballast inspection technology for the management of track substructure. J. Trans. Infrastruct. Geotechnol. 1(1), 83–109 (2014)CrossRefGoogle Scholar
  2. 2.
    Palese, J. W., Hartsough, C. H., Zarembski, A. M., Thompson, H., Ling, H. L., Pagano, W.: Life Cycle Benefits of Subgrade Reinforcement Using Geocell on a High-Speed Railway- A Case Study. American Railway Engineering Association Annual Conference, Indianapolis (2017)Google Scholar
  3. 3.
    Zarembski, A.M., Palese, J.W., Hartsough, C.H., Ling, H.I., Thompson, H.: Application of Geocell track substructure support system to correct surface degradation problems under high speed passenger railroad operations. J. Trans. Infrastruct. Geotechnol. 4(4), 106–125 (2017). CrossRefGoogle Scholar
  4. 4.
    Li, D., Hyslip, J., Sussmann, T., Chrismer, S.: Railway Geotechnics. CRC Press, Boca Raton (2016)Google Scholar
  5. 5.
    Fontal, S., et al.: Railways track characterization using ground penetrating radar. Proc. Eng. 143, 1193–1200 (2016)CrossRefGoogle Scholar
  6. 6.
    Zarembski, A.M., Yurlov, D., Palese, J.W., Attoh-Okine, N., Thompson, H.: Relationship Between Track Geometry Degradation and Subsurface Condition as Measured by GPR. American Railway Engineering Association Annual Conference, Chicago (2018)Google Scholar
  7. 7.
    Attoh-Okine, N.: Big Data and Differential Policy: Analysis Strategies for Railroad Track Engineering. John Wiley & Sons, Hoboken (2017)CrossRefGoogle Scholar
  8. 8.
    Freedman, D. A.: Statistical Models: Theory and Practice. Cambridge University Press. 2009Google Scholar
  9. 9.
    Müllner, D.: Modern hierarchical, agglomerative clustering algorithms. IEEE Signal Process Lett. 19(4), 231–234 (2011)Google Scholar
  10. 10.
    Billard, L., Kim, J.: Hierarchical clustering for histogram data. Wiley Interdiscip. Rev. Comput. Stat. 9(5), e1405 (2017)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Irpino, A., Verde, R.: Basic statistics for distributional symbolic variables: a new metric-based approach. ADAC. 9(2), 143–175 (2015)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Punj, G., Stewart, D.W.: Cluster analysis in marketing research: review and suggestions for application. J. Mark. Res. 20(2), 134 (1983)CrossRefGoogle Scholar
  13. 13.
    Irpino, A., Verde, R., De Carvalho, F.D.A.T.: Dynamic clustering of histogram data based on adaptive squared Wasserstein distances. Exp. Syst. Appl., Elsevier Ltd. 41(7), 3351–3366 (2014)CrossRefGoogle Scholar
  14. 14.
    Hosmer, D. W., Lemeshow, S., Sturdivant, R. X.: Logistic regression models for the analysis of correlated data. Appl. Log. Regress. 313–376 (2013)Google Scholar
  15. 15.
    Boryga, M., Graboś, A.: Planning of manipulator motion trajectory with higher-degree polynomials use. Mech. Mach. Theory. 44(7), 1400–1419 (2009)CrossRefzbMATHGoogle Scholar
  16. 16.
    Stehman, S.V.: Selecting and interpreting measures of thematic classification accuracy. Remote Sens. Environ. 62(1), 77–89 (1997). CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Track and Right-of-Way Israel RailwaysLodIsrael
  2. 2.Department of Civil and Environmental EngineeringUniversity of DelawareNewarkUSA
  3. 3.Federal Railroad AdministrationWashingtonUSA

Personalised recommendations