Skip to main content

Identification and estimation of latent group-level-effects in infrastructure performance modeling

Abstract

As in other panel data analyses, the presence of unobserved heterogeneity is a critical issue in the estimation of infrastructure performance models. In the literature, this issue has been addressed by formulating variable intercept, fixed or random effects models under the assumptions that (1) heterogeneity stems from facility/individual-level effects, and that (2) the coefficients are constant and homogeneous across the population. In contrast, we present mixture regression as a performance modeling framework. The approach relies on the assumption that the underlying population is comprised of a finite set of classes/segments in unknown proportions. The segmentation basis is latent meaning that the criteria to establish the number and type of segments are related to unobserved heterogeneity manifested in facility performance/deterioration. The segments are characterized by a set of commonly specified regression equations, which allows for the identification and estimation of coefficients, i.e., group-level effects, that differ in terms of their level-of-significance, magnitude or sign. We also derive an instance of the Expectation-Maximization Algorithm to estimate the associated parameters, and to assign facilities to the population segments. To illustrate the framework, we analyze the performance of a panel of 131 pavements from the AASHO Road Test. The results suggest both observed and unobserved sources of heterogeneity in the panel. The heterogeneity is captured by differential group-level effects, which we estimate and interpret. We also discuss how these effects can be exploited in the development of resource allocation strategies. We also compare the mixture regression model to established benchmarks.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

References

  1. Archilla A (2006) Repeated measurement data analysis in pavement deterioration modeling. J Infrastruct Syst 12(3):163–173

    Article  Google Scholar 

  2. Archilla A, Madanat S (2000) Development of a pavement rutting model from experimental data. J Transp Eng 126(4):291–299

    Article  Google Scholar 

  3. Ben-Akiva M, Humplick F, Madanat S, Ramaswamy R (1993) Infrastructure management under uncertainty: The latent performance approach. ASCE J Transp Eng 119(1):43–58

    Article  Google Scholar 

  4. Carnahan J, Davis W, Shahin M, Keane P, Wu M (1987) Optimal maintenance decisions for pavement management. ASCE J Transp Eng 113(5):554–573

    Article  Google Scholar 

  5. Chu C, Durango-Cohen P (2008a) Empirical comparison of statistical pavement performance models. J Infrastruct Syst 14(2):138–149

    Article  Google Scholar 

  6. Chu C, Durango-Cohen P (2008b) Estimation of dynamic performance models for transportation infrastructure using panel data. Transp Res Part B Methodol 42(1):57–81

    Article  Google Scholar 

  7. Chu C, Durango-Cohen P (2008c) Incorporating maintenance effectiveness in the estimation of dynamic infrastructure performance models. Comput Aided Civ Infrastruct Eng 23(3):174–188

    Article  Google Scholar 

  8. Daziano RA, Bolduc D (2013) Incorporating pro-environmental preferences towards green automobile technologies through a bayesian hybrid choice model. Transp A Transp Sci 9(1):74–106

    Google Scholar 

  9. Dempster A, Laird N, Rubin D (1977) Maximum likelihood from incomplete data via the em algorithm. J R Stat Soc Ser B 39(1):1–38

    Google Scholar 

  10. Gaffney S, Smyth P (1999) Trajectory clustering with mixtures of regression models. In: Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining, ACM, pp 63–72

  11. Gendreau M, Soriano P (1998) Airport pavement management systems: an appraisal of existing methodologies. Transp Res Part A 32(3):197–214

    Google Scholar 

  12. Golabi K, Kulkarni R, Way G (1982) A statewide pavement management system. Interfaces 12(6):5–21

  13. Goldberger AS (1962) Best linear unbiased prediction in the generalized linear regression model. J Am Stat Assoc 57(298):369–375

    Article  Google Scholar 

  14. Greene W (2011) Econometric Analysis, 7th edn. Prentice Hall, Upper Saddle River, NJ

    Google Scholar 

  15. Highway Research Board (1962) The AASHO Road Test, Special Report No. 61A-E. National Academy of Science, National Research Council, Washington, DC.

  16. Hong F, Prozzi J (2006) Estimation of pavement performance deterioration using bayesian approach. J Infrastruct Syst 12(2):77–86

    Article  Google Scholar 

  17. Hong F, Prozzi J (2010) Roughness model accounting for heterogeneity based on in-service pavement performance data. J Transp Eng 136(3):205–213

    Article  Google Scholar 

  18. Hsiao C (1986) Analysis of panel data. Cambridge University Press, Cambridge, UK

    Google Scholar 

  19. Hudson W, Haas R, Uddin W (1997) Infrastructure management. McGraw-Hill, New York, NY

    Google Scholar 

  20. Humplick F (1992) Highway pavement distress evaluation: modeling measurement error. Transp Res Part B Methodol 26(2):135–154

    Article  Google Scholar 

  21. Jedidi K, Jagpal HS, DeSarbo WS (1997) Finite-mixture structural equation models for response-based segmentation and unobserved heterogeneity. Mark Sci 16(1):39–59

    Article  Google Scholar 

  22. Luo Z, Chou Y (2006) Pavement condition prediction using clusterwise regression. Transp Res Rec 1974:70–77

    Article  Google Scholar 

  23. Luo Z, Yin H (2008) Probabilistic analysis of pavement distress ratings with the clusterwise regression method. Transp Res Rec 2084:38–46

    Article  Google Scholar 

  24. Madanat S, Karlaftis M, McCarthy P (1997) Probabilistic infrastructure deterioration models with panel data. J Infrastruct Syst 3(1):4–9

    Article  Google Scholar 

  25. Malyshkina NV, Mannering FL (2009) Markov switching multinomial logit model: an application to accident-injury severities. Accid Anal Prev 41(4):829–838

    Article  Google Scholar 

  26. Malyshkina NV, Mannering FL, Tarko AP (2009) Markov switching negative binomial models: an application to vehicle accident frequencies. Accid Anal Prev 41(2):217–226

    Article  Google Scholar 

  27. McGilchrista C, Yaua K (1995) The derivation of blup, ml, reml estimation methods for generalised linear mixed models. Commun Stat Theory Methods 24(12):2963–2980

    Article  Google Scholar 

  28. McNeil S, Markow M, Neumann L, Ordway J, Uzarski D (1992) Emerging issues in transportation facilities management. J Transp Eng 118(4):477–495

    Article  Google Scholar 

  29. Onar A, Thomas F, B C, Byron T (2006) Statistical mixed effects models for evaluation and prediction of accelerated pavement testing results. J Transp Eng 132(10):771–780

  30. Park BJ, Lord D (2009) Application of finite mixture models for vehicle crash data analysis. Accid Anal Prev 41(4):683–691

    Article  Google Scholar 

  31. Prozzi J, Madanat S (2003) Incremental nonlinear model for predicting pavement serviceability. J Transp Eng 129(6):635–641

    Article  Google Scholar 

  32. Späth H (1977) Computational experience with the exchange method. Eur J Oper Res 1(1):23–31

    Article  Google Scholar 

  33. Späth H (1979) Algorithm 39: Clusterwise linear regression. Computing 22:367–373

    Article  Google Scholar 

  34. Späth H (1982) Algorithm 48: A fast algorithm for clusterwise linear regression. Computing 29:175–181

    Article  Google Scholar 

  35. Van Noortwjk J, Frangopol D (2004) Deterioration and maintenance models for insuring safety of civil infrastructures at lowest life-cycle cost. In: Faber M, Adey B (eds) Frangopol D, Br\(\ddot{\text{ u }}\)hwiler E. Published in Life-Cycle Performance of Deteriorating Structures, Assessment, Design and Management, ASCE, pp 384–391

  36. Vij A, Carrel A, Walker JL (2013) Incorporating the influence of latent modal preferences on travel mode choice behavior. Transp Res Part A Polic Pract 54:164–178

    Article  Google Scholar 

  37. Washington S, Karlaftis M, Mannering F (2011) Statistical and econometric methods for transportation data analysis, 2nd edn. Chapman & Hall/CRC, Boca Raton, FL

    Google Scholar 

  38. Wedel M, Kamakura W (2000) Market segmentation: conceptual and methodological foundations. Kluwer Academic Publishers, Norwell, MA

    Google Scholar 

  39. Xiong Y, Mannering FL (2013) The heterogeneous effects of guardian supervision on adolescent driver-injury severities: A finite-mixture random-parameters approach. Transp Res Part B Methodol 49:39–54

    Article  Google Scholar 

  40. Xu L, Jordan MI, Hinton GE (1995) An alternative model for mixtures of experts. Advances in neural information processing systems pp 633–640

  41. Yu J, Chou E, Luo Z (2007) Development of linear mixed effects models for predicting individual pavement conditions. J Transp Eng 133(6):347–354

    Article  Google Scholar 

  42. Zhang W, Durango-Cohen P (2014) Explaining heterogeneity in pavement deterioration: a clusterwise linear regression model. J Infrastruct Syst 20(2):04014005 1–9

Download references

Acknowledgments

This research was partially supported by a Dissertation Year Fellowship awarded by The Northwestern University Transportation Center to the second author.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Pablo L. Durango-Cohen.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Medury, A., Zhang, W. & Durango-Cohen, P.L. Identification and estimation of latent group-level-effects in infrastructure performance modeling. EURO J Transp Logist 4, 31–55 (2015). https://doi.org/10.1007/s13676-014-0057-1

Download citation

Keywords

  • Infrastructure performance modeling
  • Mixture regression
  • EM algorithm
  • Panel data
  • Unobserved heterogeneity