Estimation of Markovian transition probabilities for pavement deterioration forecasting
- 785 Downloads
While it is impossible to estimate when a road section will collapse, the understanding of road section deterioration can help asset managers predict the condition of road sections and take appropriate actions for rehabilitations. Deterioration forecasting modeling is an essential element for an efficient pavement management system. Although the Pavement Management System (PMS) has been introduced and operated for optimal road maintenance since the late 1980s in Korea, some problems for accurate prediction of road deterioration remain due to the quality of pavement performance data and the different pavement structural, material and environmental conditions. In this paper, a methodology to estimate the Markov transition probability model is presented to forecast the deterioration process of road sections. The deterioration states of the road sections are categorized into several ranks and the deterioration processes are characterized by hazard models. The Markov transition probabilities between the deterioration states, which are defined by the non-uniform or irregular intervals between the inspection points in time, are described by the exponential hazard models. Furthermore, in order to verify the validity of the proposed method, the applicability of the estimation methodology presented in this paper is investigated by using the empirical surface data set of the national highway in Korea.
KeywordsMarkov transition probability asset management exponential hazard function road pavement deterioration forecasting pavement management system
Unable to display preview. Download preview PDF.
- Amemiya, T. (1985). Advanced Econometrics, Harvard University Press, Cambridge, MA.Google Scholar
- Butt, A. A., Shahin, M. Y., Carpenter, S. H., and Carnahan, J. V. (1994). “Application of Markov process to pavement management systems at network level.” 3rd International Conference on Managing Pavements, TRB, Vol. 2, pp. 159–172.Google Scholar
- Cesare, M. A., Santamarina, J. C., Turstra, C. J., and Vanmarcke, E. (1992). “Modeling bridge deterioration with Markov chains.” J. of Transp. Engrg., Vol. 118, No. 6, pp. 821–833.Google Scholar
- Federal Highway Administration (FHWA). (1993). PONTIS technical manual, Publ. No. FHWA-SA-94-031, U.S. Dept. of Transp., Washington, D.C.Google Scholar
- Gourieroux, C. (2000). Econometrics of qualitative dependent variables, Cambridge University Press.Google Scholar
- Greene, W. H. (1997). Econometric analysis, 3rd. Ed., Macmillan, New York.Google Scholar
- Jiang, Y., Saito, M., and Sinha, K. C. (1989). “Bridge performance prediction model using the Markov chain.” Transp. Res. Rec. 1180, TRB, pp. 25–32.Google Scholar
- Lancaster, T. (1990). The econometric analysis of transition data, Cambridge University Press.Google Scholar
- Lee, T. C., Judge, G. G., and Zellner, A. (1970). Estimating the parameters of the Markov probability model from aggregate time series data, Amsterdam, North-Holland.Google Scholar
- Loizos, A. and Karlaftis, M.G. (2005). “Prediction of pavement crack initiation from in-service pavements: A duration model approach” J. of the Transportation Research Board, No.1940, TRB, pp. 38–42.Google Scholar
- Tsuda, Y., Kaito, K., Aoki, K., and Kobayashi, K. (2005) “Estimating markovian transition probabilities for bridge deterioration forecasting.” JSCE Journal, No.801/IV-73, pp. 69–82 (in Japanese).Google Scholar