Journal of
Faculty of Civil Engineering


FACULTY JOURNALS
SCIENTIFIC MEETINGS
ARCHIVE







JOURNAL OF FACULTY OF CIVIL ENGINEERING 22, 2013.y., pp. 173-183


PROBABILISTIC MODELS OF PAVEMENT DETERIORATION
 
DOI: 10.14415/zbornikGFS22.016
UDC: 625.04
CC-BY-SA 4.0 license
Author : Dragana Zeljić, Igor Jokanović
 
 Summary:
 Pavements are exposed to repeated traffic load and the effect of the environment, which eventually leads to deformations and damage. Prediction of pavement deterioration is essential to determine the parameters for the design of pavements, and to select appropriate ways of maintaining and planning of major interventions (rehabilitation and reconstruction). Models for predicting pavement deterioration can be broadly categorized into deterministic and probabilistic depending on the method used to simulate pavement aging. This paper deals with prediction of pavement deterioration using probabilistic methods at the network level by the help of transition probability matrix and optimization of objective function. The main reason for underuse of this model by engineers in the past is the difficulty of determining transition probabilities (i.e. probability that the road at a certain level of deterioration will “pass“ to another level of deterioration) from empirical data or from observation. As a possible solution to this problem, the paper will describe methods for the determination of transition probabilities, chose optimum and give recommendations for their use.
 
 Keywords:
 Pavement condition, pavement deterioration, probabilistic methods, Markov chains, optimization, transition probabilities.