Simplified staged-homogenous Markov model for flexible pavement performance prediction

This paper presents a simplified staged-homogenous Markov model proposed for predicting future pavement conditions at the project level with minimal time and effort. The analysis period is divided into equal staged-time periods with each represented by a unique transition probability matrix. The deterioration transition probabilities are expected to increase over time due to the progressively increasing traffic loading and gradually degrading pavement structure. Therefore, the deterioration transition probabilities associated with staged-time periods are estimated from multiplying the present deterioration transition probabilities by appropriate constants called C constants. The C constants are to be estimated from minimising the sum of squared errors (SSE) defined as the differences between the predicted and observed distress ratings (DRs). A simplified sequential trial-and-error minimisation approach is proposed for obtaining the best estimates of the C constants. The presented sample problem has indicated the effectiveness of the proposed staged-homogenous Markov model in predicting future pavement DRs using five-year staged-time periods. This has been validated using three key performance indicators, namely the generated best-fit performance curves, minimum SSE, and staged-performance ratios.