Experimentation with Gray Theory for Pavement Smoothness Prediction

Abstract

Riding quality is a dominant characteristic of pavement performance. In the proposed mechanistic–empirical pavement design guide (MEPDG), the functional performance indicator is pavement smoothness as indicated by the international roughness index (IRI). The MEPDG IRI prediction models were developed on the basis of a general hypothesis in which changes in smoothness are caused in part by various distresses, which can be predicted. With pavement distress data from the Long-Term Pavement Performance (LTPP) database, traditional regression analysis was used statistically to establish the MEPDG prediction equations. The gray system theory was devised in the 1980s for modeling uncertain systems with the characteristics of partially known information. A pavement performance prediction system can fit the domain of the gray system. The gray theory–based prediction method was used to develop IRI prediction equations. With the data exported from the LTPP database, it was found that certain specific distresses significantly affect the accuracy of the predictions. Combined with the results of gray relational analysis and the gray prediction methodology, gray model–based smoothness predictions are established by using influencing factors similar to those used in MEPDG. From comparisons of results from the two prediction methodologies with actual LTPP data, it is shown that the gray model–based method provides promising results and is useful for modeling pavement performance.