In order to determine the projected rate of deterioration of pavement segments so that the District can schedule resurfacing, a study of performance prediction was conducted on the pavements in District 5, Florida. The development of performance prediction models was primarily based on the regression method. Data related to historical traffic, maximum speed limit on the pavement segment, rainfall amount, temperature, work-mix, historical condition ratings, distress types, pavement surface course thickness, pavement age, and most recent year-to-year rating decline were collected for pavements in District 5. For District 5 in Florida, most pavement distresses are cracking failure, rather than rut or ride failure. After reviewing the pavement condition survey (PCS) report, one of the nine counties in District 5 was selected as the pilot county. Based on the data of the pilot county, analysis was made to determine the appropriate response variable and predictor variables in the prediction model. Statistical tests and regression were performed in the determination of final predictor variables. A prediction model was developed for the pilot county. The validity of the pilot model was checked by comparing the predicted condition ratings with actual ones. After the pilot model was investigated and verified, the regression modeling approach was employed on all nine counties in District 5. The proposed models include predictor variables such as current year condition rating, slope of degradation of previous years, and current annual average daily traffic. An estimated pavement remaining life based on the condition survey is also presented according to the predictive condition rating. (a)
Samenvatting