The primary objective of this research was to investigate the roughness data in the Long Term Pavement Performance (LTPP) Information Management System (IMS) and to develop performance models based on roughness that could be used in a decision support system to determine the remaining pavement life. The focus of the research was on LTPP General Pavement Studies' (GPS) Asphalt Concrete (AC) pavement sections with granular base (GPS-1) and bound base (GPS-2). The models developed can be used to predict the remaining life of the pavement based on the International Roughness Index (IRI).; Six major steps were followed in completing the research. First, a comprehensive review of the existing literature was completed. Second, the LTPP IMS was studied for initial selection of all variables that may contribute to pavement distress. Third, an analysis data base was built. Fourth, initial data verification and identification of outliers was completed. Fifth, data were thoroughly analyzed, trends in IRI data were investigated, and variables with the most significant correlation to the development of roughness in LTPP sections were identified. Lastly, models were constructed and validated.; A statistical analysis approach based on empirical modeling methodology was selected for this research. After the study of numerous linear and nonlinear models, a general nonlinear model format that best fit the data set was selected. The model format includes initial IRI at the time of construction, age as the predominant variable, as well as temperature, moisture, subgrade, and other physical elements.; Six performance models were developed based on GPS-1 data and five models were developed for GPS-2 data for a total of 11 models. The GPS-1 models were categorized in freeze and no-freeze zones in dry climates and based on fine or coarse subgrades and number of days per year with precipitation in wet climates. The GPS-2 models were classified based on the treated base types.
展开▼
