| The objective of this proposed research is to provide a comprehensive study of traffic data modeling for Mechanistic-Empirical Pavement Design Guide (MEPDG). To accomplish this goal, the following research efforts will be conducted: truck weight models for Class 9 vehicle and variability impact analysis of traffic data on pavement design using Monte Carlo simulation technique. Findings of this research will provide pavement designers and traffic engineers a better tool in using traffic data for the MEPDG.;Data sets collected from Weigh-in-Motion (WIM) sites are used as sources to determine axle load spectrum traffic data for the new MEPDG. However, traffic data collected from WIM sites often have errors; sometimes rendering most WIM data sets from a Department of Transportation (DOT) not suitable for MEPDG analysis. For example, based on results from the research project TRC-0702, there are only two stations that passed quality control for weight data in Arkansas with a pool of 55 stations. In this research, a Weight Data Validating Model based on Genetic Algorithm and Clustering Technique is introduced to repair bad weight data to be useful for design analysis.;Accurate and reliable traffic data play an important role in a wide variety of transportation applications, including pavement design. Specifically, the accuracy and reliability of traffic load estimates are keys to determining a pavement's life expectancy. To better understand the impact of traffic variability on the accuracy of loading estimates, this study introduces an analysis method based on Monte Carlo simulation to study the effects of varying classification and weight data on pavement design. |
