Comparing MEPDG Distress Predictions to Automated and Manual Interpretations

The Mechanistic-Empirical Pavement Design Guide (MEPDG) software was designed on a nationally calibrated basis, thus calibrating the existing parameters to suit local conditions is necessary for accurate distress predictions. The process of calibration involves iterating distress model coefficients until the predicted results match a desired outcome. This desired outcome is typically associated with monitored pavement distress results, documented through manual methods and stored in pavement management systems. However, the calibration process can be very time consuming, thus sufficient automated methods of data collection are desirable. The cracking protocol used by MEPDG is consistent with the cracking definitions provided by the Long Term Pavement Performance (LTPP) program. Due to the difficulties involved with the development of cracking recognition algorithms for automated distress survey software, many have made attempts to promote alternative cracking protocols that are more suitable for automated technologies. Significant efforts for the enhancement of the MEPDG software have also been taking place, thus recent research has been pulled in two directions. This paper portrays the implementation of a post-processing tool used to modify the cracking interpretations of an existing surveying software called the Automated Distress Analyzer (ADA). This tool incorporates wheelpath boundaries in order to establish a means of defining cracks in regards to LTPP. The results indicated that the automated post-processing tool has the ability to replicate nationally-calibrated MEPDG distress predictions better than a surveying method involving human intervention as well as ADA by itself.