Faulting Model Improvements for MEPDG

Accurate joint transverse faulting prediction is essential for proper pavement design. Departments of transportation are encouraged to evaluate, recalibrate, and even modify the faulting model if discrepancies between predicted and field measured faulting are observed. The Pennsylvania Department of Transportation (PennDOT) identified a series of problems with Pavement ME faulting prediction for jointed plain concrete pavements (JPCP). Moreover, the current design of bonded concrete overlays on asphalt (BCOA) lacks a faulting model that simulates the different aspects of the BCOA structure. To address these issues, this project had two main objectives: reevaluate and modify Pavement ME faulting model for JPCP and develop a singular faulting model for BCOA. For JPCP, the faulting predictions of the current version of Pavement ME with various reliability levels were evaluated using PennDOT’s Road Management System (RMS). The approach adopting RMS data shows potential for a new use of these types of databases for calibration of distress prediction model evaluation. In addition, a novel methodology accounting for the so-called “survival issue” of the pavement performance databases in model evaluation was introduced. It was found that Pavement ME greatly overpredicted faulting for most cases. Thus, the faulting model was modified and calibrated accordingly improving prediction accuracy. For BCOA, a series of Artificial Neural Networks (ANNs) were developed to estimate the structural response of BCOA to vehicle and environmental loading. This response was quantified in terms of the difference between the deflection basins on the loaded and unloaded slabs and related to Differential Energy (DE). Differential energy is used to characterize damage accumulation which is related to the predicted faulting. The ANNs were then combined with field measurements of faulting to calibrate a faulting model similar to the JPCP faulting model used in Pavement ME but considering the different structural aspects of BCOA. As is the case with the modified JPCP model, the new BCOA model shows good agreement with field data.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Report
  • Features: Appendices; Figures; References; Tables;
  • Pagination: 150p

Subject/Index Terms

Filing Info

  • Accession Number: 01894613
  • Record Type: Publication
  • Report/Paper Numbers: FHWA-PA-2021-005-PITT WO 001
  • Contract Numbers: 4400018535; WO # PIT 001/511801
  • Files: NTL, TRIS, ATRI, USDOT, STATEDOT
  • Created Date: Sep 26 2023 9:09AM