Volume 35, Issue 5 (September 2007)
Verification and Calibration of VESYS5W Fatigue Cracking Model Using Results from Accelerated Pavement Testing
Cost effective design methodology demands accurate performance modeling. Fatigue cracking is one of the major performance indicators in flexible pavement systems. Through literature review, a decision was made to modify the fatigue cracking model in the original VESYS5 program. The revised model in VESYS5W has the same basic form as suggested by the Asphalt Institute (AI) and the 2002 Mechanistic-Empirical Pavement Design Guide (MEPDG). The main challenge in modeling fatigue cracking is to determine appropriate input parameters. Efforts were made to compare the experimental results of full-scale accelerated pavement testing (APT) to the VESY5W predictions using the material parameters suggested by AI and the 2002 MEPDG. The test data employed in this paper are obtained from the full-scale APT conducted by the Federal Highway Administration (FHWA) under a national pooled fund study TPF-5(019). Test results from five lanes at testing temperatures of 19°C were utilized in the comparisons. It was found that the VESYS5W prediction using the material parameters from 2002 MEPDG yielded a much better fit to the experimental results than those using AI material parameters. The VESYS5W predictions using the AI material parameters always underestimated the fatigue cracking life. To match experimental results, shift factors were applied. The shift factors ranged from 0.48 to 3.1 for the predictions using the material parameters from the 2002 MEPDG. However, the shift factors for the AI material parameters ranged from 3.6 to 14.4. In summary, VESYS5W predictions using 2002 MEPDG material parameters provided satisfactory fits to the experimental results with relatively small (⩽3.1) shift factors. Based on the FHWA’s APT test results and the VESYS5W analyses, two equations are recommended for prediction of fatigue cracking in pavements with conventional mix and with asphalt modifier.