Estimate of Fatigue Shift Factors between Laboratory Tests and Field Performance

Abstract

Mechanistic–empirical pavement design relies on empirical transfer functions to convert stresses and strains (from mechanistic analyses) to an allowable number of load repetitions until an acceptable limit of damage occurs. One method of determining transfer function for bottom-up fatigue cracking is to perform laboratory fatigue tests on the mixture or a range of potential mixtures intended for the pavement. Previous research estimated that shift factors on the order of 4 to 100 are needed to relate laboratory and field performance. There are a number of potential explanations for these shift factors, such as rest periods and healing. The structural test sections at the National Center for Asphalt Technology (NCAT) Test Track provided an opportunity to determine shift factors between transfer functions developed from laboratory beam fatigue tests and from the Guide for Mechanistic–Empirical Design of New and Rehabilitated Pavement Structures (MEPDG) and field performance. Strains input into the beam fatigue transfer function were measured using embedded pavement strain gauges and calculated using layered-elastic theory (PerRoad). The MEPDG was used to calculate strains for its transfer function. Fatigue shift factors were calculated for four of the structural sections from the 2003 NCAT Test Track. The fatigue shift factors determined using the measured strains ranged from 4.2 to 75.8. The fatigue shift factors determined with PerRoad ranged from 6.7 to 19.2. The fatigue equations developed from the laboratory testing were not used in the MEPDG; the NCHRP 1-37A-calibrated fatigue models were used. On the basis of these analyses, the MEPDG fatigue model reasonably predicts observed cracking.