Comparison of Equivalent Single-Axle Loads from Empirical and Mechanistic-Empirical Approaches

Empirical design methods such as the AASHTO’93 express pavement life in terms of the number of Equivalent Single Axle Loads (ESALs). Although the AASHTO’93 Guide still provides tables for calculating ESALs, soon after the first guide was published in 1972, the fourth power law was established as a simplification to convert mixed traffic into equivalent axle loads. Numerous posterior studies have recognized, however, that there is no a unique power value and it varies with pavement type, distress considered, failure level, contact stresses, etc. In order to address some of these dependences, an in-depth study is conducted to evaluate the existing load equivalency factors and establish optimal power values with the aid of the Mechanistic-Empirical Design Guide, which is developed under NCHRP 1-37A (herein referred to as the M-E Guide). The two most common failure criteria of flexible pavements, surface rutting and fatigue cracking, are considered. Axle load spectra from actual weigh-in-motion data are used. A variety of pavement structures with focus on varying surface layer thicknesses and subgrade materials are modeled and analyzed. The paper found that the ESALs failure determined by the fourth power law (herein referred to as empirical ESALs) generally agree well with pavement damage caused by a similar number of 18 kips single axles (herein referred to as mechanistic ESALs) when fatigue cracking is considered. However, the differences are very significant in the case of surface rutting. Under both distress types, the optimal power values vary with the thickness of the asphalt surface layer and the resilient modulus of the subgrade. In addition, it is found the optimal power values are smaller for rutting failure as compared with fatigue cracking. The results indicate that the optimal power values for rutting are significantly sensitive to the change of axle load spectra while the sensitivity for the case of fatigue cracking is less pronounced. Finally, higher power values do not necessarily mean higher number of ESALs when the power law is applied. Contrary to general expectation, when actual axle load spectra are considered, power values between 3 and 4 tend to produce the lower number of ESALs.