Estimation of the Sensitivity of Design Input Variables for the Mechanistic-Empirical Design Guide

Abstract

Many highway agencies use AASHTO methods for the design of pavement structures. Current AASHTO methods are based on empirical relationship s between traffic loading, materials, and pavement performance developed from the AASHO Road Test (1958-1961). The applicability of these methods to modern-day conditions has been questioned; in addition, the lack of realistic inputs regarding environmental and other factors in pavement design has caused concern. Research sponsored by th e National Cooperative Highway Research Program has resulted in the development o f a mechanistic-empirical design guide (M-E Design Guide) for pavement structural analysis. The new M-E Design Guide requires over 100 inputs to model traffic, environmental, materials, and pavement performance to provide estimates of pavement distress over the design life of the pavement. Many designers may lack specific knowledge of the data required. A study was performed to assess the relative sensitivity of the models used in the M-E Design Guide to inputs relating to construction materials in the analysis of flexible and rigid pavement structures. Inputs were evaluated by analyzing a standard pavement section and changing the value of each input individually, then assessing the change in predicted pavement distress (cracking, faulting, and roughness for rigid pavements; rutting, fatigue, and low-temperature cracking for flexible pavements). The evaluations may aid designers in focusing on those inputs having the most
effect on desired pavement performance.