Sensitivity Evaluation of MEPDG Performance Prediction

Abstract

This report documents and presents the results of a study to evaluate the sensitivity of pavement performance predicted by the Mechanistic-Empirical Pavement Design Guide to the values of the design inputs. Global sensitivity analyses were performed for five pavement types under five climate conditions and three traffic levels. Design inputs evaluated in the analyses included traffic volume, layer thicknesses, material properties (e.g., stiffness, strength, HMA and PCC mixture characteristics, subgrade type), groundwater depth, geometric parameters (e.g., lane width), and others. Detailed traffic inputs were not considered. Depending on the base case, approximately 25 to 35 design inputs were evaluated in the analyses. Correlations among design inputs (e.g., between PCC elastic modulus and modulus of rupture) were considered where appropriate. A normalized sensitivity index defined as the percentage change of predicted distress relative to its design limit caused by a given percentage change in the design input. The analyses found that, for all pavement types and distresses, the sensitivities of the design inputs for the bound surface layers were consistently the highest. Additional findings are also reported for each specific pavement type. Disciplines Civil and Environmental Engineering DISCLAIMERThis is an uncorrected draft as submitted by the research agency. The opinions and conclusions expressed or implied in the report are those of the research agency. They are not necessarily those of the Transportation Research Board, the National Academies, or the program sponsors.i Correlations among design inputs (e.g., between PCC elastic modulus and modulus of rupture) were considered where appropriate. A normalized sensitivity index defined as the percentage change of predicted distress relative to its design limit caused by a given percentage change in the design input. The analyses found that, for all pavement types and distresses, the sensitivities of the design inputs for the bound surface layers were consistently the highest. Additional findings are also reported for each specific pavement type. LIST OF TABLES 1 EXECUTIVE SUMMARYThe Mechanistic-Empirical Pavement Design Guide Manual of Practice (AASHTO, 2008) and related MEPDG software provide a new, more theoretically grounded methodology for the analysis and performance prediction of different types of flexible and rigid pavements. The cracking, rutting, faulting, smoothness, and other distresses predicted by the MEPDG for the anticipated climatic and traffic conditions will depend on the values of the input parameters that characterize the pavement materials, layers, design features, and condition. Knowledge of the sensitivity of predicted performance to the design input values can help identify, for specific climatic region and traffic conditions, the inputs that most influence predicted performance. This will help pavement designers determine where additional effort is justified in developing higher quality and/or more certain input values.Th...

Global Sensitivity Analysis of Jointed Plain Concrete Pavement Mechanistic–Empirical Performance Predictions

“…The rankings and µ ± 2σ values in Table 5 are judged to be the good measures of the MEPDG input sensitivities in the MEPDG. Graphical summaries of the input sensitivities by distress are available elsewhere (4).…”

“…This was pursued with a combination of insight from previous acceptable sensitivity studies and quantitative evaluations with OAT sensitivity analysis. Detailed procedures and results of MEPDG triage for JPCP inputs and OAT sensitivity analysis were given by Schwartz et al (4).…”

“…Load transfer efficiency and slab width were used to represent edge support inputs in MEPDG JPCP analyses. Details of these input considerations were given by Schwartz et al (4).…”

“…MEPDG sensitivity studies for rigid pavements began appearing in the literature immediately after the initial release of MEPDG in 2004 (4). However, previous MEPDG sensitivity studies were limited in scope, approach, and findings.…”

Sensitivity analysis frameworks for mechanistic-empirical pavement design of continuously reinforced concrete pavements

“…Detailed discussions by Schwartz et al [4] on procedures employed by previous studies and related findings indicate that the previous MEPDG sensitivity studies have been limited in scope, approach, and findings. These limitations include varying only small subsets of inputs, reliance on sensitivity analysis approaches without answering the question that "if input x goes up by n%, output y goes down by m%", neglecting any correlations and/or interactions among input parameters, fewer studies focusing on CRCP performance predictions, and analysis using earlier versions of the MEPDG software and models which are different from the latest version of MEPDG software (version 1.1) that form the main framework of AASHTOWare Pavement ME Design.…”