Framework for Development of an Improved Unbound Aggregate Base Rutting Model for Mechanistic–Empirical Pavement Design

Abstract

This paper presents findings from an ongoing research study at the University of Illinois that aims to develop and calibrate improved models for unbound aggregate rutting through laboratory characterization of aggregate materials used for unbound base and subbase applications in the state of North Carolina. Extensive triaxial laboratory testing was performed to establish a robust link between the number of load applications, stress levels, shear stress and sheer strength ratios, and permanent deformation responses. A framework was established for considering the strong correlation that commonly exists between permanent deformation and shear strength characteristics, as opposed to resilient modulus properties, in the laboratory characterization of the permanent deformation behavior of various types of aggregate materials. Trends of permanent strain accumulations from repeated load triaxial tests were adequately captured in a new rutting model whose development took into account the shear stresses applied at 25%, 50%, and 75% of the shear strength properties of these materials under similar field loading confinement conditions. The research shows that this model is an improvement on the rutting damage model for unbound aggregate currently used in AASHTO's mechanistic–empirical pavement design approach because it offers better material characterization and rutting prediction of the unbound base or subbase layer.

Dense-graded aggregate base gradation influencing rutting model predictions

“…Gidel et al [24] also proposed a stress dependent permanent deformation model based on the laboratory data. Recently, Chow et al [25][26][27] proposed a framework for predicting permanent deformation (also known as the UIUC Rutting Model) as a function of applied wheel load stress levels and aggregate shear strength under applied confinement (or ratio of the two defined as the shear stress ratio) along with the number of load applications.…”

“…The research described in this paper builds on the Chow et al [25][26][27] framework study with the goal to develop and calibrate a new rutting model to better predict rutting performance of in-service unbound aggregate pavement layers. Sixteen different aggregate materials commonly used in the state of North Carolina for pavement base/subbase applications were tested in the laboratory for shear strength and permanent deformation behavior.…”

Mechanistic-empirical approach to characterizing permanent deformation of reinforced soft soil subgrade

“…In summary, in order to estimate the subgrade permanent deformation using Equation (3), one would need the following inputs and calibration factors: 1) Material properties: water content (W c ), elastic modulus E r ; 2) Thickness of the layer or sublayer: h; 3) Outputs from the response model: average vertical strain (3 v ) or vertical elastic strain (3 v ) at the mid-depth of the layer or sublayer; 4) Parameters associated with the layer stiffness: b 1 and b 9 ; and 5) Calibration factors: b cal . It should be noted that the NCHRP permanent deformation model for unbound pavement materials does not account for the influence of stress state, although studies have shown the stress state is one of the important factors affecting the permanent deformation (Lekarp et al, 2000;Chow et al, 2014).…”

Mechanistic-empirical models for better consideration of subgrade and unbound layers influence on pavement performance

“…IRI decreases with the increase of base modulus (Masad and Little 2004) Cross-anisotropy affects total rutting and cracking, which leads to the change of IRI (Masad and Little, 2004) High shear strength results in low IRI values (AASHTO, 2008;Chow et al, 2014) Permanent deformation of unbound base is a major distress resulting in increase of surface roughness (Zhou et al 2007) Change of IRI diminishes with increase of thickness of the base layer (Masad and Little 2004) 8 (Cary and Zapata 2011;Sahin et al 2013;Gu et al 2016a) High shear strength prevents occurrence of transverse cracking (Cleveland et al 2002;Lytton et al 2010 Resistance to loadrelated cracking would be enhanced with increase of modulus of subgrade (Gupta et al 2007;Shahji 2006;Schwartz et al 2013) N/A Lower permanent deformation of subgrade reduces the probability of load-related cracking (Oh et al 2007) Thermal…”

“… Resilient modulus models of unbound layers and subgrade 1) Empirical regression models (AASHTO 1993;ARA, 2004) 2) Nonlinear stress-dependent models (Seed et al 1967;Hicks and Monismith 1971;Thompson and Robnett 1979;Drumm, 1990;Uzan 1985;Witczak and Uzan 1988;Witczak 2003;Lade and Nelson 1987) 3) Moisture-sensitive models (AASHTO 2008) 4) Moisture-sensitive and stress-dependent models (Oloo and Fredlund 1998;Lytton et al 1993;Lytton 1995;Sahin et al 2013;Yang et al 2005;Liang et al 2008;Cary and Zapata 2011;Gupta et al 2007;Oh and Fernando 2011) 5) Stress-dependent and cross-anisotropic models (Al-Qadi et al 2010;Tutumluer and Thompson 1997) 6) Moisture-sensitive, stress-dependent, and cross-anisotropic model (Gu et al 2016a) 7) Regression models for stress-dependent model coefficients (Yau and Von Quintus 2002) 6 8) Regression models for moisture-sensitive and stress-dependent model coefficients (Gu et al 2015b)  Permanent deformation models of unbound layers and subgrade 1) Non-stress-dependent mechanistic-empirical models (Kenis 1977;Uzan 2004;Tseng and Lytton 1989;Ayres and Witczak 1998) 2) Stress-dependent mechanistic-empirical models (ARA, 2004;Uzan 2004;Korkiala-Tanttu 2009;Chow et al 2014;Gu et al 2015a) 3) Regression models for Pavement ME Design model coefficients (Tseng and Lytton 1989;ARA, 2...…”