damental experimental method to characterize rutting performance of asphalt mixtures. Permanent deformation is defined as a combination of densifica- tion (volume change) and repetitive shear deformation (plastic flow without volume change). There are three stages of the permanent deformation: primary (initial consolidation), secondary (constant strain rate), and tertiary (shear deformation). The cycle number at which tertiary flow starts is referred to as the flow number (FN) (1). The relationship between permanent strain and the number of load cycles is generally expressed by a power law model: where  p = accumulated permanent strain, N = number of load cycles, and a, b = intercept and slope on log–log plot. This power law model has been widely used to describe the rutting behavior of asphalt mixtures. However, it can only represent the sec- ondary stage of permanent deformation and ignore the tertiary zone of the shear deformation. In addition, the model coefficients strongly depend on the data range selected to fit the model (5). The FN has been considered as an indicator of rutting performance of asphalt mixtures (4). In general, a higher FN value suggests better rutting performance. Although some research (1, 4) has indicated that the FN was correlated well with field rutting performance, there are some concerns regarding the FN. Since the FN represents the number of loading cycles required to initiate the tertiary flow, it may not be able to identify rutting resistance at the tertiary stage. More important, it is not an engineering property of asphalt mixtures. A comprehensive study (6), including laboratory test and finite ele- ment analysis, concluded that shape distortion induced by plastic flow was the dominant contribution to rutting. Therefore, shear properties, especially shear strength, play an important role in the development of permanent deformation for asphalt mixtures and could potentially be incorporated into a rutting prediction model. Goodman et al. (7 ) found that a high degree of correlation existed between shear strength and rut depth. Hajj et al. (8) observed that there was a close relation- ship between cumulative strain and the factor of safety, which is basi- cally the reciprocal of stress to strength ratio. All these studies gave great insights on the effects of the shear properties on the permanent deformation of asphalt mixtures. However, the studies were limited due to either complicated testing equipment (7 ) or relatively small database size based on a single mixture (8). The main objectives of this study are to evaluate the effects of the FN and shear properties on the rutting performance and to develop a  p b aN = () 1 Characterization of Permanent Deformation of Asphalt Mixtures Based on Shear Properties Qiang Li, Hyun Jong Lee, and Eui Yoon Hwang 1 A permanent deformation model was developed in this study on the basis of shear properties of asphalt mixtures, such as cohesion and friction angle. Two types of laboratory tests—triaxial compressive strength test and repeated load permanent deformation test—for three types of asphalt mixtures with various volumetric properties were performed to correlate the shear properties with rutting performance. A rational approach was proposed to determine flow number (FN) more accurately with the Hoerl model and roughness penalty method. It was observed from the test results that the FN was not accurate enough to differentiate the rut- ting characteristics of asphalt mixtures. Unlike the FN, the shear stress to strength ratio accurately identified the mixture rutting performance. Because the same value of the shear stress to strength ratio produced the same level of permanent strain regardless of mixture type, the ratio could be used as the rutting criterion for a performance-based mix design. A permanent deformation model was established on the basis of the shear stress to strength ratio. The proposed model is able to successfully pre- dict the permanent deformation of the various mixtures without chang- ing the model coefficients and could predict the permanent deformation all the way up to failure, including the tertiary flow. It is widely accepted that there is a need for mechanical property tests based on field performance to complement the Superpave ® volumet- ric mix design method. Permanent deformation, one of the most impor- tant distress types, should be evaluated during the asphalt mix design process. Recently, NCHRP Project 9-19 was initiated to recommend a simple performance test (SPT) that can describe the rutting behav- ior of asphalt mixtures (1). The three candidates of SPT for evalu- ating rutting potential are static creep permanent deformation (SCPD) test, repeated load permanent deformation (RLPD) test, and dynamic modulus ⎟ E * ⎟ test. Previous researchers (2, 3) observed that ⎟ E * ⎟ was not an accu- rate indicator for predicting mixture behavior at high temperature since it was incapable of capturing characteristics of permanent deformation, which comprised large strain and displacement phe- nomena. In addition, more emphasis has been placed on the RLPD test than the SCPD test because it offers a better simulation of field loading conditions (4). Thus the RLPD test is considered as a fun- Department of Civil and Environmental Engineering, Sejong University, 98 Kunja-dong, Kwangin-Gu, Seoul 143-747, South Korea. Corresponding author: H. J. Lee, hlee@sejong.ac.kr. Transportation Research Record: Journal of the Transportation Research Board, No. 2181, Transportation Research Board of the National Academies, Washington, D.C., 2010, pp. 1–10. DOI: 10.3141/2181-01