Since the identification of segregation on asphalt pavements has been based on visual observations around the world, it is essential to develop a quantitative method to detect the presence and severity of segregation. The nonnuclear density device was selected for the field measurements at nine test sites, and cores were taken for laboratory testing. The nonnuclear density difference between segregated and control areas was found to increase with increasing the segregation severity. For medium or heavy segregated areas, the nonnuclear density difference was more than 90 kg/m3. In medium or heavy segregated areas, the air voids increased by 20 % and the indirect tensile strength decreased by 8 % as compared to the control areas. The criterion to detect segregation was based on the statistical differences in nonnuclear measured density values with a p-value 0.05. If a statistical difference in nonnuclear density between segregated and control areas existed, the chance of aggregate gradation difference in percent passing the 3/8 in., No. 4 and No. 8 sieves for medium and heavy segregation was 78 and 64 %, respectively. In addition to the field nonnuclear density readings, the material properties of asphalt mixtures obtained from 108 cores, including texture depth and percent air voids were selected as independent variables to develop a model to predict the calibrated nonnuclear density difference between the segregated and control areas. This quantitative model holds a great promise as a tool to identify potential areas of segregation using both lab data and field nonnuclear density measurements for quality assurance purposes.