Investigation of viscoelastic behaviour and permanent deformation modelling for New Zealand hot mix asphalts.

Permanent deformation is considered as one of the major modes of failures of flexible pavements in New Zealand. Currently based on the Australian mechanisticempirical pavement design procedure, rutting is modelled solely based on the subgrade compressive strain criterion. However, the assumptions and the rationale behind the subgrade compressive strain criterion are questionable. Therefore, the main objective of this study is to characterize the permanent deformation behaviour of Hot Mix Asphalt (HMA) based on the extensive laboratory tests to find more accurate correlation with filed performance of flexible pavement. In this study, asphalt mixes were tested by Simple Performance Test (SPT) method which includes Dynamic Modulus and Repeated Creep tests. The effect of deviator stress, temperature, air void content, mix gradation and binder type were thoroughly investigated through a factorial analysis. It was found that the temperature has leading effect on mix behaviour followed by pressure, air void and binder type. Permanent to resilient strain model was proposed for the New Zealand asphalt mixes. The study proved that, models proposed elsewhere could not be directly adopted in New Zealand due to significant differences in mix design and material properties. Indirect Tensile test (IDT) was also considered in this research to study bituminous mixture modulus based on New Zealand and Australian design procedure. It was shown that, under accurate conversion backed up by viscoelastic principles, the dynamic modulus of the material could be predicted from the Indirect Tensile Modulus (IDT) counterpart. As a result, it would be feasible for the New Zealand industry to adopt Mechanistic-Empirical Pavement Design method by simply using the current IDT test setup. Resilient modulus then could be converted to the dynamic modulus equivalent. Finally, Wheel Tracking (WT) test was conducted to evaluate the validity of the SPT results regarding the permanent deformation. A total of 200 cylindrical and 30 slab specimens were fabricated over the course of this research. A modified test setup was proposed for the current wheel tracker test. More realistic responses under the new test setup were recorded as a result. In addition, the material behaviour was simulated through finite element modelling by using ABAQUS program. In order to do so, the Static Creep Test was performed for a selected number of specimens. It was found that the “Creep εodel” could closely simulate the mix performance.