Modeling and Analysis of Pavements with Asphalt Reinforcement: Development of a new computational model

Pavement systems are important for ensuring social and economic equality. These systems constantly deteriorate under the pressure of heavy traffic and weather effects, making regular maintenance or rehabilitation necessary. However, these activities challenge societies as they are economically expensive, disruptive to traffic, and require the use of natural resources. With the marked growth in global shipping, the evolution of new types of vehicles, and rapid climate change effects, these societal challenges are expected to aggravate. Accordingly, there is a need to improve the traditional pavement design and rehabilitation methods to maintain an acceptable service level of the overall road network. To meet these challenges, asphalt grid reinforcement (AGR) products have emerged as a possible solution. Previous studies have shown that AGR products can potentially improve pavement performance and prolong service life. Nonetheless, no mechanistic-empirical design method currently applies to asphalt pavements with AGR, which is needed to promote an AGR alternative.

Motivated by this need, the current PhD study focused on developing a mechanistic component (which is concerned with response evaluation) that engineers can use to subsequently quantify pavement performance. Two study objectives were outlined in the dissertation: (i) to develop a versatile, useful, and validated computational model capable of incorporating AGR effects in asphalt pavement systems and (ii) to gain valuable insights into the effects of AGR.

To achieve this, a model formulation was developed by extending the standard linear elastic theory framework to include: fragmented layers, imperfect interface bonding, viscoelasticity, moving loads, and AGR. The AGR model-component was formulated as a combination of three contributions: (i) including AGR as a thin high-modulus elastic layer, (ii) accounting for the AGR’s influence on interface bonding between adjacent layers, and (iii) capturing the AGR’s impact on the surrounding asphalt concrete properties. In total, seven modeling inputs were required to represent an AGR, and the formulation
was implemented into the computational code GRIDPAVE-MM. Several experimental campaigns were carried out for subsequent model validation of GRIDPAVE. These campaigns involved full-scale testing at a newly built test facility called DTU Smart Road. Here embedded sensors were used to collect pavement responses triggered by vehicle loadings. The collected field responses were compared against model simulations to showcase GRIDPAVE’s ability to match embedded sensor readings.

The main study findings highlighted that the developed model GRIDPAVE was able to capture the effects of installing AGR products inside an asphalt pavement. Furthermore, general model simulations indicated that adding reinforcement can potentially delay the development of cracks and ruts in road systems, hence improving the overall service life.

In conclusion, the PhD study provides a versatile, useful, and valid tool that can enable engineers, consultants, and contractors (etc.) to include AGR in asphalt pavement design. Specifically, the model can produce traffic-induced stresses, strains, and displacements in various asphalt pavement systems with AGR, which can subsequently be used to evaluate the AGR effects on pavement performance. Ultimately, the study opens up new opportunities for using AGR as a sustainable and robust solution for maintaining an acceptable pavement service level for future road systems.