Selection of asphalt mixtures based on the optimization of their mechanical and functional properties using the concept of solution dominance

This is the authors' abstract. We don't add key points for this paper.

Permeable pavements are one of the technical solutions within the stormwater management approach already applied by many countries, known as LID (Low Impact Development). By enabling exceptional functionalities, such as ease of infiltration, permeability, and water retention, among others, the main challenge during the design of permeable pavements is to simultaneously optimize their mechanical and functional properties. Specifically, regarding the surface layer of this type of pavement, porous asphalt mixtures are one of the most commonly used types of surfacing. The idea, therefore, is to obtain the best combination of aggregate gradation and binder content that can optimize the functional properties of the asphalt mixture, without neglecting the mechanical attributes imposed by technical standards. Thus, unequivocally, the problem can be diagnosed as a multiobjective optimization problem, and the therapy is to apply optimization methods in an attempt to solve it. In this context, the objective of this study was to formulate optimization problems involving, specifically, porous asphalt mixtures under different boundary conditions, such as traffic level, climate, temperature, and roadway surface drainage conditions, and to apply a search method based on the concept of solution dominance to select the most suitable mixtures for each condition within the available decision space. The aggregate gradation ranges and binder contents of the asphalt mixture were the decision variables considered for the optimization model, and the objective functions were the optimization of the mechanical and functional properties of interest for each scenario. The results demonstrated what is generally observed in multiobjective optimization problems, namely, that there is no single solution for this type of problem. Among the available solutions, mixtures analogous to those with a maximum aggregate size of 12.5 mm performed reasonably well in virtually all multiobjective problem scenarios conceived in this study. Although variations in binder content were observed, the CPA Range I mixture with a binder content of 5.0% stood out as the most recurrent optimal solution. In terms of the applied modeling, it was possible to conclude that the degree of dominance of a solution is not defined by the number of solutions it dominates, but by the number of times that solution is dominated.