Experimental evaluation of fiber-reinforced porous asphalt mixtures through the design of experiments integrated with multi-criteria decision-making analysis

Abstract

Porous asphalt mixture (PA) is a type of mixture characterized by having large air voids content, with aggregates of high quality and a small number of fines, commonly used as wearing course in pavement structures due to the various benefits that this mixture provides in terms of safety and environmental care. In general, PA mixture allows the water drainage through its structure, improving the skid resistance and mitigating noise pollution. Despite the multiple functional benefits, the durability of this mixture is lower in comparison to dense-graded asphalt mixtures. In Europe and some countries of America, the use of polymer-modified binder (PMB) is quite common to improve the durability of PA mixture. However, other additives such as synthetic fibers could be presented as an innovate alternative to increase the overall performance of PA mixture. The following study investigated the effect of adding synthetic fibers on the performance properties of PA mixtures employing an innovative methodology which combines the Design of Experiment (DOE) concept with the Multi-Criteria Decision-Making (MCDM) analysis. In the first phase of the research, a complete laboratory assessment plan was developed to evaluate the effectiveness of adding polyolefin-aramid (POA) and Polyacrylonitrile (PAN) synthetic fibers in PA mixtures in terms of functionality and mechanical performance. As the bituminous mortar plays a dominant role in the cohesive forces of the mix to keep the aggregates together, the effect of synthetic fibers at asphalt mortar scale was also analyzed. Once the most promising fiber was selected, in a second phase of the study, a novel methodology that integrates the Taguchi DOE concept with the CRITIC-TOPSIS hybrid MCDM analysis was proposed to evaluate the impact of fibers taking into account other control factors such as the binder type, the fiber content and the binder content. Among the most relevant results found in the research is that both POA and PAN fibers increase the raveling resistance and indirect tensile strength in dry conditions. However, the higher mechanical performance was provided by the addition of POA fibers. At the asphalt mortar scale, significant improvement of strength at low temperature (-15°C) was observed when adding 0.3% of POA and PAN fibers. Concerning the novel DOE-MCDM method proposed, the main effects plot for means obtained from the Taguchi method, allowed to determine the proper levels of the control factors for the different responses carried out. As multiple responses were obtained, the MCDM analysis allowed to transform the multi-response optimization problem into single-one response optimization problem. CRITIC was contemplated as an objective weighting approach whereas TOPSIS method contributed to identify the optimal combination of the control factors and the preference ranking among the experimental designs. From the results obtained, POA fibers acted very well as a stabilizer agent and as reinforcement since they reduced the particle loss in dry conditions. In the same way, the modified binder increased the raveling resistance in both dry and wet conditions without affecting the functional performance of the mixture and without presenting a risk of binder drain down. Although the first positions of the order of preference refers to experiments with mixes using polymer modified binder, admissible results can be also obtained using a conventional binder as long as the proper proportions of fibers are applied. The integration of DoE techniques and MCDM analysis can be considered a powerful tool for the evaluation of the impact of different control variables on different responses.