• Marta Vila-Cortavitarte
  • Daniel Jato-Espino
  • Amir Tabakovic
  • Daniel Castro-Fresno

Self-healing within asphalt pavements is the process whereby road cracks can be repaired automatically when thermal and mechanical conditions are met. To accelerate and improve this healing process, metal particles are added to asphalt mixtures. However, this approach is costly both in economic and environmental terms due to the use of virgin metallic particles. So, even though the self-healing of asphalt mixtures has been widely addressed in experimental terms over the years, there is a lack of research aimed at modelling this phenomenon, especially with the purpose of optimizing the use of metal particles through the valorization of industrial by-products. As such, the goal of this study was to develop a statistical methodology to model the healing capacity of asphalt concrete mixtures (AC-16) from the characteristics of the metal particles added and the time and intensity used for magnetic induction. Five metal particles were used as heating inductors, including four types of industrial by-products aimed at transforming waste products into material for use in the road sector. The proposed approach consisted of a combination of cluster algorithms, multiple regression analysis and response optimization, which were applied to model laboratory data obtained after testing asphalt concrete mixtures containing these inductors. The results proved the accuracy of the statistical methods used to reproduce the experimental behaviour of the asphalt mixtures, which enabled the authors to determine the optimal amount of industrial by-products and time needed to make the self-healing process as efficient as possible.

Original languageEnglish
Article number116715
Number of pages11
JournalConstruction and Building Materials
Volume228
DOIs
Publication statusPublished - 2019

    Research areas

  • Asphalt mixtures, Cluster analysis, Industrial by-products, Multiple regression analysis, Response optimization, Self-healing, Waste valorization

ID: 56485043