Modelling the Indirect Tensile Strength of Asphalt Mixtures with the Use of Artificial Intelligence

  • Ali Saleh Széchenyi István Egyetem, Építész-, Építő- és Közlekedésmérnöki Kar, Közlekedésépítési és Vízgazdálkodási Tanszék
  • László Gáspár Széchenyi István Egyetem, Építész-, Építő- és Közlekedésmérnöki Kar, Közlekedésépítési és Vízgazdálkodási Tanszék; KTI Magyar Közlekedéstudományi és Logisztikai Intézet Nonprofit Kft.,
Keywords: foamed bitumen, warm mix asphalt, neural network, Support Vector Regression (SVR), Machine learning

Abstract

The authors estimate the indirect tensile strength of asphalt mixtures containing recycled asphalt and foamed bitumen using linear regression and neural network models. By comparing the random forest and the neural network model, the applicability of machine learning techniques in this field was proven. In the course of the research work, three models were developed, which are able with a high R2 value to predict the relationship between the ITS (wet and dry) value and two factors affecting it, namely the foamed bitumen content and the Reclaimed Asphalt Pavement (%).

Published
2024-08-01
How to Cite
SalehA., & GáspárL. (2024). Modelling the Indirect Tensile Strength of Asphalt Mixtures with the Use of Artificial Intelligence. Scientific Review of Transport, 74(4), 8-23. https://doi.org/10.24228/KTSZ.2024.4.1
Section
Cikkek