Use of probabilistic machine learning for assessing rutting in asphalt mixtures
DOI:
https://doi.org/10.58922/transportes.v34.e3165Keywords:
Asphalt mixtures; Permanent deformation; Probabilistic machine learning.Abstract
Permanent deformation is a distress that affects the durability and quality of asphalt pavements, being influenced by the properties of the binder and aggregates in asphalt mixtures used on surface courses. The uniaxial repeated load test, which provides the Flow Number, is used to evaluate the mixture rutting resistance. An alternative to this test is the use of machine learning algorithms, an approach that enables the prediction of the traffic level equivalent to the Flow Number using data from various parameters of asphalt mixtures. Another option is the use of probabilistic models, which consider the uncertainty of the predicted variables. Among these, Gaussian Processes stand out, excelling with small amounts of data. The study herein aims to assess the performance of various probabilistic models in classifying asphalt mixtures concerning permanent deformation. The goal is to assist in material selection and pavement design. A database, containing 252 mixtures, was constructed by gathering information from various laboratory tests. Using regression and classification models to predict the equivalent traffic level, the results obtained demonstrate the feasibility of this approach, achieving an accuracy of up to 90%.
Downloads
References
Ahmed, T. M., P. L. Green, & H. A. Khalid (2017). Predicting fatigue performance of hot mix asphalt using artificial neural networks. Road Materials and Pavement Design 18, 141–154. DOI:10.1080/14680629.2017.1306928.
Apeagyei, A. K. (2011). Rutting as a function of dynamic modulus and gradation. Journal of Materials in Civil Engineering 23(9), 1302–1310. DOI:10.1061/(ASCE)MT.1943-5533.0000309.
Barros, L. M. (2022). Implementação do Ensaio Stress Sweep Rutting e do Shift Model para a Previsão da Deformação Permanente de Misturas Asfálticas Brasileiras. Doutorado (tese), Universidade Federal do Rio de Janeiro, Rio de Janeiro. URL: http://hdl.handle.net/11422/28334 [visitado 13/08/2025].
Bastos, J. B. S., J. B. Soares, & L. A. H. Nascimento (2017). Critérios para os resultados do ensaio uniaxial de carga repetida de misturas asfálticas em laboratório a partir do desempenho em campo. Transportes 25(2), 29–40. DOI:10.14295/transportes. v25i2.1284.
Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Cambridge: Springer.
Carvalho, P. H. J. (2025). Coleta, estruturação e análise de um banco de dados de ensaios sobre misturas asfálticas e seus componentes. Monografia (bacharelado), Universidade Federal do Ceará, Crateús. URL: http://repositorio.ufc.br/ handle/riufc/80054 [visitado 13/08/2025].
DNIT (2018). DNIT 184/2018-ME: Pavimentacão - Misturas Asfálticas - Ensaio Uniaxial de Carga Repetida para Determinação da Resistência à Deformação Permanente – Método de Ensaio. Rio de Janeiro: Departamento Nacional de Infraestrutura de Transportes (DNIT).
El-Badawy, S., R. A. El-Hakim, & A. Awed (2018). Comparing artificial neural networks with regression models for hot-mix asphalt dynamic modulus prediction. Journal of Materials in Civil Engineering 30(7), 04018128. DOI:10.1061/(ASCE)MT. 1943-5533.0002282.
Faccin, C. (2018). Concretos asfálticos em utilização no Rio Grande do Sul: Comportamento mecânico e desempenho em campo quanto à deformação permanente. Mestrado (dissertação), Universidade Federal de Santa Maria, Santa Maria. URL: http://repositorio.ufsm.br/handle/1/15888 [visitado 13/08/2025].
Ferreira, J., L. Babadopulos, J. Bastos, & J. Soares (2020). A tool to design rutting resistant asphalt mixes through aggregate gradation selection. Construction & Building Materials 236, 117531. DOI:10.1016/j.conbuildmat.2019.117531.
Gandomi, A. H., A. H. Alavi, M. R. Mirzahosseini, & F. M. Nejad (2011). Nonlinear genetic-based models for prediction of flow number of asphalt mixtures. Journal of Materials in Civil Engineering 23(3), 248–263. DOI:10.1061/(ASCE)MT.1943-5533.0000154.
Gardner, J. R., G. Pleiss, D. Bindel, K. Q. Weinberger, & A. G. Wilson (2018). GPyTorch: Blackbox matrix-matrix Gaussian process inference with GPU acceleration. In Advances in Neural Information Processing Systems. arXiv:1809.11165. DOI:10.48550/arXiv.1809.11165.
Gomes, O. J. F., J. B. Soares, & J. B. S. Bastos (2019). Relação das diferentes escalas da mistura asfáltica com a sua resistência à deformação permanente. In 20º Congresso Ibero-Latino-Americano do Asfalto (CILA), Guadalajara, México. CILA.
Kim, D. (2015). Modulus and Permanent Deformation Characterization of Asphalt Mixtures and Pavements. Doutorado (tese), North Carolina State University, Raleigh. URL: http://www.lib.ncsu.edu/resolver/1840.16/10587 [visitado 13/08/2025].
Liu, J., F. Liu, Z. Wang, E. O. Fanijo, & L. Wang (2023). Involving prediction of dynamic modulus in asphalt mix design with machine learning and mechanical-empirical analysis. Construction & Building Materials 407, 133610. DOI:10.1016/j. conbuildmat.2023.133610.
Mariano, A. L. G. (2023). Uso de aprendizado de máquina interpretável para avaliação da deformação permanente em misturas asfálticas. Mestrado (dissertação), Universidade Federal do Ceará, Fortaleza. URL: http://repositorio.ufc.br/handle/ riufc/75193 [visitado 13/08/2025].
McKinney, W. (2010). Data structures for statistical computing in Python. In Proceedings of the 9th Python in Science Conference, Austin, TX, pp. 56–61. SciPy Proceedings. DOI:10.25080/Majora-92bf1922-00a.
Milios, D., R. Camoriano, P. Michiardi, L. Rosasco, & M. Filippone (2018). Dirichlet-based Gaussian processes for large-scale calibrated classification. In Advances in Neural Information Processing Systems. arXiv:1805.10915. DOI:10.48550/arXiv. 1805.10915.
Mirzahosseini, M., Y. M. Najjar, A. H. Alavi, & A. H. Gandomi (2015). Next-generation models for evaluation of the flow number of asphalt mixtures. International Journal of Geomechanics 15(6), 04015009. DOI:10.1061/(ASCE)GM.1943-5622.0000483.
Murphy, K. P. (2012). Machine Learning: A Probabilistic Perspective. Cambridge: The MIT Press. Murphy, K. P. (2022). Probabilistic Machine Learning: An Introduction. Cambridge: The MIT Press.
Onofre, F. C. (2012). Avaliação do comportamento mecânico de misturas asfálticas produzidas com ligantes asfálticos modificados por ácido polifosfórico e aditivos poliméricos, enfatizando a resistência à deformação. Mestrado (dissertação), Universidade Federal do Ceará, Fortaleza. URL: http://repositorio.ufc.br/handle/riufc/11166 [visitado 13/08/2025].
Ouni, A. E., A. Dony, & J. Colin (2014). Probabilistic parametric approach for rutting evaluation: Application to hot and warm asphalt. The International Journal of Pavement Engineering 15(1), 58–65. DOI:10.1080/10298436.2012.725473.
Ozturk, H. I. & M. E. Kutay (2014). An artificial neural network model for virtual Superpave asphalt mixture design. The International Journal of Pavement Engineering 15(2), 151–162. DOI:10.1080/10298436.2013.808341.
Pedregosa, F., G. Varoquaux, A. Gramfort, V. Michel, & B. Thirion (2011). Scikit-learn: machine learning in Python. Journal of Machine Learning Research 12, 2825–2830. URL: https://www.jmlr.org/papers/v12/pedregosa11a.html [visitado 21/5/2026].
Portela Neto, M. G. (2018). Avaliação da faixa de agregados dominantes e componentes intersticiais em misturas asfálticas. Mestrado (dissertação), Universidade Federal do Ceará, Fortaleza. URL: http://repositorio.ufc.br/handle/riufc/49901 [visitado 13/08/2025].
Singh, P. & A. K. Swamy (2020). Probabilistic approach to characterise laboratory rutting behaviour of asphalt concrete mixtures. The International Journal of Pavement Engineering 21(3), 384–396. DOI:10.1080/10298436.2018.1480780.
Titsias, M. K. (2009). Variational learning of inducing variables in sparse Gaussian processes. Journal of Machine Learning Research 5, 567–574.
Williams, C. K. I. & C. E. Rasmussen (2006). Gaussian Processes for Machine Learning. Cambridge: The MIT Press. Witczak, M. W. (2007). Specification Criteria for Simple Performance Tests for Rutting, Volume II: Flow Number and Flow
Time. NCHRP Report 580, Transportation Research Board, Washington, D.C. URL: https://trid.trb.org/View/844079 [visitado 21/5/2026].
Zhao, Y., K. Zhang, Y. Zhang, Y. Luo, & S. Wang (2022). Prediction of air voids of asphalt layers by intelligent algorithm.
Construction & Building Materials 317, 125908. DOI:10.1016/j.conbuildmat.2021.125908.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Pedro Luiz Ribeiro Rocha, Jorge Barbosa Soares, Iuri Sidney Bessa, César Lincoln Cavalcante Mattos, Carlos David Rodrigues Melo
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who submit papers for publication by TRANSPORTES agree to the following terms:
- The authors retain the copyright and grant Transportes the right of first publication of the manuscript, without any financial charge, and waive any other remuneration for its publication by ANPET.
- Upon publication by Transportes, the manuscript is automatically licensed under the Creative Commons License CC BY 4.0 license. This license permits the work to be shared with proper attribution to the authors and its original publication in this journal.
- Authors are authorized to enter into additional separate contracts for the non-exclusive distribution of the version of the manuscript published in this journal (e.g., publishing in an institutional repository or as a book chapter), with recognition of the initial publication in this journal, provided that such a contract does not imply an endorsement of the content of the manuscript or the new medium by ANPET.
- Authors are permitted and encouraged to publish and distribute their work online (e.g., in institutional repositories or on their personal websites) after the editorial process is complete. As Transportes provides open access to all published issues, authors are encouraged to use links to the DOI of their article in these cases.
- Authors confirm they have obtained all required employer permissions for the publication and CC BY 4.0 licensing of the manuscript, particularly if the employer holds any claim to the manuscript's copyright. Authors assume all responsibility for employer-related copyright issues, releasing ANPET and Transportes from any related liability.
- Authors assume full responsibility for the content of the manuscript, including the necessary and appropriate authorizations for the disclosure of collected data and obtained results, releasing ANPET and Transportes from any responsibility in this regard.
Last update: 11/27/2025
