An alternative approach to obtaining AASHTO soil classification using ann

Authors

DOI:

https://doi.org/10.14295/transportes.v29i1.2176

Keywords:

AASHTO classification, Low volume roads, Perceptron, Soil survey

Abstract

The soil survey and laboratory tests to analyze the soil general rating as subgrade of roads using the AASHTO classification, usually have a high financial cost for roads projects, in Ceará state the cost of geotechnics services for pavement design is estimated at 30%. An alternative way to identify preliminarily a soil’s qualities rapidly just with field soil analysis would be positive to paving. The aims of this paper are an artificial neural network framework that processes qualitative field test data to prediction AASHTO soil classification. The data of the Visual-Manual classification of soils, which makes it possible to verify preliminary the particle size and color of the material, were used as explanatory variables. Thus, was created a database with 1790 soil samples, which were extracted from pre-existing projects, provided by National Department of Transportation Infrastructure and Department of Transportation of Ceará state. The proposed model presented an accuracy rate of 94.5%, in the average of the estimates for the AASHTO classification, and an error of the order of 0.04, considered the mean square of errors (MSE).

Downloads

Download data is not yet available.

References

AASHTO (2004) Mechanistic-Empirical pavement design guide: of new and rehabilitated pavement structures – final report. Washington D.C.: American Association of State Highway and Transportation Officials.

AASHTO (2008) Mechanistic-Empirical pavement design guide-a manual of practice. Washington D.C.: American Association of State Highway and Transportation Officials.

ABNT (1984) NBR 7181: Solo - Análise Granulométrica. Rio de Janeiro.: Associação Brasileira de Normas Técnicas.

ABNT (2016) NBR 7182: Solo – Ensaio de compactação. Rio de Janeiro.: Associação Brasileira de Normas Técnicas.

ABNT (2016) NBR 7180: Solo – Determinação do limite de plasticidade. Rio de Janeiro.: Associação Brasileira de Normas Técnicas.

ABNT (2016) NBR 9895: Solo - Índice de suporte Califórnia (ISC) - Método de ensaio. Rio de Janeiro.: Associação Brasileira de Normas Técnicas.

Alawi, M. e M. Rajab (2013) Prediction of Califórnia bearing ratio of sub base layer using multiple linear regression models. Road Materials and Pavement Design, v. 14, n. 1, p. 211-2019. DOI: 10.1080/14680629.2012.757557.

ASTM D2488-00 (2000) Standard Practice for Description and Identification of Soils (Visual-Manual Procedure). West Conshohocken, PA.: American Society for Testing and Materials.

Barroso, S. H. A. (2002) Estudo dos solos da região metropolitana de Fortaleza para aplicação na engenharia rodoviária. Tese (Doutorado). Programa de Pós-graduação em Engenharia de Transportes, Universidade de São Paulo. São Paulo. DOI:10.11606/T.18.2016.tde-01062016-112348.

Bhargavi, P. e J. Jyothi (2009) Applying naive bayes data mining technique for

classification of agricultural land soils, IJCSNS International Journal of Computer Science and Network Security, v. 9, n. 8, p.117–122.

Bhargavi, P. e S. Jyothi (2010) Soil classification using GATREE. International Journal of Computer Science & Information Tech-nology, v. 2, n. 5, p. 184-191. DOI: 10.5121/ijcsit.2010.2514.

Boruvka, L. e V. Penizek (2007) A test of an artificial neural network allocation procedure using the czech soil Survey of agri-cultural land data. In Lagacherie, P.; A.B. Mcbratney e M. Voltz (eds). Digital Soil mapping: an introductory perspective. Am-sterdam: Elsevier, p. 415-424. DOI: 10.1007/978-1-4020-8592-5_1.

Chaves, F. J. (2000) Caracterização geotécnica de solos da formação barreiras da região metropolitana de fortaleza para aplica-ção em obras rodoviárias. Dissertação (Mestrado). Programa de Pós-graduação em Engenharia de Transportes, Universi-dade Federal do Rio de Janeiro. Rio de janeiro.

Cybenko, G. (1989) Approximation by superposition of a sigmoidal function. Matemática. Control Signal Systems, v. 2, n. 4, p. 303-314.

Erzin, Y. e D. Turkoz (2016) Use of neural networks for the prediction of the CBR value of some aegean sands. Neural Compu-ting and Applications, v. 27, n. 5, p. 1415–1426. DOI: 10.1007/s00521-015-1943-7.

Guilherme, A. T. P. (2016) Análise comparativa de modelos para previsão e mapeamento de propriedades geotécnicas dos solos da microrregião de Mossoró-RN. Dissertação (Mestrado). Programa de Pós-graduação em Engenharia de Transportes, Uni-versidade Federal do Ceará. Fortaleza. Disponível em: <http://www.repositorio.ufc.br/handle/riufc/23174> (Acesso em 25/07/2019).

Gunaydin, O.; A. Gokoglu e M. Fener (2010) Prediction of artificial soil’s unconfined compression strength test using statisti-cal analyses and artificial neural networks. Advances in Engineering Software, v. 41, p. 1115-1123. DOI: 10.1016/j.advengsoft.2010.06.008.

Haykin, S.O. (2001) Neural Networks, A Comprehensive Foundation (2ª ed.). Ontario: Pearson Education.

Johari, A.; A. A. Javadi e G. Habibagahi (2011) Modelling the mechanical behaviour of unsaturated soils using a genetic algo-rithm-based neural network. Computers and Geotechnics, v. 38, n. 1, p. 2-13. DOI: 10.1016/j.compgeo.2010.08.011.

Kaniu, M. I. e K. H. Angeyo (2015) Challenges in rapid soil quality assessment and opportunities presented by multivariate chemometric energy dispersive X-ray fluorescence and scattering spectroscopy. Geoderma, v. 241–242, p. 32-40. DOI: 10.1016/j.geoderma.2014.10.014.

Kayadelen, C.; O. Gunaydin.; M. Fener.; A. Demir e A. Ozvan (2009) Modeling of the angle of shearing Resistance of soils using soft computing systems. Expert Systems With Applications, v. 36, n. 9, p. 11814-11826.DOI: 10.1016/j.eswa.2009.04.008.

Mascarenhas, I. M. N. (2016) Caracterização geotécnica de solos na região metropolitana do Cariri/CE para uso em pavimenta-ção. Dissertação (Mestrado), Programa de Pós-graduação em Engenharia de Transportes, Universidade Federal do Cea-rá. Fortaleza. Disponível em: <http://www.repositorio.ufc.br/handle/riufc/49825> (Acesso em 27/07/2019).

Minke, G. (2006) Building with earth: design and technology of a sustainable architecture (2ª ed.). Basel: Birkhaeuser.

Mollahasani. A.; A. H. Alavi.; A. H. Gandomi e A. Rashed (2010) Nonlinear neural-based modeling of soil cohesion intercept. Ksce Journal of Civil Engineering, v. 15, n. 5, p. 831-840. DOI: 10.1007/s12205-011-1154-4.

Nazzal, M. D. e O. Tatari (2013) Evaluating the use of neural networks and genetic algorithms for prediction of subgrade resilient modulus. International Journal of Pavement Engineering, v. 14, n. 4, p. 364–373. DOI: 10.1080/10298436.2012.671944.

Neto, L. B.; A. C. C. F. Sieira.; B. R. Danziger e J. G. S. Silva (2006) Neuro-CPT: classificação de solos usando-se Redes Neurais Artificiais. Engevista, v. 8, n. 1, p. 37-48. DOI: 10.22409/engevista.v8i1.170.

Neto, S. A. D.; M. V. Silveira.; L. B. Amâncio e G. J. M. Anjos (2014) Pile settlement modeling with multilayer perceptrons. Elec-tronic Journal of Geotechnical Engineering, v. 19, p. 4517-4518.

Nguyen, B. T. e A. Mohajerani (2017) Possible estimation of resilient modulus of fine grained soils using a dynamic light-weight cone penetrometer. International Journal of Pavement Engineering.v. 18, n. 6 p. 473-484. DOI: 10.1080/10298436.2015.1095899.

Park, H. I.; G. C. Kweon e S. R. Lee (2009) Prediction of resilient modulus of granular subgrade soils and subbase materials using artificial neural network. Road Materials and Pavement Design, v. 10, n. 3, p. 647- 665. DOI:10.1080/14680629.2009.9690218.

Pavini, M. (2002) Artificial neural network modeling of pavement performance using expert judgement. Road Material and Pavement Design, v. 3, n. 4, p. 373-384. DOI: 10.1080/14680629.2002.9689931.

Pinto, C.S. (2006) Curso básico de mecânica dos solos em 16 aulas (3ª ed.). São Paulo: Oficina textos.

Rakesh, N.; A. K. Jain.; M. Amaranatha Reddy e K. Sudhakar Reddy (2006) Artificial neural networks- genetic algorithm based model for backcalculation of pavement layer moduli. International Journal of Pavement Engineering, v. 7, n. 3, p. 221-230. DOI: 10.1080/10298430500495113.

Ribeiro, A. J. A.; C. A. U. Da Silva e S. H. A. Barroso (2015) Neural Estimation of Localization and Classification of Soils for Use in Low-Traffic- Volume Roads. Transportation Research Record, v. 2473 n. 2473, p. 98-106, 2015. DOI:10.3141/2473-12.

Ribeiro, A. J. A.; C. A. U. Da Silva e S. H. A. Barroso (2018) Metodologia de baixo custo para mapeamento geotécnico aplicado à pavimentação. Transportes (Rio de Janeiro), v. 26, n. 2, p. 84-100. DOI:10.14295/transportes.v26i2.1491.

Rigassi, V. (1985) Compressed earth blocks: manual of production (Vol I). Eschborn: Vieweg.

Sadrossadat, E.; H. Ali e B. Ghorbani (2018) Towards application of linear genetic programming for Indirect estimation of the resilient modulus of pavements subgrade soils. Road Materials and Pavement Design, v. 19, n. 1, p. 139-153. DOI: 10.1080/14680629.2016.1250665.

Sadrossadat, E.; H. Ali e O. Saeedeh (2016) Prediction of the resilient modulus of flexible pavement Subgrade soils using adaptive neuro-fuzzy inference systems. Construction and Building Materials, v. 123, p. 235-247. DOI: 10.1016/j.conbuildmat.2016.07.008.

Sitton, J. D.; Y. Zeinali e B.A. Story (2017) Rapid soil classification using artificial neural networks for use in constructing com-pressed earth blocks. Construction and Building Materials, v. 138, p. 214-221. DOI: 10.1016/j.conbuildmat.2017.02.006.

Taskiran, T. (2010) Prediction of California bearing ratio (CBR) of fine grained soils by AI methods. Advances in Engineering Software, v. 41, n. 6, p. 886-892. DOI: 10.1016/j.advengsoft.2010.01.003.

Tenpe, A. R. e A. Patel (2018) Application of genetic expression programming and artificial neural network for prediction of CBR. Road materials and pavement design, v. 19, n. 1, p. 1-18. DOI: 10.1080/14680629.2018.1544924.

Yildirim, B. e O. Gunaydin (2011) Estimation of California bearing ratio by using soft computing systems. Expert Systems with Applications, v. 38, p. 6381-6391. DOI: 10.1016/j.eswa.2010.12.054.

Zhang, H. e T. Yu (2016) Prediction of subgrade elastic moduli in different seasons based on BP neural network technology. Road Materials and Pavement Design, v. 19, n. 8, p. 1-18. DOI: 10.1080/14680629.2016.1259122.

Zeghal, M. e W. Khogali (2005) Predicting the resilient modulus of unbound granular materials by neural networks. In Na-tional Research Council Canada (NRCC-47704). Trondheim : BCRA, 27-29, p. 1-9.

Published

2021-04-30

How to Cite

de Souza, W. M., Alves Ribeiro, A. J., & da Silva, C. A. U. (2021). An alternative approach to obtaining AASHTO soil classification using ann. TRANSPORTES, 29(1), 41–54. https://doi.org/10.14295/transportes.v29i1.2176

Issue

Section

Artigos