Gaussian Process Regression (GPR) for Auto-Estimation of Resilient Modulus of Stabilized Base Materials

Document Type : Regular Article

Authors

1 Associate Professor, Department of Civil Engineering, Sirjan University of Technology, Sirjan, Iran

2 Research Assistant, Department of Civil Engineering, Sirjan University of Technology, Sirjan, Iran

3 School of Civil Engineering, Iran University of Science and Technology, Tehran, Iran

Abstract

The resilient modulus of different pavement materials is one of the most important parameters for the pavement design using the mechanistic-empirical (M-E) method. The resilient modulus is generally determined by a triaxial test, which is expensive and time-consuming and requires special laboratory facilities. This study aims to develop a model based on the Gaussian Process Regression (GPR) to predict the resilient modulus of stabilized base material with different additives under wetting-drying cycles. For this purpose, a laboratory dataset containing 704 records have been used. The input parameters were considered as the wetting-drying cycles, free lime to silica ratio, Alumina and iron oxide compounds in the additives, maximum dry density to optimum moisture content ratio, deviator stress, and confining stress. The results indicate high accuracy of the GPR method with a regression coefficient of 0.997 and 0.986 respectively for train and test data and 0.995 for all datasets. Comparing the developed model based on the GPR method with the developed models in the literature based on the artificial neural network methods and the support vector machines shows higher accuracy of the GPR method.

Highlights

Google Scholar | Scopus | PlumX Metrics | Mendeley

Keywords

Main Subjects

References

[1]     AASHTO. AASHTO Guide for Design of Pavement Structures. 1986.
[2]     Zaman M, Solanki P, Ebrahimi A, White L. Neural network modeling of resilient modulus using routine subgrade soil properties. Int J Geomech 2010;10:1–12. doi:10.1061/(ASCE)1532-3641(2010)10:1(1).
[3]     NCHRP. Guide for Mechanistic–Empirical Design of New and Rehabilitated Pavement Structures. Washington, DC United States: 2004. doi:Final Report for Project 1-37A.
[4]     Witczak MW, Qi X, Mirza MW. Use of nonlinear subgrade modulus in AASHTO design procedure. J Transp Eng 1995;121:273–82. doi:10.1061/(ASCE)0733-947X(1995)121:3(273).
[5]     Kim DS, Kweon GC, Rhee S. Alternative method of determining resilient modulus of subbase soils using a static triaxial test. Can Geotech J 2001;38:117–24. doi:10.1139/cgj-38-1-117.
[6]     Kim D-S, Stokoe KH. Characterization of resilient modulus of compacted subgrade soils using resonant column and torsional shear tests. Transporation Res Rec 1992;1307:90–8.
[7]     George KP. Resilient Testing of Soils Using Gyratory Testing Machine. Transp Res Rec J Transp Res Board 1992;1369:63–72.
[8]     George KP. Prediction of Resilient Modulus from Soil Index Properties. Rep FHWA/MS-DOT-RD-04-172 2004:72.
[9]     Ghanizadeh AR, Rahrovan M. Application of Artifitial Neural Network to Predict the Resilient Modulus of Stabilized Base Subjected to Wet Dry Cycles. Comput Mater Civ Eng 2016;1:37–47.
[10]    Solanki P. Artificial neural network models to estimate resilient modulus of cementitiously stabilized subgrade soils. Int J Pavement Res Technol 2013;6:155–64. doi:10.6135/ijprt.org.tw/2013.6(3).155.
[11]    Vadood M, Johari MS, Rahai A. Developing a hybrid artificial neural network-genetic algorithm model to predict resilient modulus of polypropylene/polyester fiber-reinforced asphalt concrete. J Text Inst 2015;106:1239–50. doi:10.1080/00405000.2014.985882.
[12]    Kim SH, Yang J, Jeong JH. Prediction of subgrade resilient modulus using artificial neural network. KSCE J Civ Eng 2014;18:1372–9. doi:10.1007/s12205-014-0316-6.
[13]    El-Ashwah AS, Mousa E, El-Badawy SM, Abo-Hashema MA. Advanced characterization of unbound granular materials for pavement structural design in Egypt. Int J Pavement Eng 2020:1–13. doi:10.1080/10298436.2020.1754416.
[14]    Amiri H, Nazarian S, Fernando E. Investigation of Impact of Moisture Variation on Response of Pavements through Small-Scale Models. J Mater Civ Eng 2009;21:553–60. doi:10.1061/(asce)0899-1561(2009)21:10(553).
[15]    Pal M, Deswal S. Extreme Learning Machine Based Modeling of Resilient Modulus of Subgrade Soils. Geotech Geol Eng 2014;32:287–96. doi:10.1007/s10706-013-9710-y.
[16]    Sadrossadat E, Heidaripanah A, Ghorbani B. Towards application of linear genetic programming for indirect estimation of the resilient modulus of pavements subgrade soils. Road Mater Pavement Des 2018;19:139–53. doi:10.1080/14680629.2016.1250665.
[17]    Ghanizadeh AR, Amlashi AT. Prediction of Fine-grained Soils Resilient Modulus using Hybrid ANN-PSO, SVM-PSO and ANFIS-PSO Methods. J Transp Eng 2018;9:159–82.
[18]    Heidarabadizadeh N, Ghanizadeh AR, Behnood A. Prediction of the resilient modulus of non-cohesive subgrade soils and unbound subbase materials using a hybrid support vector machine method and colliding bodies optimization algorithm. Constr Build Mater 2021;275:122140. doi:10.1016/j.conbuildmat.2020.122140.
[19]    AASHTO. Mechanistic-Empirical Pavement Design Guide. Washington, DC, USA: 2008. doi:10.1201/b17043-11.
[20]    Maalouf M, Khoury N, Laguros JG, Kumin H. Support vector regression to predict the performance of stabilized aggregate bases subject to wet-dry cycles. Int J Numer Anal Methods Geomech 2012;36:675–96. doi:10.1002/nag.1023.
[21]    Nunan TA, Humphrey DH. a Review and Experimentation of Gravel Stabilization Methods. Executive Summary. Tranportation Research Board, Washington, DC: 1990.
[22]    Berg KC. Durability and strength of activated reclaimed Iowa Class C fly ash aggregate in road bases. Iowa State University, 1998.
[23]    Zaman MM, Zhu JH, Laguros JG. Durability effects on resilient moduli of stabilized aggregate base. Transp Res Rec 1999:29–38. doi:10.3141/1687-04.
[24]    Khoury NN. Durability of Cementitiously Stabilized Aggregate Bases for Pavement Application. THE UNIVERSITY OF OKLAHOMA, 2005.
[25]    Filizzola F, Student CG, Edil TB, Chairman CHB, Camargo B. Strength and Stiffness of Recycled Base Materials Blended With Fly Ash. Washington, DC, USA: 2008.
[26]    Guthrie WS, Michener JE, Wilson BT, Eggett DL. Effects of environmental factors on construction of soil-cement pavement layers. Transp Res Rec 2009:71–9. doi:10.3141/2104-08.
[27]    George KP, Davidson DT. Development of a Freeze-Thaw Test for Design of Soil-Cement. Highw Res Rec 1963:77–96.
[28]    Miller GA, Zaman M. Field and laboratory evaluation of cement kiln dust as a soil stabilizer. Transp Res Rec 2000:25–32. doi:10.3141/1714-04.
[29]    Khoury NN, Zaman MM. Effect of wet-dry cycles on resilient modulus of class C coal fly ash-stabilized aggregate base. Transp Res Rec 2002:13–21. doi:10.3141/1787-02.
[30]    Khoury N, Zaman M. Durability effects on flexural behavior of fly ash stabilized limestone aggregate. J Test Eval 2006;34:167–75. doi:10.1520/jte14085.
[31]    Khoury, N., Zaman M. Influences of various cementitious agents on the performance of stabilized aggregate base subjected to wet dry cycles. Int J Pavement Eng 2007;8:265–76.
[32]    Kaloop MR, Kumar D, Samui P, Gabr AR, Hu JW, Jin X, et al. Particle Swarm Optimization algorithm-Extreme Learning Machine (PSO-ELM) model for predicting resilient modulus of stabilized aggregate bases. Appl Sci 2019;9. doi:10.3390/app9163221.
[33]    Wang J. An intuitive tutorial to Gaussian processes regression. ArXiv 2020.
[34]    Cheng MY, Huang CC, Roy AF Van. Predicting project success in construction using an evolutionary gaussian process inference model. J Civ Eng Manag 2013;19. doi:10.3846/13923730.2013.801919.
[35]    Omran BA, Chen Q, Jin R. Comparison of Data Mining Techniques for Predicting Compressive Strength of Environmentally Friendly Concrete. J Comput Civ Eng 2016;30:04016029. doi:10.1061/(ASCE)CP.1943-5487.0000596.
[36]    Pal M, Deswal S. Modelling pile capacity using Gaussian process regression. Comput Geotech 2010;37:942–7. doi:10.1016/j.compgeo.2010.07.012.
[37]    Rasmussen CE, Williams CKI. Gaussian Processes for Machine Learning. Adaptive Computation and Machine Learning series. Comput Sci 2005;3176:63–71.
[38]    Ebden M. Gaussian Processes: A Quick Introduction. ArXiv Preprint ArXiv: 2015.
[39]    Yang Y, Zhang Q. A hierarchical analysis for rock engineering using artificial neural networks. Rock Mech Rock Eng 1997;30:207–22. doi:10.1007/BF01045717.
[40]    Ghanizadeh AR, Abbaslou H, Amlashi AT, Alidoust P. Modeling of bentonite/sepiolite plastic concrete compressive strength using artificial neural network and support vector machine. Front Struct Civ Eng 2019;13:215–39. doi:10.1007/s11709-018-0489-z.

Files

Cited by

History

  • Receive Date: 18 January 2021
  • Revise Date: 05 March 2021
  • Accept Date: 17 March 2021

Share

How to cite

Statistics