# Frontiers of Structural and Civil Engineering

 Front. Struct. Civ. Eng.    2020, Vol. 14 Issue (5) : 1083-1096     https://doi.org/10.1007/s11709-020-0654-z
 TRANSDISCIPLINARY INSIGHT
Estimation of flexible pavement structural capacity using machine learning techniques
1. Civil Engineering Department, Shahrood University of Technology, Shahrood 3619995161, Iran
2. Department of Civil Engineering, Ferdowsi University of Mashhad, Mashhad 9177948974, Iran
3. Department of Elite Relations with Industries, Khorasan Construction Engineering Organization, Mashhad 9185816744, Iran
4. Department for Management of Science and Technology Development, Ton Duc Thang University, Ho Chi Minh City, Vietnam
5. Faculty of Information Technology, Ton Duc Thang University, Ho Chi Minh City, Vietnam
 Download: PDF(1993 KB)   HTML Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks
 Abstract The most common index for representing structural condition of the pavement is the structural number. The current procedure for determining structural numbers involves utilizing falling weight deflectometer and ground-penetrating radar tests, recording pavement surface deflections, and analyzing recorded deflections by back-calculation manners. This procedure has two drawbacks: falling weight deflectometer and ground-penetrating radar are expensive tests; back-calculation ways has some inherent shortcomings compared to exact methods as they adopt a trial and error approach. In this study, three machine learning methods entitled Gaussian process regression, M5P model tree, and random forest used for the prediction of structural numbers in flexible pavements. Dataset of this paper is related to 759 flexible pavement sections at Semnan and Khuzestan provinces in Iran and includes “structural number” as output and “surface deflections and surface temperature” as inputs. The accuracy of results was examined based on three criteria of R, MAE, and RMSE. Among the methods employed in this paper, random forest is the most accurate as it yields the best values for above criteria (R=0.841, MAE=0.592, and RMSE=0.760). The proposed method does not require to use ground penetrating radar test, which in turn reduce costs and work difficulty. Using machine learning methods instead of back-calculation improves the calculation process quality and accuracy. Corresponding Author(s): Hosein GHASEMZADEH TEHRANI,Shahaboddin SHAMSHIRBAND Just Accepted Date: 28 July 2020   Online First Date: 14 September 2020    Issue Date: 16 November 2020
 Tab.1  Summary of SNeff prediction models Tab.2  Deflection bowl parameters [40] Tab.3  Statistical characteristics of the study dataset Fig.1  Distributions of the study variables: a) D0, b) D20, c) D30, d) D45, e) D60, f) D90, g) D120, h) D150, i) D180, j) Surfacetemp, and k) SNeff. Tab.4  Tests of normality for SNeff Tab.5  Spearman’s correlation coefficient between input variables and SNeff Tab.6  Target and seven input categories selected for prediction Tab.7  Performance metrics values for all seven input categories in GPR, M5P, and RF methods Fig.2  Observed and predicted ?values of SNeff for GPR-1, M5P-2, and RF-1 models. Fig.3  Scatter plots of observed ??and predicted SNeff values for GPR-1, M5P-2, and RF-1 models. Fig.4  Taylor diagrams of predicted SN values.