Backpropagation Neural Network to estimate pavement performance: dealing with measurement errors

The objective of this study is to apply the Backpropagation Neural (BPN) network with Generalised Delta Rule learning algorithm for reducing the measurement errors of pavement performance modelling. The Multi-Layer Perceptron network and sigmoid activation function are applied to build the BPN network of Pavement Condition Index (PCI). Collector and arterial roads of both flexible and rigid pavements in Montreal City are taken as a case study. The input variables of the PCI are Average Annual Daily Traffic (AADT), Equivalent Single Axle Loads (ESALs), Structural Number (SN), pavement age, slab thickness and difference in the PCI between the current and preceding year (ΔPCI). The BPN networks estimate that the PCI has inverse relationships with AADT, ESALs and pavement age. The PCI has positive relationships with these variables for roads that have recent treatment operations. The PCI has positive relationships with SN and slab thickness that imply the increase in pavement condition with increasing structural strength and slab thickness. The ΔPCI significantly influences the estimation of PCI values. The AADT and ESALs have considerable importance; however, pavement age and structural characteristics of the pavement have an insignificant influence in determining the PCI values, except in the case of flexible arterial roads.

[1]  J M Jansen,et al.  PERFORMANCE MODELS AND PREDICTION OF INCREASE IN OVERLAY NEED IN THE DANISH STATE HIGHWAY PAVEMENT MANAGEMENT SYSTEM, BELMAN , 1994 .

[2]  Kathryn A Cation,et al.  PERFORMANCE PREDICTION DEVELOPMENT USING THREE INDEXES FOR NORTH DAKOTA PAVEMENT MANAGEMENT SYSTEM , 1992 .

[3]  N O Attoh-Okine,et al.  PREDICTING ROUGHNESS PROGRESSION IN FLEXIBLE PAVEMENTS USING ARTIFICIAL NEURAL NETWORKS , 1994 .

[4]  A. O. Abd El Halim,et al.  The influence of age on interlayer shear properties , 2015 .

[5]  H Von Quintus,et al.  DESIGN PAMPHLET FOR THE DETERMINATION OF LAYERED ELASTIC MODULI FOR FLEXIBLE PAVEMENT DESIGN IN SUPPORT OF THE 1993 AASHTO GUIDE FOR THE DESIGN OF PAVEMENT STRUCTURES , 1997 .

[6]  Serdal Terzi,et al.  Modeling the pavement serviceability ratio of flexible highway pavements by artificial neural networks , 2007 .

[7]  Ralph Haas,et al.  Investigation of Relationship Between Deterministic and Probabilistic Prediction Models in Pavement Management , 1997 .

[8]  David M. Skapura,et al.  Neural networks - algorithms, applications, and programming techniques , 1991, Computation and neural systems series.

[9]  Nii O. Attoh-Okine,et al.  Analysis of learning rate and momentum term in backpropagation neural network algorithm trained to predict pavement performance , 1999 .

[10]  Thomas J. Van Dam,et al.  Proposed Pavement Performance Models for Local Government Agencies in Michigan , 2000 .

[11]  H Jamsa.  MUNICIPAL PAVEMENT MANAGEMENT SYSTEM , 1988 .

[12]  Jemal H. Abawajy,et al.  An Accelerated Particle Swarm Optimization Based Levenberg Marquardt Back Propagation Algorithm , 2014, ICONIP.

[13]  Moshe Ben-Akiva,et al.  Infrastructure management under uncertainty: Latent performance approach , 1993 .

[14]  A R Shekharan.  Assessment of Relative Contribution of Input Variables to Pavement Performance Prediction by Artificial Neural Networks , 1999 .

[15]  Ralph Haas,et al.  RELIABILITY-BASED PROCESSING OF MARKOV CHAINS FOR MODELING PAVEMENT NETWORK DETERIORATION , 1996 .

[16]  Md. Shohel Reza Amin.  The Pavement Performance Modeling: Deterministic vs. Stochastic Approaches , 2015 .

[17]  S. Emerson,et al.  AASHTO (American Association of State Highway and Transportation Officials). 2001. A Policy on Geometric Design of Highways and Streets. Fourth Edition. Washington, D.C. , 2007 .

[18]  Kelvin C. P. Wang,et al.  Probabilistic Behavior of Pavements , 1994 .

[19]  Litao Liu,et al.  Bayesian Model for Predicting the Performance of Pavements Treated with Thin HMA Overlays , 2014 .

[20]  Alan Cheetham,et al.  Enhancement of Arizona Pavement Management System for Construction and Maintenance Activities , 2006 .

[21]  Mohamed M El-Basyouny,et al.  Probabilistic Performance-Related Specifications Methodology Based on Mechanistic–Empirical Pavement Design Guide , 2010 .

[22]  Michael S Mamlouk,et al.  Mechanistic Roughness Model Based on Vehicle-Pavement Interaction , 2000 .

[23]  Suleiman Ashur,et al.  Macroscopic optimum system for management of pavement rehabilitation , 2001 .

[24]  K P George,et al.  MODELS FOR PREDICTING PAVEMENT DETERIORATION , 1989 .

[25]  Luis Amador-Jimenez,et al.  Pavement management with dynamic traffic and artificial neural network: a case study of Montreal , 2016 .

[26]  F Finn.  PAVEMENT MANAGEMENT SYSTEMS--PAST, PRESENT, AND FUTURE , 1998 .

[27]  Timo Hartmann,et al.  Prediction of IRI in short and long terms for flexible pavements: ANN and GMDH methods , 2016 .

[28]  K. Sudhakar Reddy,et al.  Artificial neural networks—genetic algorithm based model for backcalculation of pavement layer moduli , 2006 .