Use of random forests regression for predicting IRI of asphalt pavements

Abstract Random forest is a powerful machine learning algorithm with demonstrated success. In this study, the authors developed a random forests regression (RFR) model to estimate the international roughness index (IRI) of flexible pavements from distress measurements, traffic, climatic, maintenance and structural data. To validate the model, more than 11,000 samples were collected from the database of long-term pavement performance (LTPP) program, with 80% randomly sampled data for training and 20% of them for testing the RFR model. The performance of the RFR model was then compared with that of the regularized linear regression model. The results showed that the RFR model significantly outperformed the linear regression model, with coefficients of determination ( R 2 ) greater than 0.95 in both the training and test sets. The variable importance score obtained from the RFR revealed that the initial IRI was the most important factor affecting the development of the IRI. In addition, the transverse cracking, fatigue cracking, rutting, annual average precipitation and service age had important influences on the IRI. Other distresses such as longitudinal cracking, edge cracking, aggregate polishing, and potholes exerted little impact on the evolution of the IRI.

[1]  Weng Tat Chan,et al.  Constraint handling methods in pavement maintenance programming , 2001 .

[2]  Charles E Dougan,et al.  Models Relating Pavement Quality Measures , 2004 .

[3]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[4]  Hongren Gong,et al.  Effectiveness Analyses of Flexible Pavement Preventive Maintenance Treatments with LTPP SPS-3 Experiment Data , 2016 .

[5]  Baoshan Huang,et al.  Local calibration of the fatigue cracking models in the Mechanistic-Empirical Pavement Design Guide for Tennessee , 2017 .

[6]  Danny X Xiao,et al.  Calibration of the Mechanistic–Empirical Pavement Design Guidefor Flexible Pavement Design in Arkansas , 2011 .

[7]  Teresa M Adams,et al.  Pavement Roughness Modeling Using Back‐Propagation Neural Networks , 2004 .

[8]  Samuel Labi,et al.  Evaluating the Cost Effectiveness of Flexible Rehabilitation Treatments Using Different Performance Criteria , 2009 .

[9]  Satish Chandra,et al.  Relationship between Pavement Roughness and Distress Parameters for Indian Highways , 2013 .

[10]  William D O Paterson.  A TRANSFERABLE CAUSAL MODEL FOR PREDICTING ROUGHNESS PROGRESSION IN FLEXIBLE PAVEMENTS , 1989 .

[11]  Philippe Nitsche,et al.  Comparison of Machine Learning Methods for Evaluating Pavement Roughness Based on Vehicle Response , 2014, J. Comput. Civ. Eng..

[12]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[13]  Gonzalo R. Rada,et al.  Relating Ride Quality and Structural Adequacy for Pavement Rehabilitation and Management Decisions , 2012 .

[14]  Yuhong Wang.  Ordinal Logistic Regression Model for Predicting AC Overlay Cracking , 2013 .

[15]  Jay N. Meegoda,et al.  Roughness Progression Model for Asphalt Pavements Using Long-Term Pavement Performance Data , 2014 .

[16]  Samuel Labi,et al.  Effectiveness of Highway Pavement Seal Coating Treatments , 2004 .

[17]  Samuel Labi,et al.  Part 1: Pavement Management: Analysis of Long-Term Effectiveness of Thin Hot-Mix Asphaltic Concrete Overlay Treatments , 2005 .

[18]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[19]  Kathleen T. Hall,et al.  Performance of Flexible Pavement Rehabilitation Treatments in the Long-Term Pavement Performance SPS-5 Experiment , 2003 .

[20]  Gaetano Bosurgi,et al.  An ANN model to correlate roughness and structural performance in asphalt pavements , 2017 .

[21]  Hehua Zhu,et al.  Optimal Thresholds for Pavement Preventive Maintenance Treatments Using LTPP Data , 2017 .

[22]  Xuedong Yan,et al.  Exploring precrash maneuvers using classification trees and random forests. , 2009, Accident; analysis and prevention.

[23]  Yuhong Wang.  The effects of using reclaimed asphalt pavements (RAP) on the long-term performance of asphalt concrete overlays , 2016 .

[24]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[25]  Samuel Labi,et al.  Aggregate and disaggregate statistical evaluation of the performance-based effectiveness of long-term pavement performance specific pavement study-5 (LTPP SPS-5) flexible pavement rehabilitation treatments , 2010 .

[26]  Kathleen T. Hall,et al.  Performance of Flexible Pavement Maintenance Treatments in the Long-Term Pavement Performance SPS-3 Experiment , 2003 .

[27]  T D Gillespie,et al.  THE INTERNATIONAL ROAD ROUGHNESS EXPERIMENT: A BASIS FOR ESTABLISHING A STANDARD SCALE FOR ROAD ROUGHNESS MEASUREMENTS , 1986 .

[28]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[29]  Mohamed Abdel-Aty,et al.  Assessing Safety on Dutch Freeways with Data from Infrastructure-Based Intelligent Transportation Systems , 2008 .

[30]  José Pablo Aguiar-Moya,et al.  Calibration of Mechanistic–Empirical Pavement Design Guide Permanent Deformation Models , 2009 .

[31]  George Wang,et al.  Impact of Climate Conditions on Effectiveness of Asphalt Pavement Preservation Techniques , 2012 .

[32]  Samuel Labi,et al.  Determining the Service Life of Thin Hot-Mix Asphalt Overlay by Means of Different Performance Indicators , 2009 .

[33]  Linda M Pierce,et al.  Practical Guide for Quality Management of Pavement Condition Data Collection , 2013 .

[34]  Baoshan Huang,et al.  Field performance evaluation of asphalt mixtures containing high percentage of RAP using LTPP data , 2018, Construction and Building Materials.