Loading...

Evaluation of the Efficiency of Flexible Pavement Maintenance and Rehabilitation Activities Using the Roughness Index

Karbalaei Poorsafari, Mahdi | 2024

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58108 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Tabatabaee, Nader
  7. Abstract:
  8. Determining the effect of maintenance and repair activities on pavement roughness, the service life of pavement sections, and the economic efficiency of these activities are crucial aspects in selecting the appropriate maintenance and repair strategies of pavements. Developing a suitable model for predicting pavement performance is a prerequisite for addressing this issue. In this study, a prediction model for the International Roughness Index (IRI) of flexible pavements at the network level was developed based on long-term pavement performance data using machine learning algorithms. Two models were developed; one for predicting the IRI of pavement sections without maintenance, and another for sections that underwent rehabilitation. The significance of input data was assessed using the p-value test, while the accuracy of the models was evaluated using the Root Mean Square Error (RMSE) and the coefficient of determination (R²). Among the machine learning algorithms, the Gradient Boosting Machine algorithm demonstrated the highest accuracy in predicting the IRI compared to other machine learning models. It was also found that the predicted IRI for rehabilitated pavements showed the highest correlation with the measured IRI and the time elapsed since the last rehabilitation. Furthermore, the effect of maintenance activities, including preventive maintenance and asphalt overlays, on IRI changes after rehabilitation was analyzed for several pavement sections, and the added service life of these sections was calculated. For this purpose, data from the Long-Term Pavement Performance (LTPP) database were used, and maintenance activities were classified into eight categories to assess their effect on pavement sections. Depending on the overlay thickness and the milling status before rehabilitation, asphalt overlays led to reduction of IRI from 37% to 51%. Additionally, the service life of pavement sections increased from 9.2 to 15.4 years. Milling before overlay application resulted in reduction of IRI by an average of 16%, and for thin overlays, milling before rehabilitation had a more significant effect on reducing the IRI index after rehabilitation
  9. Keywords:
  10. International Roughness Index ; Pavement Performance ; Preventive Maintenance ; Machine Learning ; Root Mean Squared Error ; Pavement Mnagement System ; Pavement Maintenance and Rehabilitation

 Digital Object List

 Bookmark

No TOC