Parametric Study of Pavement Deterioration Using Machine Learning Algorithms

Pavement performance models play a vital role in any pavement management system. The Regional Government of Biscay (RGB) (Spain) manages a 1200 km road network and conducts pavement data collections, including the International Roughness Index (IRI) values. The aim of the paper is to develop an IRI performance model for two-lane roads with flexible pavement until the first maintenance and/or rehabilitation activity is performed. Due to the huge amount of available information, a deterministic model was selected. A literature review of deterministic models showed that, apart from age and traffic volumes, the pavement structure is a key factor. Therefore, it was decided to analyze the only road stretches whose entire pavement section was known (surface layer + base + subbase). Various variables related to age, traffic volumes and employed materials were introduced as possible factors. The multiple linear regression model with the highest coefficient of determination and all the variables significant included the real pavement age, the cumulated heavy traffic and the total thickness of bituminous layers. As the material employed in the surface layer could affect roughness progression, a qualitative variable was introduced to consider various surface materials. The model improved its accuracy, indicating that the surface layer material is also an influencing factor on IRI evolution.

https://www.mdpi.com/2079-6412/10/2/97/pdf

IRI Performance Models for Flexible Pavements in Two-Lane Roads until First Maintenance and/or Rehabilitation Work

Prediction of the remaining service life (RSL) of pavement is a challenging task for road maintenance and transportation engineering. The prediction of the RSL estimates the time that a major repair or reconstruction becomes essential. The conventional approach to predict RSL involves using non-destructive tests. These tests, in addition to being costly, interfere with traffic flow and compromise operational safety. In this paper, surface distresses of pavement are used to estimate the RSL to address the aforementioned challenges. To implement the proposed theory, 105 flexible pavement segments are considered. For each pavement segment, the type, severity, and extent of surface damage and the pavement condition index (PCI) were determined. The pavement RSL was then estimated using non-destructive tests include falling weight deflectometer (FWD) and ground-penetrating radar (GPR). After completing the dataset, the modeling was conducted to predict RSL using three techniques include support vector regression (SVR), support vector regression optimized by the fruit fly optimization algorithm (SVR-FOA), and gene expression programming (GEP). All three techniques estimated the RSL of the pavement by selecting the PCI as input. The correlation coefficient (CC), Nash–Sutcliffe efficiency (NSE), scattered index (SI), and Willmott’s index of agreement (WI) criteria were used to examine the performance of the three techniques adopted in this study. In the end, it was found that GEP with values of 0.874, 0.598, 0.601, and 0.807 for CC, SI, NSE, and WI criteria, respectively, had the highest accuracy in predicting the RSL of pavement.

Comparative Analysis of Machine Learning Models for Prediction of Remaining Service Life of Flexible Pavement

Pavement performance models, a vital part of pavement management systems and life-cycle analysis, are generally divided into deterministic and probabilistic ones. Among probabilistic models, the Markov chains are attracting great attention. Transition probability matrices were developed for flexible pavement road network of the Republic of Moldova using the IRI values collected twice a year, in spring and in autumn, from 2013 to 2015. Consequently, a half-year cycle time was established. The aim of this paper is to demonstrate that it is feasible to develop transition probability matrices for an entire flexible pavement network using data from a short data collection period, and simultaneously carrying out maintenance and rehabilitation activities, if some assumptions are made. Results showed that road sections can drop two or three states in one cycle time, not only remaining in the same state or evolving to the next one, as it is usually assumed in pavement performance modeling. These models are proposed for countries in similar circumstances; a network with no new roads constructed in last decades, pavements maintained or rehabilitated in different moments during their service life, invalid or useless pavement condition data from previous years and unknown pavement structure in most of the sections.

Transition Probability Matrices for Flexible Pavement Deterioration Models with Half-Year Cycle Time

Due to the extension of road networks and the complexity of problems in various parts of the pavement management system (PMS), the traditional system and classical techniques cannot create an effective system. Accordingly, the application of automatic and expert systems in PMS is unavoidable to create a more efficient and optimal system. The application of computational intelligence (CI) in PMS leads to solve a complex problem and create more expert and optimal systems. This chapter, first, presents a brief explanation of CI frameworks (artificial neural network, fuzzy logic, evolutionary computation, swarm intelligence, and hybrid method), then provides a big-picture from CI frameworks and techniques. Also, it describes the methodology of the newest and more efficient techniques in the different applications of CI, such as learning from the data, solving the problem with uncertainty, and optimization. Finally, it provides a comprehensive view of the CI applications in a different part of PMS.

Computational intelligence for modeling of asphalt pavement surface distress

The objective of this paper was to develop probabilistic models based on the transition probability matrices obtained through knowledge and experience from those responsible for highway management ...

Development of probabilistic models for predicting roughness in asphalt pavement

The objective of this paper was to develop fuzzy models for asphalt pavement performance. The fuzzy logic can convert linguistic or qualitative variables into quantitative values. This feature makes it possible to gather experts’ experience about the knowledge they have on factors that affect the pavement performance and its state condition. Forms developed in an organized way were applied for acquiring the knowledge from experts on pavement construction and maintenance. The variables pavement age, traffic, International Roughness Index (IRI) and Flexible Pavement Condition Index (FPCI) were associated with numerical scales and linguistic concepts such as new, old, light, heavy, good, fair, and poor. From the information obtained through the application of forms, variables were modeled with the aid of software InFuzzy and fuzzy models were developed for IRI and FPCI. For validating the model, a straight line adjustment was used to relate the predicted to the observed data. Also, the corresponding correlation coefficient (r) was calculated and residuals were analyzed. The developed models fitted to observed data and correlation coefficient r = 0.71 and 0.70, respectively.

/pdf/development-of-fuzzy-models-for-asphalt-pavement-performance-37krf1vyrz.pdf

Igor Castro Sá de Oliveira

Papers

Development of probabilistic models for predicting roughness in asphalt pavement

Development of fuzzy models for asphalt pavement performance