Electronic Theses and Dissertations

Date of Award


Document Type


Degree Name

Ph.D. in Engineering Science

First Advisor

Hakan Yasarer

Second Advisor

Yacoub Najjar

Third Advisor

Hunain Alkhateb


University of Mississippi

Relational Format



Transportation infrastructures account for a considerable portion of public investments, which serve as the backbone of a country’s economy by providing essential services to businesses and people. In the United States, public investments in transportation infrastructure assets represent trillions of dollars. The U.S road network consists of about 4 million miles, being the world’s largest, longest, and biggest transportation system. Paved roads account for 2.6 million miles, and 93% of them are surfaced with asphalt. However, a portion of the paved roads consists of asphalt overlaid concrete pavements, also known as composite pavements. When concrete pavements start to fail, they are overlaid with Hot Mix Asphalt (HMA). Compared to flexible or rigid pavements, this offers better performance measures both structurally and functionally, and accordingly, it can be considered a cost-effective alternative.Several performance indicators have been used to assess pavement surface conditions, but the Pavement Condition Rating (PCR) and the International Roughness Index (IRI) are the most widely used and well-recognized pavement performance indicators. Transportation agencies use these indexes to evaluate and classify the conditions for the road networks in the long term. If maintenance and rehabilitation (M&R) interventions are not performed timely, the pavement damage caused by environmental impacts and traffic repetitions can lead the roads to early deterioration. Billions of dollars are spent every year on M&R. However, a shortage in federal and state funds led roads and bridges to poor conditions since M&R interventions were not carried out timely. Therefore, there is a need to develop pavement performance prediction models that can support and allow decision-makers to prioritize M&R actions due to the limited budget allocation and estimate the rate of pavement deterioration. Traditionally, linear, non-linear, multiple linear regression analysis, Markov chains, mechanistic-empirical relations, survivor curves, semi-Markov, and Bayesian models have been used for predicting pavement performance. However, simple statistical approaches do not account for the complex relations among input variables and pavement performance. A growing body of literature is exploring the use of more advanced modeling techniques for pavement performance prediction. Among these techniques, the Artificial Neural Networks (ANNs) approach has shown the most significant improvements with consistent and reliable results. However, most performance models did not consider M&R history in the model development. This doctoral research presents new pavement performance models incorporating the M&R history and activities for composite pavements of the LTPP database. Additionally, a more comprehensive approach was developed for flexible, rigid, and composite pavements of the Mississippi Department of Transportation (MDOT) database, accounting for the influence of M&R history. This dissertation successfully utilized the ANNs modeling technique to obtain accurate and promising prediction results for pavement performance. Furthermore, the development of a simple, low-cost, and easy-access graphical user interface (GUI) tool brings a significant contribution to the enhancement of agencies' pavement management system (PMS) by predicting future pavement conditions, identifying rehabilitation needs, and allowing a better budget allocation for critical pavement sections without the need of distress data.


Civil Engineering



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.