School of Civil and Environmental Engineering

Ph.D. Thesis Defense Announcement

Optimizing Project-level Pavement Asset Management: Predictive and Precision-based Maintenance with 3D Pavement Surface Data

By Ryan Salameh

Advisor:

Dr. Yichang (James) Tsai

Committee Members: Dr. Adjo A. Amekudzi-Kennedy (CEE), Dr. Baabak Ashuri (BC), Dr. John E. Taylor (CEE), Dr. Georgene Geary (GGfGA Engineering), Mr. Andrew Mergenmeier (FHWA)

Date and Time: March 24, 2025. 3 PM EST

Location: SEB 122

ABSTRACT
The increasing demand for U.S. roadway networks places significant strain on pavement infrastructure, particularly as it deteriorates due to aging, environmental stressors, and heavy traffic loads. Rising funding levels remain inadequate to keep pace with the accelerating deterioration of the nation's pavement conditions. Jointed Plain Concrete Pavement (JPCP) surfaces are commonly used on critical routes such as interstates and freight corridors due to their durability and long lifespan. However, as they age, these pavements require timely maintenance and rehabilitation (M&R) to preserve their structural integrity and minimize disruptions from unexpected deterioration and unplanned maintenance. To overcome these challenges, transportation agencies must transition from reactive maintenance approaches to predictive and precision-based asset management strategies. Most state highway agencies have adopted 3D laser imaging technology for automated, high-resolution pavement distress surveying. This survey data is primarily used for network-level condition assessments and planning, where distress information is aggregated at segment-level resolutions (e.g., 1-mile sections). This dissertation proposes a JPCP slab-based, project-level asset management framework that leverages high-resolution distress data collected over multiple survey cycles to enable precise condition monitoring and predictive maintenance decision-making.
This research introduces a JPCP slab-based condition monitoring system designed to enhance project-level pavement management. The system includes a structured framework for storing and managing slab- and joint-level condition data, a slab-based registration methodology to track individual slab performance across multiple surveys, a replaced slab classification and identification method to assess maintenance interventions, and an as-built slab layout reconstruction method to monitor the performance of the original pavement. A spatial-temporal deterioration analysis is conducted using six years of high-resolution slab- and joint-level distress data collected for a 10-mile JPCP project. The analysis provides insights into faulting behavior, capturing its variability across individual transverse joints along the segment and characterizing joint-based faulting deterioration trends. For slab-based cracking, findings reveal significant deterioration variability across 1-mile segments with similar characteristics (e.g., design, age, traffic, and environmental conditions). Markov Chain analysis highlights that different slab cracking states exhibit distinct transition rates and expected lifetimes. Additionally, several factors that increase the risk of slab failure due to cracking are identified, including longer slab length, steep road grades, higher joint faulting, proximity to a cracked slab, and deteriorated adjacent slab conditions. To support data-driven decision-making, a predictive M&R framework is developed, integrating an enhanced life-cycle cost analysis (LCCA) to evaluate the long-term cost-effectiveness of various strategies. A machine learning-based model is introduced to predict slab replacement needs using high-resolution condition data. A feasibility assessment with limited data demonstrates strong potential for accurately forecasting future maintenance demands. Finally, this dissertation presents a novel methodology for automating Concrete Pavement Restoration (CPR) pre-construction planning and design, extending the application of 3D pavement surface data into the construction phase. This approach leverages slab-based inventory and distress condition data to optimize localized treatment selection for CPR activities, including slab and joint replacement, joint repair, and crack sealing, improving efficiency and accuracy in the design process. A software tool was developed to implement and validate these methodologies.

By leveraging high-resolution 3D pavement surface data, this study provides a transformative framework for predictive and precision-based project-level pavement asset management, equipping transportation agencies with the tools necessary to improve condition assessments, optimize maintenance interventions, and enhance the longevity and resilience of roadway infrastructure.