Longitudinal cracks, those long fissures that run parallel to highway center lines, are more than just a nuisance to drivers. They accelerate pavement deterioration and cost city, state, and federal governments millions in repairs. Yet until now, little effort has been made to model and predict this specific type of road failure.
Two University of Pittsburgh engineers are tackling this challenge to provide a nationwide infrastructure solution. Lev Khazanovich and Julie Vandenbossche secured a highly competitive $600,000, three-year grant from the National Academies' National Cooperative Highway Research Program (NCHRP) to pinpoint the root causes of longitudinal cracking and create predictive models for integration into design software.
The project, “ Development of Longitudinal Cracking Models for Concrete Pavements ,” will improve how engineers and planners across the country design roads and predict their performance.
“For thousands of years, all great civilizations have built great roads, and the U.S. interstate system is an engineering miracle,” said Khazanovich, the Anthony Gill Chair Professor in the Swanson School of Engineering Department of Civil and Environmental Engineering and principal investigator for this project. “However, political, budgetary, and labor constraints affect how roads are designed and how long they last.”
“Roads are designed to perform for a pre-determined period of time,” said Vandenbossche, professor of civil engineering, Associate Chair of Research, and co-principal investigator. “When they fail too soon or last too long, it skews the lifecycle cost analysis used for selecting the most cost-effective design. An accurate prediction of the performance life is imperative for these lifecycle cost analyses to be accurate.”
An overlooked road failure
Longitudinal cracking occurs on jointed plain concrete pavements (JPCP) that are commonly used on roads and interstates. These cracks can form from poor construction, weather, and heavy loads.
To improve how roads are built and to better predict their lifespans, engineers have developed mechanistic-empirical design models, which incorporate principles based on physics and engineering as well as vast amounts of data collected during testing and observation. Although these models have evolved to meet 21 st -century demands and reduce environmental impact, the current version of the American Association of State Highway and Transportation Officials (AASHTO) pavement mechanistic-empirical design (PMED) software does not account for longitudinal cracking in the design of original concrete pavements.
“On the modeling side, there has been a greater focus on more prevalent failure modes,” said Vandenbossche, “but in focusing on some modes and neglecting others, longitudinal cracking has become more common.”
Expanding the model
To improve the model and design software, which is used nationwide, Khazanovich and Vandenbossche will conduct a thorough analysis to identify the causes of longitudinal cracking. In addition to reviewing the existing literature, models, documentation, and databases, they will survey transportation officials from across the country and will also identify and research stretches of road experiencing longitudinal cracking.
From their findings, the team will develop and test mechanistic-empirical models that predict this persistent failure mode. These models will be compatible with the PMED software so they can be easily implemented into it, improving a tool vital to road engineering.
“This project builds off our previous work to improve how pavements are designed, so that we can develop safer, more efficient roads,” said Khazanovich. “We’re honored to receive this competitive award from the National Academies of Science.”
“Billions of dollars are invested in our roads,” said Vandenbossche, “and this research will provide a better tool to achieve the best investment. It’s a great opportunity for the U.S. to account for this neglected failure mode and to advance a national design procedure.”