U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-HRT-06-121
Date: November 2006

Long-Term Pavement Performance (LTPP) Data Analysis Support: National Pooled Fund Study Tpf-5(013)

Chapter 2. Background

In phase 1 a literature review was conducted to provide guidance and support to the work on this project. The available literature directly related to this investigation was quite limited. While most of the literature regarding frost effects deals with quantifying the change in material properties and performance characterization on particular projects, studies have been conducted on modeling pavement performance using LTPP data. The literature review is included in appendix A.

As part of phase 1, the contractor was asked to complete the following task:

Identify the specific LTPP data to be used in the analysis, acquire the data, and process it as necessary to create the analysis database to be used in subsequent analysis.

The analysis database was first developed to evaluate the potential of using only data from Seasonal Monitoring Program (SMP) test sections for the study. This evaluation showed that including additional test sections beyond SMP sites would increase the dataset to better represent the large number of variables and lead to improved results. With the approval from Federal Highway Administration (FHWA) and the pooled fund States (PFS) panel, the sample set was expanded to include GPS-1 and GPS-6 experiments for phase 1 analysis. The analysis database was expanded for phase 2, which included accumulating additional variables, combining data that had been segregated into environmental zones for phase 1, and thoroughly reviewing the data. Chapter 3 contains a more detailed discussion of the database development.

Based on phase 1 findings, it was evident that the available data could support the study, and that performance differences did exist. Because the analysis dataset required additional data, it was proposed that it be expanded to include not only GPS-1 and GPS-6 sites, but also GPS-2, SPS-1, and SPS-8 projects for an investigation of AC sections for phase 2. Similarly, GPS-3, SPS-2, and SPS-8 test sections were proposed to model PCC pavements. The analysis datasets for phase 2 were designed to combine the data into a more comprehensive form where the environmental factors could be fully addressed in a statistical analysis. This approach was presented to, and accepted by, the PFS panel.

Another consideration from the phase 1 study was the use of frost depths, information that is collected only at SMP sites. The database could not be expanded from SMP sites without using a substitute for frost depths. The annual freezing index (FI) was shown to correlate quite well to frost depths, and it was subsequently used in both the phase 1 and phase 2 studies to represent relative freezing conditions. A complete analysis of these findings is reported in chapter 3

An investigation of the interaction between FTCs and FI found that moderate- and deep-freeze zones experience approximately the same number of annual freeze cycles; therefore, the initial assumption that each was mutually exclusive, that the moderate-freeze zone would experience multiple FTCs while sections in the deep-freeze zone would experience few cycles, is not confirmed by the data. Consideration of this finding in phase 2 analysis was essential to making performance comparisons, described in chapter 5.

An initial trend analysis was conducted to determine if the rate of deterioration varied between environmental settings. Linear regression was performed on the preliminary dataset and differences in performance were observed; however, because of the large spread of data, most differences would not likely be statistically significant at the 95-percent confidence interval. Moreover, simple linear regression does not consider the large amount of independent variables that contribute to pavement performance. Considering this, a more complex investigation (consisting of multivariate regression analysis) was initiated in phase 2.

A preliminary cost investigation was performed to determine the amount of cost data available for future analysis. Sufficient amounts of cost data were found to be available to make cost comparisons between States in different frost conditions; however, the unit descriptions were not consistent among the agencies. Therefore, phase 2 work consisted of determining standard roadway sections and identifying unit descriptions that were consistent. This was accomplished through inquiries with each SHA.

Previous | Contents | Next