TECHBRIEF

Help with Converting Pavement
Smoothness Specifications

Report No. FHWA-RD-02-112
FHWA Contact: Mark Swanlund, HIPT-20, (202) 366-1323

Introduction

Pavement smoothness is a key factor in determining highway user satisfaction. To adequately represent drivers' opinions of roadway conditions, many highway agencies are transitioning to the International Roughness Index (IRI) or to the Profile Index (PI) using a 0.0-millimeter (mm) blanking band (PI0.0) from the PI using a 5-mm or 2.5-mm blanking band (PI5-mm or PI2.5-mm). This has led to the need for correlation equations to relate the current PI5-mm or PI2.5-mm smoothness specification levels with those to be used in the new specifications.

To address this issue, research was initiated to develop a practical tool to assist in the transition from PI5-mm or PI2.5-mm to IRI or PI0.0. Information developed in this research can be used by highway agencies to estimate the level of IRI and PI0.0 smoothness associated with their current pavement smoothness specifications.

Key Findings

Key findings of this research include:

Data

The FHWA Long-Term Pavement Performance (LTPP) Program provided profile data used in this research. Since 1996, LTPP has collected pavement surface profile data from more than 1,700 test sections (typically 152.5 meters (m) long) across the United States and Canada. LTPP uses a high-speed inertial profiler, rated as class I according to ASTM E-950-98, to record profile data at 25-mm intervals. IRI values for each profile are computed using LTPP software according to the ASTM E-1926-98 protocols. Researchers obtained climatic and pavement type information from the LTPP database.

Analysis

To determine the PI values from the profiles, researchers employed a commonly used simulation method. Lightweight profilers simulating PI values have been in use for more than 5 years, and several manufacturers have proven that they can correlate well with PIs obtained from profilographs. For this study, researchers used the algorithm employed by the lightweight profiler that the LTPP profiler manufacturer developed. Because the same sensors, filtering, and sampling interval are used for the manufacturer's high-speed profiler and lightweight profiler, this simulation held the best promise of matching profilograph PI output.

Results

Based on the data shown in figures 1 and 2, a linear relationship is evident between IRI and PI0.0 for both asphalt concrete (AC) and portland cement concrete (PCC) pavements. Rather good correlation coefficients result from these direct comparisons, with an R2 of 0.87 and 0.83 for AC and PCC pavements, respectively.

Figure 1. IRI vs. PI0.0 for AC pavements.

Figure 1. IRI versus PI (0.0 millimeter blanking band) for asphalt concrete (AC) pavements. The figure shows a graph with simulated PI (0.0 millimeter blanking band), meters per kilometer, on the horizontal axis; and IRI, meters per kilometer, on the vertical axis. The linear regression line starts at an IRI of 0.2 and a PI of 0.0 and ends at an IRI of 5.2 and a PI of 1.8. The full-pay range is shown to be within an IRI of 0.6 to 1.5 and a PI of 0.15 to 0.50.

Figure 2. IRI vs. PI0.0 for PCC pavements.

Figure 2. IRI versus PI (0.0 millimeter blanking band) for portland cement concrete (PCC) pavements. The figure shows a graph with simulated PI (0.0 millimeter blanking band), meters per kilometer, on the horizontal axis; and IRI, meters per kilometer, on the vertical axis. The linear regression line starts at an IRI of 0.8 and a PI of 0.15 and ends at an IRI of 4.5 and a PI 1.85. The full-pay range is shown to be within an IRI of 0.8 to 1.5 and a PI of 0.15 to 0.50.

Statistical analysis reveals that, in certain instances, a small portion of the variability in the relationships between IRI and PI can be related to climate and pavement type. To provide agencies with the most accurate equations, these variables were included in the PI to IRI models shown in table 1.

Table 1. PI-to-IRI index conversion equations and variability indices.

Pavement TypeClimateaBlanking Band (mm)Correlation Equation (IRI = mm/km, PI = mm/km)NSEER2
AC1,2,3,40.0IRI = 2.66543*PI0.0 + 213.0114,170200.170.89
AC1,2,3,45.0IRI = 3.78601*PI5-mm + 887.5113,775292.260.77
AC/AC10.0IRI = 2.74599*PI0.0 + 265.421,854191.970.91
AC/AC20.0IRI = 2.68169*PI0.0 + 274.671,494184.640.81
AC/AC3,40.0IRI = 2.42295*PI0.0 + 301.905,126178.810.84
AC/AC15.0IRI = 4.25316*PI5-mm + 957.801,824288.170.79
AC/AC25.0IRI = 4.39478*PI5-mm + 883.201,345308.230.45
AC/AC3,45.0IRI = 3.42671*PI5-mm + 876.804,906265.850.63
AC/PCC1,2,3,40.0IRI = 2.40300*PI0.0 + 292.934,156205.580.79
PCC1,30.0IRI = 2.12173*PI0.0 + 439.7612,039259.630.84
PCC20.0IRI = 2.58454*PI0.0 + 423.091,448176.540.88
PCC40.0IRI = 2.3582*PI0.0 + 317.192,888236.510.84
PCC1,35.0IRI = 2.62558*PI5-mm + 1,205.7311,946305.960.77
PCC25.0IRI = 3.51673*PI5-mm + 1,226.351,364268.700.72
PCC45.0IRI = 2.87407*PI5-mm + 1,229.632,885297.370.74

aClimatic zones: 1=DF, 2=DNF, 3=WF, 4=WNF.

Climatic regions used in these tables are defined as dry freeze (DF), dry non-freeze (DNF), wet freeze (WF), and wet non-freeze (WNF). Climates with average annual rainfall of more than 508 mm per year were considered as wet. Freezing climates were designated as having a freezing index of more than 66°C-days per year.

As expected, the relationship between PI5-mm and PI0.0 is also linear, as figures 3 and 4 illustrate. The variability in these relationships is an indication of the pavement roughness data that is lost when a wider blanking band is used. The R2 value for all AC pavements is 0.85, while it is increased to 0.90 for all PCC pavements.

Figure 3. PI0.0 vs. PI5.0 for AC pavements.

Figure 3. PI (0.0 millimeter blanking band) versus PI (5.0 millimeter blanking band) for AC pavements. The figure shows a graph with simulated PI (5 millimeter blanking band), meters per kilometer, on the horizontal axis; and simulated PI (0.0 millimeter blanking band), meters per kilometer, on the vertical axis. The linear regression line starts at a PI 0.0 millimeters of 0.25 and a PI 5 millimeters of 0.0 and ends at a PI 0.0 millimeters of 2.0 and a PI 5 millimeters of 1.2. The full-pay range is shown to be within a PI 0.0 millimeters of 0.25 to 5.2 and a PI 5 millimeters of 0.0 to 0.2.

Figure 4. PI0.0 vs. PI5.0 for PCC pavements.

Figure 4. PI (0.0 millimeter blanking band) versus PI (5.0 millimeter blanking band) for PCC pavements. The figure shows a graph with simulated PI (5 millimeter blanking band), meters per kilometer, on the horizontal axis; and simulated PI (0.0 millimeter blanking band), meters per kilometer, on the vertical axis. The linear regression line starts at a PI 0.0 millimeters of 0.4 and a PI 5 millimeters of 0.0 and ends at a PI 0.0 millimeters of 1.8 and a PI 5 millimeters of 1.2. The full-pay range is shown to be within a PI 0.0 millimeters of 0.4 to 5.2 and a PI 5 millimeters of 0.0 to 0.2.

No statistical differences in the correlations were noted for jointed-plain, jointed-reinforced, and continuously reinforced concrete (JPC, JRC, and CRC) pavements. However, statistical differences were noted between original AC pavements, AC overlays of AC, and AC overlays of PCC. Also, climatic effects accounted for some of the variability in the equations. Therefore, the correlations in table 2 include several pavement types and climatic regions.

Table 2. PI-to-PI 0.0 index conversion equations and variability. PI-to-IRI index conversion equations and varibility indices.

Pavement TypeClimateaCorrelation Equation(PI = mm/km)NSEER2
AC1,3PI0.0 = 1.35776*PI5-mm + 275.485,68483.580.88
AC2,4PI0.0 = 1.46417*PI5-mm + 240.098,09371.730.86
AC/AC1PI0.0 = 1.56038*PI5-mm + 250.891,82673.740.88
AC/AC2PI0.0 = 1.75837*PI5-mm + 222.841,34779.320.66
AC/AC3,4PI0.0 = 1.45876*PI5-mm + 233.594,90871.530.81
PCC1PI0.0 = 1.39512*PI5-mm + 343.082,18271.190.87
PCC2PI0.0 = 1.36715*PI5-mm + 313.251,36666.420.86
PCC3PI0.0 = 1.20723*PI5-mm + 367.919,76486.730.91
PCC4PI0.0 = 1.19909*PI5-mm + 390.492,88585.190.85

aClimatic zones: 1=DF, 2=DNF, 3=WF, 4=WNF.

Applications

These correlation equations can be used to assist agencies transition from the PI5-mm smoothness index to IRI or to PI0.0. For example, a wet-freeze State has specified a full-pay range for new PCC pavements of PI5-mm between 79 and 110 mm/kilometer (km). State officials can use equations 14 and 25 to determine that a comparable IRI range would be 1,413 to 1,495 mm/km and a comparable PI0.0 range would be 463 to 501 mm/km.

Similarly, for AC pavement smoothness specifications (full-pay is 64 to 110 mm/km), the State could use equations 2 and 17 to estimate a comparable IRI range of 1,130 to 1,304 mm/km and a comparable PI0.0 range of 362 to 425 mm/km.

Recommendations

To make this research useful, it is recommended that agencies:

  1. Evaluate the validity of the research results based on agency conditions and experience.

  2. Use the correlation equations and variability information to estimate the required level of smoothness for a specification that transitions to IRI or PI0.0. It is recommended that the derived specification limits be adjusted to reflect agency implementation practices such as segment length, segment averaging, scope of application, and index computation method.

  3. Track the results of the new smoothness specification and adjust agency smoothness requirements to meet the abilities of contractors and the benefit-cost of smoother roadway surfaces.

Additional Information

Researcher -- This study was performed by ERES Consultants, 505 West University Avenue, Champaign, Illinois 61820-3915, (217) 356-4500. Contract No. GS-10F-0298K.

Distribution -- This TechBrief is being distributed according to a standard distribution. Direct distribution is being made to the Resource Centers and Divisions.

Availability -- The publication linked to this TechBrief is Pavement Smoothness Index Relationships, Final Report (Report No. FHWA-RD-02-057). It is available from the National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161. A limited number of copies will be available from the R&T Report Center, FHWA, 9701 Philadelphia Court, Unit Q, Lanham, MD 20706, telephone: (301) 577-0818, fax: (301) 577-1421.

Key Words -- International Roughness Index, Profile Index, correlation, specification, smoothness.

Notice -- This TechBrief is disseminated under the sponsorship of the Department of Transportation in the interest of information exchange. This TechBrief provides a synopsis of the study's final publication. The TechBrief does not establish policies or regulations, nor does it imply FHWA endorsement of the conclusions or recommendations. The U.S. Government assumes no liability for the contents or their use.

FHWA-RD-02-112
NOVEMBER 2002

U.S. Department of Transportation
Federal Highway Administration
Research, Development, and Technology
Turner-Fairbank Highway Research Center
6300 Georgetown Pike
McLean, VA 22101
www.tfhrc.gov

This RD&T TechBrief summarizes an experiment, its results, data collected from analysis, or findings and recommendations.

FHWA
TFHRC Home | FHWA Home | Feedback



United States Department of Transportation - Federal Highway Administration