U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

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

 
REPORT
This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-17-049    Date:  October 2017
Publication Number: FHWA-HRT-17-049
Date: October 2017

 

Investigation of Increase in Roughness Due to Environmental Factors in Flexible Pavements Using Profile Data From Long-Term Pavement Performance Specific Pavement Studies 1 Experiment

CHAPTER 3. IRI DATA USED FOR THE STUDY

 

OBTAINING DATA FOR ANALYSIS

As indicated previously, since December 1996, center of the lane profile data at the test sections in the LTPP Program have been collected along with the wheelpath profile data, starting with use of the K.J. Law T-6600 profiler. The 0.98-inch interval profile data collected at the SPS-1 projects shown in table 2 since December 1996 were used for analysis. As described previously, profile data files containing the 0.98-inch interval profile data in ERD format have been stored in the AIMS. These ERD files were obtained through the LTPP customer support service over the period of October to November 2014. In most cases, profile data for seven to nine repeat runs for each profile date were available.

The profile data collection dates at LTPP sections and the ride parameters computed from the profile data have been stored in the MON PROFILE MASTER table in the PPDB.(1) Records for SPS-1 sections were extracted from this table from Standard Data Release (SDR) 28. The information obtained was used to determine the following information for each test section: the date when the section was first profiled, the date when the section was first profiled with the K.J. Law T-6600 profiler, and the last profile date. The EXPERIMENT SECTION table in the PPDB was used to obtain the date when each LTPP section was opened to traffic and to determine whether the section was out of the LTPP study (i.e., deassigned), and if so, what date the section was deassigned (i.e., deassign date). Appendix A of this report contains a table that shows the following information for each SPS-1 test section:

The information contained in appendix A was obtained as described previously from the MON PROFILE MASTER table and the EXPERIMENT SECTION table in the PPDB. For sections that were still active in the LTPP study, the last profile date shown in appendix A represents the most recent profile date indicated in the MON PROFILE MASTER table obtained from SDR 28.

The profile dates for SPS-1 sections in the MON PROFILE MASTER table were used to check whether the ERD files corresponding to all profile dates were obtained. The table obtained from SDR 28 indicated 2,250 site visits had been made to SPS-1 sections during which center of the lane profile data were collected, and it was possible to obtain ERD files for approximately 98 percent of these site visits.

The LTPP Program’s SMP collected data in the different seasons as a subset of LTPP sections over a time period. Some SPS-1 sections were included in the SMP. The sections in the SMP were profiled four times a year during each season (i.e., spring, summer, fall, and winter) while the SMP program was active. Table 4 shows the SPS-1 sections that were included in the SMP. This table shows the following information for the section: State where the section was located, LTPP section number, whether the section was deassigned, date when profile data were first collected at the section, date when profile data were first collected at the section by the K.J. Law T-6600 profiler, date when the section was last profiled, and number of times profile data were collected at the site that included center of the lane data.

Table 4. SPS-1 sections that were SMP sites.

State LTPP
Section
Number
Deassigned? First Profile Date Last
Profile
Date
Number of Times
Profiled With
Center of the
Lane Data
At the
Section
With
T‑6600
Arizona 040113 Yes 1/27/1994 7/3/1997 3/27/2006 27
Arizona 040114 Yes 1/27/1994 1/23/1997 3/27/2006 27
Montana 300114 No 11/19/1998 11/19/1998 9/5/2012 32
Nevada 320101 Yes 12/3/1996 12/3/1996 6/24/2009 30
Virginia 510114 Yes 4/24/1996 10/9/1997 3/27/2012 29

 

COMPUTATION OF IRI VALUES

The profile data in the ERD files were used to compute the left wheelpath, right wheelpath, and the center of the lane IRI values for all profile runs made at SPS-1 sections. Thereafter, for each site visit, the IRI values for the repeat runs were averaged to obtain an average IRI for each profiled path (i.e., left wheelpath, right wheelpath, and center of the lane). These averaged IRI values were used for analysis.

As an example, table 5 shows the IRI values for each profiled path for the seven runs collected at section 010101 on 7/3/1997 and the average IRI values for each path. The IRI values shown for each run in table 5 are about 3 to 4 percent greater than the IRI values shown in the MON PROFILE MASTER table in the PPDB for the corresponding run.

The IRI values included in the PPDB were computed by smoothening the 0.98-inch interval profile data with an 11.8-inch moving average, extracting data at 5.9-inch intervals, and using the resulting profile to compute the IRI values. The IRI computation program has a 9.8‑inch moving average built into it to address tire enveloping before computing IRI. If the profile data used in the IRI computation had already been subjected to a moving average, the 9.8‑inch moving average in the IRI program should not be applied to that data.(18) However, this procedure was not followed when computing IRI from the 5.9-inch interval profile data in the LTPP Program. The application of the 9.8-inch interval moving average in the IRI program to data that had already been subjected to a moving average would cause further smoothening of the data, resulting in a decrease in the IRI value. This was the reason IRI computed from the 0.98‑inch interval profile data was slightly greater than IRI computed from the 5.9-inch interval profile data.

Table 5. IRI values at section 010101 for profile data collected on 7/3/1997.

Run
Number
IRI (Inches/mi)
Left Wheelpath Right Wheelpath Center of the Lane
1 47.8 43.6 41.7
2 45.9 44.7 40.4
3 47.2 44.5 41.3
4 45.0 44.0 38.9
5 46.1 43.7 40.6
6 45.4 46.1 41.0
7 45.8 45.1 40.9
Average 46.2 44.5 40.7

 

The center sensor data were not available or the center sensor data values were zeros for the cases shown in table 6. Study researchers verified with the regional contractors that the center sensor in the profiler was not functioning for the cases shown in table 6.

Table 6. Sections and profile dates for which center sensor data were not available.

State Sections Profile Date Comment
Michigan 13–24 8/10/2010 Center sensor values are zero
Ohio 1–12 8/11/2010 Center sensor values are zero
Montana 13–24 7/21/2011 No center sensor data
Montana 13–24 9/15/2012 No center sensor data

 

CONSTRUCTION NUMBER OF LTPP SECTIONS

The construction number “1” was assigned to an SPS-1 section initially after construction. Whenever any maintenance or rehabilitation activity was performed at an SPS-1 section, the construction number was incremented, and the date the maintenance or rehabilitation was performed, including the type of maintenance or rehabilitation activity, was recorded in the PPDB table EXPERIMENT SECTION. Maintenance activities at a section could include crack sealing, slurry sealing, application of an aggregate seal coat, or patching. Such activity could cause a decrease in IRI, cause an increase in IRI, or have no impact on IRI. Rehabilitation performed at a section typically involved the placement of an overlay, which would usually cause a sharp reduction in the IRI of the pavement. It was possible that some maintenance or rehabilitation activities performed at a test section might not have been recorded in the EXPERIMENT SECTION table in the PPDB.

Evaluation of the information contained in the table EXPERIMENT SECTION indicated there were only three SPS-1 projects where no maintenance or rehabilitation activities had been performed on any test section in that project during the monitoring period. These SPS-1 projects were those located in Florida, Louisiana, and Wisconsin.

EVALUATION OF IRI VALUES

IRI versus pavement age plots were developed for each test section to evaluate how the IRI of the left wheelpath, right wheelpath, and center of the lane changed over time. Figure 12 shows an example of such a plot for section 010106. The horizontal axis of this plot represents the pavement age, with the pavement age being assigned a value of 0 for the traffic open date. The first IRI value shown in this plot corresponds to the date when center of the lane profile data were first collected at this section. As described previously, center of the lane profile data were first collected at the LTPP sections starting with the K.J. Law T-6600 profilers that went into operation in December 1996. At most test sections, profile data were collected previously along the two wheelpaths with the K.J. Law DNC690 profilers, but as described previously, this profiler did not have a sensor to collect the center of the lane data.

Click for description

Figure 12. Graph. IRI progression at section 010106.

 

The time sequence IRI values at each test section were visually examined to evaluate the changes in IRI over time. Particular attention was paid to the following items when evaluating the IRI plots:

The evaluation of time-series IRI data indicated that the sudden increases or decreases in IRI between profile visits could be attributed to one of the following reasons:

The following subsections describe examples illustrating the influence of each of these conditions on IRI. Crack sealing is also included although it did not appear to be a factor that caused a noticeable change in IRI.

Equipment Errors

This section describes several examples of instances where bad data were collected because of equipment problems. The time-sequence IRI values for section 010104 are shown in table 7, while the time-sequence IRI plot for these data is shown in figure 13. The right wheelpath IRI values for 4/23/1998, 12/7/1998, and 3/14/2001 were 42, 62, and 39 inches/mi, respectively. As seen from these values, IRI for 12/7/1998 was higher than IRI for previous and subsequent profiling dates. Figure 14 shows the left wheelpath, center of the lane, and right wheelpath profile data collected by the profiler on 12/7/1998 for a portion of this section. This plot shows the profile data for the right sensor had “noise” compared with the data from the other two sensors, which indicates that the sensor was not functioning properly.

Table 7. Time-sequence IRI values at section 010104.

Profile
Date
Pavement
Age (Years)
IRI (Inches/mi)
Left Right Center
7/3/1997 4.3 47 38 41
1/27/1998 4.9 45 40 38
4/23/1998 5.1 47 42 38
12/7/1998 5.8 50 62 38
3/14/2001 8.0 46 39 41
3/10/2002 9.0 48 45 46
1/29/2003 9.9 48 40 41
4/27/2004 11.2 47 40 42
5/4/2005 12.2 49 41 42

 

Click for description

Figure 13. Graph. Time-sequence IRI values at section 010104.

 

Click for description

Figure 14. Graph. Profile data collected at section 010104 on 12/7/1998.

 

The time-sequence IRI values for section 260115 for the first five site visits with the K.J. Law T‑6600 profiler are shown in table 8. The CLIRI value for 12/30/1996 was 177 inches/mi, while this value for the subsequent profile date of 4/16/1997 was 50 inches/mi. The IRI values for the left and right wheelpaths did not show such a drop between these two visits, which indicates there might have been a problem with the center sensor during the 12/30/1996 data collection. Figure 15 shows the left wheelpath, right wheelpath, and center of the lane profile data collected by the profiler on 12/30/1996 for a portion of the test section. This plot shows the profile data for the center sensor had a sinusoidal pattern indicating that the sensor was not functioning properly.

Table 8. IRI from first five site visits with K.J. Law T-6600 profiler at site 260115.

Profile Date Pavement Age
(Years)
IRI (Inches/mi)
Left Right Center
12/30/1996 1.2 48 47 177
4/16/1997 1.5 48 46 50
6/25/1997 1.6 48 47 49
10/27/1998 3.0 57 49 52
4/8/1999 3.4 57 51 54

 

Click for description

Figure 15. Graph. Profile data collected at section 260115 on 12/30/1996.

 

Evaluation of time-sequence IRI values at all sections indicated only a few cases where a sudden increase in IRI could be attributed to errors associated with equipment. Table 9 shows the cases where equipment-related issues affected IRI, and these data were not used for analysis.

Table 9. Cases where equipment-related issues affected the collected data.

State State Code Section Profile Date Issue
Alabama 01 4 12/7/1998 Right wheelpath sensor had noise
Michigan 26 15–18, 20, 21, 23, 24 12/30/1996 Center sensor collected bad data
Ohio 39 3, 4, 6, 8–12 12/8/1997 Center sensor collected bad data
Ohio 39 3, 4, 6, 8–12 12/27/1996 Center sensor collected bad data

 

Data for approximately 2,250 visits to SPS-1 test sections were evaluated in this study. As shown in table 9, only 25 visits to test sections were identified where equipment issues affected the IRI values.

Data Assigned to the Section From an Incorrect Location

Several examples of assignment of incorrect data to a test section are presented in this section. Table 10 shows the time-sequence IRI values for section 010102, while a plot of these IRI values is shown in figure 16.

Table 10. Time-sequence IRI values at section 010102.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
7/3/1997 4.3 58 59 54
1/27/1998 4.9 59 70 55
4/23/1998 5.1 60 69 55
8/5/1998 5.4 67 91 58
12/7/1998 5.8 63 86 57
3/14/2001 8.0 85 94 58
3/10/2002 9.0 57 58 56
1/29/2003 9.9 140 144 58
4/27/2004 11.2 171 214 90
5/4/2005 12.2 194 211 86

 

Click for description

Figure 16. Graph. Time-sequence IRI values at section 010102.

 

The left and right wheelpath IRI values showed a sharp drop on 3/10/2002 compared with IRI for the previous visit. Thereafter, the left and right wheelpath IRI values showed a sharp increase in IRI for 1/29/2003 compared with the IRI values obtained on 3/10/2002. Figure 17 shows the left wheelpath profile data plots for 3/14/2001, 3/10/2002, and 1/29/2003 for a portion of the section. These plots show that the profile trace for the 3/10/2002 data was completely different from the traces obtained before and after this date. Evaluation of the right and center of the lane data also showed similar results. Therefore, it was concluded that the data for 3/10/2002 were not associated with section 010102.

Click for description

Figure 17. Graph. Left wheelpath profile data at section 010102.

 

Table 11 shows the time-sequence IRI values for section 050122. A sudden increase in IRI occurred for the right wheelpath and center of the lane data on 9/21/2008 compared with IRI obtained on 5/22/2007. There was no record of rehabilitation or maintenance activities performed on this section between these two dates.

Table 11. Time-sequence IRI values at section 050122.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
7/1/1997 2.8 48 49 41
1/22/2001 6.4 67 55 43
4/1/2002 7.6 64 50 53
4/23/2003 8.6 75 54 48
3/19/2004 9.6 71 55 48
4/6/2005 10.6 78 58 50
5/22/2007 12.7 84 65 54
9/21/2008 14.1 87 96 92

 

Figure 18 shows the right wheelpath profile plots for 4/6/2005, 5/22/2007, and 9/21/2008. Evaluation of these plots showed that the plot for the 9/21/2008 data was not similar to the plots obtained for the two previous site visits. Therefore, it was concluded that the data for 9/21/2008 were not associated with section 050122.

Click for description

Figure 18. Graph. Right wheelpath profile data at section 050122.

 

The profile data for all sections in an SPS-1 project were collected in a single run, and thereafter, the collected data were subsectioned to extract the profile corresponding to each test section. Incorrect subsectioning can be a reason for assigning incorrect data to a test section. Evaluation of time-sequence IRI plots indicated only a very few cases where a sudden change in IRI could be attributed to assignment of incorrect data to a section. The cases that were identified are shown in table 12, and these data were not used for analysis.

Table 12. Cases where data from an incorrect location were assigned to sections.

State State Code Section Profile Date
Alabama 01 2 3/10/2002
Arkansas 05 13–18, 22–24 9/21/2008
Oklahoma 40 13–24 10/11/2001

 

In this study, a comparison of profile data collected during different site visits was not performed to investigate whether the profile data matched because it was beyond the scope of this project. It is possible for the profile data not to match between visits because of incorrect subsectioning and for the IRI plot not to show any anomalous trends. This could occur if the IRI of the incorrectly subsectioned section had a value that was close to the IRI of the test section.

Overlay Placed on the Section

An overlay placed on a test section should result in a decrease in the IRI value unless the section is so smooth that an overlay would not have much of an impact on IRI. (Note that an increase in IRI could occur if the contractor did not follow best practices for constructing a smooth pavement.) A decrease in IRI was usually noted for all three paths (i.e., left wheelpath, right wheelpath, and center of the lane); however, the change in CLIRI might not be very noticeable if the center of the pavement was smooth. The magnitude of the decrease in IRI caused by an overlay would depend on the IRI of the pavement before the overlay. Sections with a high IRI before an overlay would typically show a greater decrease in IRI compared with sections with lower IRI values. In the LTPP Program, if an SPS-1 test section continued to be monitored after an overlay, the experiment number would be changed from SPS-1 to GPS-6. The date when the overlay was placed, as well as the date when the experiment number was changed, was shown in the PPDB table EXPERIMENT SECTION. Although a change in experiment number was made after the overlay, the section number of the section was not changed. Several examples of reduction in IRI caused by an overlay are presented in this section.

Table 13 shows the time-sequence IRI values at section 260117, while a plot of these data is shown in figure 19. At this section, the first overlay was placed on 10/1/2002, while the second overlay was placed on 6/1/2012. The IRI of the left wheelpath before the overlay on 10/1/2002 was more than double the IRI for the right wheelpath and center of the lane. The overlay placed on 10/1/2002 caused a reduction in IRI for all three paths, with the reduction being greatest for the left wheelpath that had an IRI of 137 inches/mi before the overlay. The overlay on 6/1/2012 also caused a reduction in IRI for all three paths. This overlay resulted in a very smooth pavement that had a MIRI of 27 inches/mi.

Table 13. Time-sequence IRI values at section 260117.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
4/16/1997 1.5 52 50 54
6/25/1997 1.6 55 50 50
10/27/1998 3.0 107 59 54
4/8/1999 3.4 95 61 53
4/19/2000 4.5 114 61 56
5/29/2002 6.6 137 63 60
10/1/2002 First overlay
4/10/2003 7.4 58 50 54
4/21/2005 9.5 69 50 56
6/2/2006 10.6 66 50 57
4/22/2008 12.5 69 55 59
10/14/2009 14.0 72 59 60
8/24/2011 15.8 73 59 61
4/25/2012 16.5 73 60 58
6/1/2012 Second overlay
9/11/2012 16.9 27 27 31

 

Click for description

Figure 19. Graph. Time-sequence IRI values at section 260117.

 

Table 14 shows the time-sequence IRI values at section 350101. IRI for all three paths at this section showed a steady increase in IRI, and thereafter, a sudden drop in IRI occurred for the last profile date of 5/15/2008. The PPDB table EXPERIMENT SECTION indicated that this test section was deassigned on 6/15/2009 but made no mention of any rehabilitation performed on this test section between 4/24/2006 and 5/15/2008. This test section was in the New Mexico SPS-1 project, and evaluation of the IRI data for all test sections in the New Mexico SPS-1 project showed a sudden decrease in IRI between 4/24/2006 and 5/15/2008. The only possible reason for this decrease was either an overlay was placed on the test section or grinding was performed on the test section to eliminate rutting. However, because rutting in this section was 0.25 inches based on the transverse profile data collected 3/28/2006, it was unlikely that grinding was performed to eliminate rutting. Therefore, it was very likely that an overlay was placed on this section between 4/24/2006 and 5/15/2008 that resulted in the decrease in IRI. Changes in rut depth at this section after 3/28/2006 could be used to verify whether an overlay was placed, but no transverse profile data used to compute rut depths were collected at this site after 4/24/06.

Table 14. Time-sequence IRI values at section 350101.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
3/11/1997 1.4 36 41 34
5/21/2000 4.6 50 42 44
5/4/2001 5.5 80 49 51
1/9/2003 7.2 74 64 62
6/9/2004 8.6 81 69 71
3/11/2005 9.4 102 86 93
4/24/2006 10.5 112 92 100
5/15/2008 12.5 50 69 59

 

If a section was overlaid, only the data up to the point before the overlay were used for analysis in this project. If the section was monitored after the overlay and sufficient post-overlay IRI values were available, the increase in IRI of the overlaid pavement was analyzed separately.

As indicated previously, information about overlays placed on SPS-1 sections was contained in the PPDB table EXPERIMENT SECTION. Evaluation of the time-sequence IRI data indicated overlays placed on SPS-1 sections were correctly identified in this table except for the possible overlay placed at all sections on the New Mexico project.

Table 15 shows the SPS-1 sections that were overlaid and the overlay date. All sections in the Delaware SPS-1 project were overlaid immediately after opening to traffic. The first IRI values available for the sections in this project were after the overlay on 6/10/1997, and the IRI values on the overlaid sections were used for analysis. For the sections in Kansas, Ohio, and Virginia, the IRI values up to the first overlay were used for analysis. The sections in Michigan and Texas SPS-1 project were overlaid twice. For these two projects, two sets of IRI data were created for each test section. The first dataset consisted of IRI values up to the first overlay, while the second dataset contained IRI values after the first overlay up to the second overlay. These datasets were treated separately for analysis purposes to determine roughness progression of the original sections and the overlaid sections.

Table 15. SPS-1 sections that were overlaid.

State State Code Sections Overlay Date Comment
Delaware 10 1–12 9/22/1996
Kansas 20 3, 4, 5, 6, 8, 9, 10, 11, 12 10/1/2001
Michigan 26 15 to 18, 20, 21, 23, 24 10/1/2002 First overlay
6/1/2012 Second overlay
New Mexico 35 1–12 Between 4/24/2006 and 5/15/2008
Ohio 39 11 5/30/2007
4, 12 6/1/2012
Texas 48 13–24 4/29/2002 First overlay
3/31/2007 Second overlay
Virginia 51 14–24 7/1/2011
—Indicates there is no comment.

 

Grinding Performed on the Section

A few SPS-1 sections were ground to eliminate rutting. All active sections in the Nebraska SPS‑1 project were ground on 7/12/2000. The grinding reduced the IRI along the two wheelpaths at all sections except for one where it increased the IRI. The effect of grinding on CLIRI was mixed, with grinding having no impact on IRI on some sections, reducing IRI on others, and increasing IRI on still others.

Table 16 shows the time-sequence IRI values at section 310118 in the SPS-1 project in Nebraska. Grinding reduced the IRI of all three paths at this section, with a reduction in IRI of 9, 18, and 20 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively.

Table 16. Time-sequence IRI values at section 310118.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
2/18/1997 1.6 87 83 88
5/16/1998 2.8 89 83 89
5/7/1999 3.8 86 82 89
10/13/1999 4.2 91 77 95
3/20/2000 4.6 85 83 93
7/12/2000 Grinding
5/16/2001 5.8 76 65 63
4/24/2002 6.7 65 55 68

 

Table 17 shows the SPS-1 sections where grinding was performed according to the information in the PPDB table EXPERIMENT SECTION. For the sections in Nebraska, the IRI data up to the point when grinding was performed were used for analysis. Only two IRI data points were available after grinding for the test sections in the Nebraska SPS-1 project. Therefore, sufficient data were not available to investigate IRI progression on the ground pavement. Investigation of IRI values at the two sections in New Mexico shown in table 17 did not show evidence of grinding having an effect on the IRI. Therefore, all available time-sequence data were used for analysis at these two sections. The three sections in Texas showed an increase in IRI after grinding; the average increase in MIRI of a section was 32 inches/mi. Only one IRI data point was available before grinding for these three sections. For these three sections, the pre-grinding IRI was ignored, and only the IRI values obtained after grinding were used for analysis.

Table 17. Sections where grinding had been performed.

State Sections Grinding Date
Nebraska 13–24 7/12/2000
New Mexico 2, 11 3/15/2005
Texas 15, 16, 19 7/7/1998

 

Patching at Sections

Several SPS-1 sections had patching performed on them. The date when patching was performed, the number of patches, and the area of patches were available in the PPDB. Some patches were placed across the entire lane, which affected the IRI for all three paths. In some cases, patching might only be performed along a single wheelpath, and in such cases, only the IRI of that wheelpath would be affected. This section presents several examples of how patching affected the IRI.

The time-sequence IRI values at section 200105 are shown in table 18, while a plot of these data are shown in figure 20. CLIRI showed a sudden increase of 22 inches/mi between 3/21/1999 and 3/17/2000, with the IRI along the left and the right wheelpath reduced by 27 and 9 inches/mi, respectively. The EXPERIMENT SECTION table in the PPDB indicated that full-depth patching had been performed at this section on 6/1/1999. Figure 21 shows the continuous IRI plots based on a 25-ft base length for the center of the lane data collected on 3/21/1999 and 3/17/2000. A data point on a continuous IRI plot shows the average IRI over the base length centered at that location. For example, the IRI shown on the plot at 100 ft was the average IRI from 87.5 to 112.5ft. This plot shows that CLIRI agreed up to about 225 ft for the two dates, and thereafter they were different. This difference occurred because of patching performed on the test section after 225 ft. The patching performed on 6/1/1999 caused the IRI of the left and right wheelpaths to be reduced by 27 and 9 inches/mi, respectively, but increased CLIRI by 22 inches/mi.

Table 18. Time-sequence IRI values at section 200105.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
2/13/1997 3.3 96 95 59
8/21/1998 4.8 129 142 63
3/21/1999 5.4 173 149 68
6/1/1999 Full-depth patching
3/17/2000 6.4 146 140 90
5/12/2001 7.5 180 176 95

 

Click for description

Figure 20. Graph. Time-sequence IRI values at section 200105.

 

Click for description

Figure 21. Graph. Continuous IRI plot for center of the lane for section 200105.

 

The time-sequence IRI values at section 200106 are shown in table 19. The IRI for all three paths showed a sudden increase in IRI after 3/21/1999, with the IRI increasing by 24, 33, and 37 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively. The EXPERIMENT SECTION table in the PPDB did not indicate any maintenance activities occurred at this section between these two profile dates. However, the distress survey performed at this section on 12/6/1999 indicated the section was patched from 0 to 50 ft and from 278 to 500 ft. Figure 22 shows continuous IRI plots based on a 25-ft base length for the center of the lane for the data collected on 3/21/1999 and 3/17/2000. The two plots agreed well with each other up to about 270 ft, and thereafter, they were different. This difference was attributed to the patching that was performed at the section, which caused the CLIRI to increase. Hence, the changes in IRI noted at this section between 3/21/1999 and 3/17/2000 were attributed to the patching that was performed at this section. The patching caused the IRI of the left wheelpath, right wheelpath, and center of the lane to increase by 24, 33, and 37 inches/mi, respectively.

Table 19. Time-sequence IRI values at section 200106.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
8/21/1998 4.8 71 81 67
3/21/1999 5.4 71 79 67
3/17/2000 6.4 95 112 104
5/12/2001 7.5 107 95 104

 

Click for description

Figure 22. Graph. Continuous IRI plot for center of the lane for section 200106.

 

The time-sequence IRI values at section 050119 are shown in table 20. The IRI for all three paths showed a sharp increase in IRI between 7/1/1997 and 1/22/2001, with the IRI increasing by 67, 34, and 61 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively. The EXPERIMENT SECTION table in the PPDB indicated that full-depth patching was performed on this section on 5/15/1999. The distress data indicated that this section had two patches with a total patch area of 565 sq ft. Figure 23 shows the continuous IRI plots based on a 25-ft base length for the center of the lane for the data collected on 7/1/1997 and 1/22/2001. The two plots agreed well with each other except for two localized areas. These were the areas where patching was performed that resulted in an increase in IRI. The increase in IRI noted at this section along all three paths between 7/1/1997 and 1/22/2001 was primarily attributed to the patching that was performed at this section.

Table 20. Time-sequence IRI values at section 050119.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
7/1/1997 2.8 53 51 43
5/15/1999 Full-depth patching
1/22/2001 6.4 120 85 104
4/1/2002 7.6 121 85 105
4/23/2003 8.6 148 87 105
3/19/2004 9.6 167 102 106

 

Click for description

Figure 23. Graph. Continuous IRI plot for center of the lane for section 050119.

 

The time-sequence IRI values at section 010102 are shown in table 21. The IRI for all three paths showed a sharp increase in IRI between 1/29/2003 and 4/27/2004, with the IRI increasing by 31, 70, and 32 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively. The EXPERIMENT SECTION table in the PPDB indicated that full-depth patching was performed on this section between these two profile visits. The distress data indicated that this section had 19 patches with a total patch area of 3,378 ft2.

Table 21. Time-sequence IRI values at section 010102.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
7/3/1997 4.3 58 59 54
1/27/1998 4.9 59 70 55
4/23/1998 5.1 60 69 55
8/5/1998 5.4 67 91 58
12/7/1998 5.8 63 86 57
3/14/2001 8 85 94 58
1/29/2003 9.9 140 144 58
4/17/2003 Full-depth patching
4/27/2004 11.2 171 214 90
5/4/2005 12.2 194 211 86

 

Figure 24 shows the continuous IRI plot based on a 25-ft base length for the center of the lane for the data collected on 1/29/2003 and 4/27/2004. The two plots agreed well with each other up to about 275 ft, and thereafter, they were different. Between about 275 and 390 ft, the IRI for 4/27/2004 was much greater than that for 1/29/2003, and this increase in IRI was attributed to the patching that was performed at the section. The increase in IRI noted at this section between 1/29/2003 and 4/27/2004 for all three paths was attributed to the patching that was performed at this section.

Click for description

Figure 24. Graph. Continuous IRI plot for center of the lane for section 010102.

 

The four examples previously described showed CLIRI increasing after patching for all cases, with the left and right wheelpath IRI increasing after patching for three cases. For many cases where patching was performed such that the patch extended to the center of the lane, CLIRI increased.

An example of a case where the IRI of all three paths decreased after patching was noted for section 190110. The time-sequence IRI values at section 190110 are shown in table 22. The IRI for all three paths showed a decrease between 5/29/2001 and 8/6/2001, with the IRI decreasing by 13, 20, and 15 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively.

Table 22. Time-sequence IRI values at section 190110.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
9/25/1997 4.3 84 88 87
10/13/1998 5.4 90 69 94
7/19/1999 6.1 96 100 101
5/22/2000 7.0 99 101 100
5/29/2001 8.0 106 110 107
7/1/2001 Strip patching
8/6/2001 8.2 93 90 92
11/22/2002 9.5 101 101 92
9/22/2004 11.3 116 115 106

 

The EXPERIMENT SECTION table in the PPDB indicated that strip patching was performed on this section on 7/1/2001. The distress data indicated this section had five patches with a total patch area of 2,895 ft2. Figure 25 shows the continuous IRI plots based on a 10-ft base length for the center of the lane for the data collected on 5/29/2001 and 8/6/2001.

Click for description

Figure 25. Graph. Continuous IRI plot for center of the lane for section 190110.

 

The two plots agreed well with each other except between 140 and 160 ft and 300 and 320 ft, with a much lower IRI on 8/6/2001 in these two areas compared with the IRI obtained on 5/29/2001. This reduction in IRI was attributed to patching performed in these two areas, and the decrease in IRI noted at this section between 5/29/2001 and 8/6/2001 was attributed to the patching that was performed at this section.

In many cases when patching was performed, it appeared that sufficient attention had not been paid to achieving a smooth pavement in the patched areas. As seen from the examples shown in this section, patching could cause a significant increase in IRI in the patched areas, which increased the overall IRI of the section. The IRI of a section could be reduced by patching when the patch repaired the damaged pavement that contributed to roughness. However, sufficient attention should be paid to properly construct the patch and compact the patch adequately to ensure a smooth patch. Otherwise, although the damaged pavement was repaired, the patch could create localized roughness. The effect of patching on the IRI would depend on the number of patches performed as well as the total area of the patches. In some cases, the patch might not have an effect on the IRI because the constructed patch had the same roughness level as the original pavement or because the area of the patch was too small to have any impact on the overall IRI of the pavement.

The effect of patching on IRI was evaluated for all cases where patching was recorded in the PDBB table EXPERIMENT SECTION by evaluating the time-sequence IRI values. If the patching had an effect in reducing or increasing the IRI of the section that would impact analysis of roughness progression at the section, only the IRI data up to the patch date were considered for analysis. Table 23 shows the sections that were identified where patching had an impact on the overall section IRI either by increasing or reducing the IRI of the section substantially. For these sections, only the IRI data up to the patch date were used for analysis.

Table 23. Sections whose IRI was identified as affected by patching.

State State Code Sections Patch Date
Alabama 01 2 4/27/2004
7 3/14/2001
Arkansas 05 19, 20 5/5/1999
Iowa 19 1–12 7/1/2001
Kansas 20 5, 6 6/1/1999
Nevada 32 8 4/1/2009
Virginia 51 24 10/1/2004

 

Surface Treatments Applied to Sections

Surface treatments applied to SPS-1 sections have included slurry seals and aggregate seals. The cases where a surface treatment was applied to SPS-1 sections are shown in table 24. If distresses were present on the pavement, a slurry seal or an aggregate seal could cover up the distresses, which could cause the IRI to decrease. However, a slurry seal or an aggregate seal might also increase the IRI of the pavement, especially if the pavement was smooth.

Table 24. Cases where a surface treatment was applied to SPS-1 sections.

State State Code Sections Activity Date
Arizona 4 13, 14, 16, 18, 20, 21, 22 Slurry seal 5/1/2002
Kansas 20 5 Aggregate seal 6/1/2004
Montana 30 13–24 Aggregate seal 7/24/2004
Texas 48 13–24 Aggregate seal 6/15/2011

 

Evaluation of the IRI values for the Arizona SPS-1 sections before the slurry seal and after the slurry seal indicated the right wheelpath IRI of all sections decreased after the slurry seal. However, the left wheelpath IRI and CLIRI increased at all sections except for both of these paths at section 040113 and the center of the lane of section 040119.

The time-sequence IRI values at section 040118, which is a section in the Arizona SPS-1 project that was subjected to the slurry seal on 5/1/2002, are shown in table 25, while figure 26 shows the time-sequence IRI values at this section. The application of the slurry seal reduced the IRI of the right wheelpath by 37 inches/mi but caused the left wheelpath IRI and CLIRI to increase by 16 and 14 inches/mi, respectively. As shown in table 25 and figure 26, CLIRI had the same value on 11/6/2001 and 2/20/2002 and then increased by 14 inches/mi because of the slurry seal.

Table 25. Time-sequence IRI values at section 040118.

Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
1/23/1997 3.5 35 73 38
4/8/1998 4.7 36 69 46
12/4/1998 5.3 35 72 40
11/17/1999 6.3 38 74 38
12/19/2000 7.4 37 78 38
11/6/2001 8.3 38 101 38
2/20/2002 8.6 38 103 38
5/1/2002 Slurry seal
3/2/2003 9.6 54 66 52
3/10/2004 10.6 54 74 68
3/15/2005 11.6 56 69 59
3/27/2006 12.7 57 71 62

 

Click for description

Figure 26. Graph. Time-sequence IRI values at section 010418.

 

Because the application of a slurry seal had a noticeable impact on the IRI at all sections in the Arizona SPS-1 project shown in table 24, only the data up to the application of the slurry seal were considered for analysis.

All sections in the SPS-1 project in Montana were subjected to an aggregate seal on 7/24/2004. An evaluation of the IRI values before and after the aggregate seal was performed to assess the effect of the aggregate seal on IRI. At seven sections, the IRI along the right wheelpath decreased after the application of the aggregate seal, with an average reduction in IRI of 7 inches/mi. The reduction in IRI was greater for the sections that had higher IRI values. The reduction in IRI was attributed to the aggregate seal covering the distress that was present along the right wheelpath. At the other five sections, the IRI of the right wheelpath increased after the application of the aggregate seal, with an average increase in IRI of 4 inches/mi. For the center of the lane, the aggregate seal caused the IRI to increase at all sections, while for the left wheelpath, the aggregate seal caused the IRI to increase at all sections except for three. The average increase in left wheelpath IRI and CLIRI were 3.3 and 4.7 inches/mi, respectively.

The time-sequence IRI values at section 300119 in Montana that was subjected to an aggregate seal on 7/24/2004 are shown in table 26, while figure 27 shows a plot of the time-sequence IRI values at this section. Evaluation of IRI values obtained before and after the aggregate seal showed the aggregate seal reduced the right wheelpath IRI by 8 inches/mi, increased CLIRI by 4 inches/mi, and caused no change in the left wheelpath IRI.

Table 26. Time-sequence IRI values at section 300119.

Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
11/19/1998 0.1 62 69 60
5/22/1999 0.6 61 66 57
7/10/2000 1.8 64 64 57
8/10/2001 2.9 61 64 59
9/25/2002 4.0 62 65 57
8/27/2003 4.9 63 79 58
7/21/2004 5.8 64 104 59
7/24/2004 Aggregate seal
8/16/2004 5.9 64 96 63
6/4/2005 6.7 65 103 66
7/26/2008 9.8 70 103 63
7/18/2009 10.8 71 111 66
7/17/2010 11.8 75 109 69

 

Click for description

Figure 27. Graph. Time-sequence IRI values at section 300109.

 

The time-sequence IRI values at section 300123, another section in the Montana SPS-1 project that was subjected to an aggregate seal on 7/24/2004, are shown in table 27, while figure 28 shows a plot of the time-sequence IRI values at this section. Evaluation of IRI values obtained before and after the aggregate seal showed the aggregate seal increased the IRI of all three paths, with an increase in IRI of 6, 4, and 7 inches/mi for the left wheelpath, right wheelpath, and center of the lane, respectively.

Table 27. Time-sequence IRI values at section 300123.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
11/19/1998 0.1 48 55 48
5/22/1999 0.6 49 51 48
7/10/2000 1.8 50 51 48
8/10/2001 2.9 51 52 47
9/25/2002 4.0 54 53 50
8/27/2003 4.9 55 62 50
7/21/2004 5.8 55 63 51
7/24/2004 Aggregate seal
8/16/2004 5.9 61 67 58
6/4/2005 6.7 61 68 60
7/26/2008 9.8 64 69 65
7/18/2009 10.8 64 69 68
7/17/2010 11.8 65 72 72

 

Click for description

Figure 28. Graph. Time-sequence IRI values at section 300123.

 

As seen in figure 27 and figure 28, the change in IRI caused by the application of the aggregate seal was small. Therefore, the evaluation of the time-sequence IRI data at these two sections would provide information about the long-term trend of IRI progression. Similar trends were observed for the other sections in the Montana SPS-1 project. Therefore, the research team decided to use all available IRI data for analyzing the IRI changes at all SPS-1 sections in the Montana SPS-1 project.

For the section in Kansas listed in table 24, the aggregate seal had been applied after the section was overlaid. Because only the IRI data up to the overlay were used in the analysis, the application of the aggregate seal had no impact on the data selected for analysis.

For the sections in Texas listed in table 24, the aggregate seal was applied after the second overlay had been placed on the sections. Because data after the second overlay were not analyzed in this study, the application of the aggregate seal had no impact on the data used for analysis.

Variability in the Profiled Path

On a pavement that had significant transverse variability in IRI, variability in the longitudinal path followed by the profiler could have a significant impact on IRI. Also, on a pavement that had cracking along the wheelpaths, transverse variability between runs could cause the distress to be captured or missed, thus affecting IRI. A section that had significant rutting could also result in variability in the IRI values among the runs. Evaluation of time sequence data at SPS-1 sections indicated only a very few cases where the IRI on a particular date would be much higher or lower than that for the preceding and subsequent profile date. Pavements that had cracking along the wheelpaths generally had high IRI values, and although there could be variability among the runs, these differences would get averaged out when the IRI values from repeat runs were averaged. Therefore, for pavements that had an accelerating trend in IRI, the time sequence IRI plots did not typically show an IRI value that did not fit the trend in roughness development.

All sections in the Louisiana SPS-1 project had one data point that clearly had been affected by transverse variability in the profiled path. The time-sequence IRI values at section 220116, which was a test section in the Louisiana SPS-1 project, are shown in table 28, while figure 29 shows a plot of the corresponding time-sequence IRI values.

Table 28. Time-sequence IRI values at section 220116.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
11/17/1997 0.4 45 39 34
10/16/2004 7.3 48 44 45
8/7/2006 9.1 53 46 43
9/25/2007 10.2 49 47 47
4/9/2008 10.8 50 48 46
2/3/2010 12.6 65 68 60
3/16/2011 13.7 52 50 45
1/16/2012 14.6 54 50 48

 

Click for description

Figure 29. Graph. Time-sequence IRI values at section 220116.

 

The IRI values on 2/3/2010 for all three paths were higher than the previous and subsequent IRI values. All other sections in the Louisiana SPS-1 project that were profiled on this date showed a similar trend in IRI. No maintenance activities had been performed at this section. This section did not have any distress along the wheelpaths, and rutting on this section was less than 0.25 inches. Evaluation of the profile data did not show any cause for the high IRI values along all three paths on 2/3/2010.

Table 29 shows the IRI values of the nine runs obtained at section 220116 on 2/3/2010, and a plot of these IRI values is shown in figure 30. The IRI values showed considerable variability for all three paths. Therefore, the cause for the high IRI values on 2/3/2010 was attributed to variability in the profiled path that affected the average IRI computed for this date. Runs 7 and 8 were the only runs made on 2/3/2010 that had IRI values that were close to the IRI values obtained for the previous and subsequent profile dates.

Table 29. IRI values obtained at section 220116 on 2/3/2010.

Run IRI (Inches/mi)
Left Right Center
1 82 83 85
2 54 57 52
3 58 55 42
4 72 71 68
5 76 81 74
6 73 83 65
7 49 46 45
8 51 54 43
9 75 84 70

 

Click for description

Figure 30. Graph. IRI values obtained at section 220116 for different runs on 2/3/2101.

 

The IRI data for 2/3/2010 were omitted from analysis at all Louisiana test sections. This was the only case where data were omitted from analysis because of transverse variability in the profiled path. There were several cases where the standard deviation in IRI for the repeat runs for a particular visit was high, and these usually occurred at sections that had high IRI values. However, evaluation of the time-sequence IRI data at these sections did not show an IRI data point that did not fit the trend in IRI progression at the site.

Increase in Distress Affecting Roughness at a Section

Table 30 shows the time sequence IRI values at section 320107, and a rapid increase in IRI was seen at this section for the two wheelpaths after 8/4/2004. The time sequence IRI values at this section are shown in figure 31. Figure 32 shows a continuous IRI plot for the left wheelpath based on a 10-ft base length for the last three profile dates shown in table 30. This plot shows how the roughness progressed for the left wheelpath at this section for the last three profile dates, which caused the overall IRI for the left wheelpath to increase from 63 inches/mi on 8/4/2004 to 119 inches/mi on 6/25/2009. The plot shows there were several localized areas within the section that had very high roughness values, and the roughness at these locations increased with time. Figure 33 shows a plot of the left wheelpath profile data collected at this section on 6/25/2009. The downward spikes in the plot represent transverse cracks, and the data showed that a dip occurred at the crack locations. The high localized roughness areas seen in figure 32 correspond to these cracks seen in figure 33. Hence, the large increase in IRI that was observed at this section along the wheelpaths was attributed to the dips occurring at the crack locations.

Table 30. IRI values obtained at section 320107.

Profile Date Pavement Age (Years) IRI (Inches/mi)
Left Right Center
4/22/1997 1.6 48 59 52
11/18/1997 2.2 48 58 53
8/28/1998 3 47 60 52
10/16/1999 4.1 45 60 52
6/14/2000 4.8 44 59 51
6/19/2001 5.8 46 58 52
6/10/2002 6.8 48 60 53
7/31/2002 6.9 48 60 53
10/13/2003 8.1 55 62 52
4/15/2004 8.6 60 64 54
8/4/2004 8.9 63 67 55
8/27/2006 11 94 84 63
6/25/2009 13.8 119 107 81

 

Click for description

Figure 31. Graph. Time sequence IRI values obtained at section 320107.

 

Click for description

Figure 32. Graph. Continuous IRI plot for left wheelpath of section 320107.

 

Click for description

Figure 33. Graph. Left wheelpath profile data collected on 6/25/2009 at section 320107.

 

Effect of Crack Sealing

Cracks could appear as downward features on the profile data, with the magnitude of the depth of the downward feature depending on the depth of the crack. Although profilers record data at 0.98-inch intervals, the height sensor reading that was used to compute the profile elevation at 0.98‑inch intervals was an average elevation over 0.98 inch. Therefore, narrow cracks that were not deep might not appear in the profile because of this averaging. Also, for a narrow and deep crack, the magnitude of the depth of the crack in the profile would be less than the actual depth of the crack because of this averaging.

The effect of cracks on IRI would depend on the width of the crack, depth of the crack, whether a dip occurs at the crack, and the number of cracks traversed when collecting the profile data. Sealing the cracks would eliminate the effect of the crack on the profile data and might result in a reduction in IRI of the section, depending on the depth and width of the crack and the number of cracks traversed when collecting profile data. However, if there was a dip at the crack location, sealing a crack might not have much of an effect in reducing the IRI. The IRI could increase after crack sealing if the cracks were overfilled, which could cause a slight bump at each crack location. Crack sealing of transverse cracks that spanned the entire lane could affect the IRI of the left wheelpath, right wheelpath, and center of the lane. Crack sealing performed on longitudinal cracks or alligator cracks present along a wheelpath could affect the IRI of the wheelpath. Crack sealing performed along longitudinal cracks that are present along the center of the lane could affect CLIRI.

An evaluation was performed at sections that had been subjected to crack sealing to evaluate the impact of crack sealing on IRI. There were not many sections where sealed transverse cracks or sealed cracks along the wheelpath had been recorded. An evaluation of IRI before and after crack sealing at several of these sections did not provide a clear indication that IRI was affected by the crack sealing. Therefore, it was decided to use all available IRI values at sections that had crack sealing performed for analysis.

DATASET FOR ANALYSIS

An IRI dataset that was suitable for analysis was assembled after evaluation of the time-sequence IRI values. For sections where maintenance or rehabilitation affected the IRI as described previously in this chapter, only the IRI data before the maintenance or rehabilitation were considered for analysis.

Data analysis feedback reports were provided to LTPP indicating cases where equipment errors were observed in the data and for cases where data had been assigned to a section from an incorrect location so that the erroneous data could be removed from the PPDB.

As indicated in the previous section entitled Overlay Placed on the Section, the SPS-1 projects in Michigan and Texas have been subjected to two overlays. For these two SPS-1 projects, data up to the first overlay and data after the first overlay up to the second overlay were treated separately.

Appendix B of this report contains a table that shows the details about the dataset that was used for analysis. The following information is shown in this table for each SPS-1 test section:

The following observations regarding deassigning of SPS-1 test sections were noted based on the information in the SDR 28 EXPERIMENT SECTION table in the PPDB:

In an SPS-1 project, all test sections were not necessarily deassigned on the same date. Individual sections could be deassigned at different times when they were rehabilitated. In such projects, the period over which profile data were collected at test sections in the project could vary, with test sections that were deassigned earlier being monitored over a shorter time period when compared with sections that were deassigned later.

As shown in table 2, 18 SPS-1 projects were built. Because each project had 12 test sections, this meant a total of 216 core test sections were constructed for the SPS-1 experiment. Center of the lane data were never collected at 15 test sections because these test sections had been deassigned before the K.J. Law T-6600 profiler that had the capability to collect center of the lane data was used in the LTPP Program. The test sections for which center of the lane data were never collected are shown in table 31.

Table 31. Test sections where center of the lane data were never collected.

State State Code Sections
Arkansas 05 19, 20
Kansas 20 1, 2, 5, 6, 7
Michigan 26 13, 14, 19, 22
Ohio 39 1, 2, 5, 7

 

Table 32 shows the following information for each SPS-1 project: age of the pavement when center of the lane data were first collected at the project, average age of the sections corresponding to the last profile date used for analysis, and the range in age of the sections at the last profile date used for analysis. For sections that were subjected to maintenance or rehabilitation that affected its IRI value, the last profile date used for analysis was the date when data were last collected at the section before the maintenance or rehabilitation activity. For sections that were not subjected to maintenance or rehabilitation, the last profile date used for analysis was the date when data were last collected at the section. The average age of the sections corresponding to the last profile date used for analysis shown in table 32 is the average value for all sections and provides a general idea about the age of the sections in the project at the last profile date used for analysis. The range in age of the sections at the last profile date shown in table 32 indicated the range in age of the test sections in the SPS-1 project at the last profile date used for analysis.

Table 32. Average age of sections in SPS-1 projects at last profile date used for analysis.

SPS-1 Project Pavement Age When
Center of the Lane
Data Were First
Collected (Years)
Average Age of
Sections at Last
Profile Date Used
for Analysis (Years)
Range of Average Age
of Sections at Last
Profile Date Used for
Analysis (Years)
Alabama 4.3 11.4 4.9–12.2
Arizona 3.5 10.3 8.6–12.7
Arkansas 2.8 12.4 9.6–12.7
Delaware 1.1 10.1
Florida 1.2 16.3
Iowa 4.3 8.0
Kansas 3.3 7.5
Louisiana 0.4 14.6
Michigan 1.6 6.6
Montana 0.1 11.8
Nebraska 1.6 4.7 4.6–4.9
Nevada 1.6 12.1 8.6–13.6
New Mexico 1.4 10.5
Ohio 3.0 10.1 6.0–10.6
Oklahoma 0.4 13.1 9.8–13.7
Texas 1.3 4.3
Virginia 4.6 16.1 8.4–17.4
Wisconsin 0.1 10.4 7.9–10.6
—Indicates all sections were the same age at last profile date.

 

Table 33 shows the SPS-1 projects classified according to the average age of the sections at last profile date used for analysis. As shown in table 33, the average age of the sections at last profile date used for analysis was less than 8 years for five SPS-1 projects, between 10 and 12 years for seven SPS-1 projects, and between 12 and 18 years for six SPS-1 projects.

Table 33. SPS-1 projects classified based on average age of sections at last profile date.

Average Age of Sections at
Last Profile Date Used for
Analysis (Years)
SPS-1 Project
4–6 Nebraska, Texas
6–8 Iowa, Kansas, Michigan
8–10 None
10–12 Alabama, Arizona, Delaware, Montana, NewMexico, Ohio, Wisconsin
12–14 Arkansas, Nevada, Oklahoma
14–16 Louisiana
16–18 Florida, Virginia

 

Figure 34 shows a cumulative frequency distribution of the age of the test sections at last profile date used for analysis. Data for 201 test sections where center of the lane data were collected were used to generate this plot. It shows the pavement age of the test sections at the last profile date used for analysis was less than 6 years for 12 percent of the sections, between 6 and 10 years for 23 percent of the sections, and greater than 10 years for 65 percent of the sections.

Click for description

Figure 34. Graph. Cumulative frequency distribution of sections based on pavement age at last profile date used for analysis.

 

Table 34 shows the average time span over which center of the lane data were collected at each SPS-1 project. The average time span for an SPS-1 project was computed by obtaining the difference in years between the last profile data used for analysis and the first date when center of the lane data were collected at each section in the SPS-1 project and then averaging the values. Table 34 also shows the range in time span when center of the lane data were collected for the sections in the SPS-1 project. As described previously, the time span over which center of the lane data were collected for the various sections in an SPS-1 project could be different because sections could be deassigned at different times, or maintenance or a construction activity that affected the IRI of the section might have occurred for different sections at different times.

Table 34. Time span over which center of the lane data were collected at SPS-1 projects.

SPS-1
Project
Pavement Age
When Center of the
Lane Data Were
First Collected
(Years)
Average Time Span
Over Which Center of
the Lane Data Were
Collected (Years)
Range of the Time
Span Over Which
Center of the Lane
Data Were Collected at
Sections (Years)
Alabama 4.3 7.0 0.6–7.8
Arizona 3.5 6.8 5.1–9.2
Arkansas 2.8 9.6 6.7–9.9
Delaware 1.1 9.0
Florida 1.2 15.1
Iowa 4.3 3.7
Kansas 3.3 4.2
Louisiana 0.4 14.2
Michigan 1.6 5.0 3.6–5.1
Montana 0.1 11.7
Nebraska 1.6 3.1
Nevada 1.6 10.4 7.0–12.6
New Mexico 1.4 9.2
Ohio 3.0 7.1 3.0–13.5
Oklahoma 0.4 12.7 9.4–13.3
Texas 1.3 3.0 1.9–3.5
Virginia 4.6 11.4 3.8–12.8
Wisconsin 0.1 10.3 7.8–10.5
—Indicates all sections monitored with center of the lane data over same time span.

 

Table 35 shows the SPS-1 projects classified according to the average time span over which center of the lane data were available for the project. As shown in table 35, the average time span over which center of the lane data were collected was less than 6 years for five SPS-1 projects, between 6 and 10 years for six SPS-1 projects, and over 10 years for seven SPS-1 projects.

Table 35. SPS-1 projects classified based on average time span over which center of the lane data were available.

Average Period Over
Which Center of the Lane
Data Collected (Years)
SPS-1 Project
2–4 Iowa, Nebraska, Texas
4–6 Kansas, Michigan
6–8 Alabama, Arizona, Ohio
8–10 Arkansas, Delaware, New Mexico
10–12 Montana, Nevada, Virginia, Wisconsin
12–14 Oklahoma
14–16 Florida, Louisiana

 

Figure 35 shows a cumulative frequency distribution of time span over which center of the lane data were collected at the test sections. Data for 201 test sections where center of the lane data were available were used to generate this plot. It shows that the time span over which center of the lane data were collected was less than 5 years for about 25 percent of the test sections, between 5 and 10 years for about 35 percent of the test sections, and greater than 10 years for about 40 percent of the test sections.

Click for description

Figure 35. Graph. Cumulative frequency distribution of sections based on time span over which center of the lane data were collected.

 

 

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101