Standard Practice for
Operating Inertial
Profiling Systems
AASHTO Designation: R 57-141
American Association of State Highway and Transportation Officials
444 North Capitol Street N.W., Suite 249
Washington, D.C. 20001
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
Standard Practice for
Operating Inertial Profiling Systems
AASHTO Designation: R 57-14 1
1. SCOPE
1.1. This practice describes the procedure for operating and verifying the calibration of an inertial
profiling system. This practice is meant to be performed as a quality control/quality assurance
(QC/QA) test for use with the appropriate smoothness specification for paving operations and for
network-level data collection. It is not meant to be required as a QA procedure to monitor daily
paving operations; however, it is recommended when inertial profiling systems are used for QC
testing.
2. REFERENCED DOCUMENTS
2.1. AASHTO Standards:
M 328, Inertial Profiler
R 56, Certification of Inertial Profiling Systems
2.2. ASTM Standards:
E 867, Terminology Relating to Vehicle-Pavement Systems
E 1926, Standard Practice for Computing International Roughness Index of Roads from
Longitudinal Profile Measurements
E 2560, Standard Specification for Data Format for Pavement Profile
2.3. Other Document:
Sayers, M. W. “On the Calculation of International Roughness Index from Longitudinal Road
Profile.” Transportation Research Record 1501. TRB, National Research Council,
Washington, DC, 1995
3. TERMINOLOGY
3.1. Definitions:
3.1.1. International Roughness Index (IRI)—a statistic used to determine the amount of roughness in a
measured longitudinal profile. The IRI is computed from a single longitudinal profile using a
quarter-car simulation at 50 mph (Sayers 1995). Computer programs to calculate the IRI from a
longitudinal profile are referenced in ASTM E 1926.
3.1.2. longitudinal profile—the vertical deviations of the pavement surface taken along a line in the
direction of travel referenced to a horizontal datum.
3.1.3. roughness—according to ASTM E 867, the deviation of a surface from a true planar surface with
characteristic dimensions that affects vehicle dynamics, ride quality, dynamic loads, and drainage.
TS-5a R 57-1 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
4. SIGNIFICANCE AND USE
4.1. This practice outlines standard procedures for operating inertial profiling systems for project-level
QC/QA determinations and for network-level data collection and analyses.
5. EQUIPMENT
5.1. Minimum Requirements—The inertial profiling system must meet all requirements and
specifications found in M 328 and must be currently certified in accordance with R 56.
Note 1—For consistent pavement profile determination, maintain air pressure in the tires of the
host vehicle according to the manufacturer’s specification. The host vehicle and all system
components will be in good repair and proven to be within the manufacturer’s specifications. The
operator of the inertial profiler shall have all tools and components necessary to adjust and operate
the inertial profiler according to the manufacturer’s instructions.
5.2. Repair, Adjustment, and Upgrade of Inertial Profilers:
5.2.1. Major component repairs or replacement to an inertial profiler that would require the
recertification of the equipment include, but are not limited to, the following:
The accelerometer and its associated hardware;
The noncontact height sensor and its associated hardware;
Any printed circuit board necessary for the collection of raw sensor data or the processing of
the inertial profiles;
Change of host vehicle in the case of permanently mounted equipment; or
Major repairs to the host vehicle.
5.2.2. The operator of the inertial profiler will be allowed to make minor adjustments to the equipment
without having to complete the recertification process as long as the adjustments allow the
equipment to fulfill the verification of calibration process in Section 5.3. Minor adjustments to the
system include, but are not limited to, the following:
Inspecting, resoldering, or replacing connectors;
Cleaning components, normal adjustments to power supply voltage levels as required by the
manufacturer;
Setting software parameters and/or scale factors as required by the manufacturer in a routine
calibration procedure; or
Mounting on different suitable host vehicle(s) in the case of portable equipment.
5.3. Verifying Calibration—The following verification procedure is required for an inertial profiler
used for QA testing of smoothness of new construction and network-level profiling, and is
recommended when an inertial profiler is to be used as a QC instrument on a daily basis.
5.3.1. Standards:
[Link]. Longitudinal:
[Link].1. The longitudinal verification standard shall be a straight and level roadway test section of at least
528 ft in length with sufficient lead-in distance for the vehicle to attain a constant speed before the
start of the test section and a safe stopping distance.
[Link].2. Measure this length accurately to within 0.05 percent using a measurement tape since the distance
to be measured is the travel length the wheels encounter.
TS-5a R 57-2 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
[Link]. Vertical:
[Link].1. The vertical measurement standard shall be flat plates or blocks of known thickness and low
thermal expansion.
[Link].2. Mark the blocks with the known thickness. Certify the block thickness accurate to within 0.001 in.
As a minimum, test a smooth base plate and both a 1-in. and a 2-in. gauge block.
5.3.2. Procedure:
[Link]. Frequency of Verifying Calibration:
[Link].1. For QC and QA testing of smoothness of new construction, perform the longitudinal and vertical
verification of calibration of the inertial profiler prior to use on each paving project in accordance
with the manufacturer’s recommendation and at such times as the Owner-Agency determines that
verification is necessary. For network-level profiling, it is recommended that the longitudinal
verification of calibration and the block check for verification of vertical calibration be performed
at least monthly, and the bounce test for verification of vertical calibration be performed daily.
[Link].2. Check the cold tire air pressure on the wheels of the host vehicle at least daily and maintain
according to the manufacturer’s recommendations.
[Link]. Longitudinal Verification of Calibration:
[Link].1. Perform the longitudinal verification of calibration by navigating the inertial profiler over a
measured test section of at least 528 ft.
[Link].2. The inertial profiler’s distance measuring subsystem, after tire and electronic warm-up in
accordance with the manufacturer’s recommendations, must measure the length of the test section
to within 0.15 percent of its actual length.
[Link].3. If the specified criterion is not met, adjust the inertial profiler’s distance measurement subsystem
according to the manufacturer’s guidelines as needed.
[Link].4. Failure to meet the specified tolerance, after attempts to adjust have been performed, will require
recalibration, and then reverification as described in Section [Link].
[Link]. Vertical Verification of Calibration—Vertical verification of calibration consists of performing
two tests, the block test and the bounce test.
[Link].1. Block Test—Vertical height sensor check tests shall be performed in accordance with the
manufacturer’s recommended procedures. In the absence of manufacturer’s procedures, perform
the block test as described in Section [Link].[Link] height sensor check tests are run after the
profiler has reached operational stability as specified by the manufacturer. This test will be
conducted with the inertial profiler on a relatively flat and level area. Its purpose is to check the
height measurements (in inches or millimeters) from the height sensor(s) of the test vehicle using
blocks of known heights. During the test, do not lean on the profiler or cause it to move in any
way. Under windy conditions, it may be necessary to perform this test indoors. The test procedure
consists of the following steps:
Position a smooth base plate under the height sensor of the profiler and allow the system to
take height measurements.
Position a 0.25-in. block underneath the height sensor on top of the base plate and allow the
system to take height measurements.
TS-5a R 57-3 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
Carefully remove the 0.25-in. block from the base plate and replace it with a 0.50-in. block.
Make another set of height measurements.
Carefully replace the 0.50-in. block with a 1.00-in. block and take another set of height
measurements.
Finally, carefully replace the 1.00-in. block with a 2.00-in. block and take the last set of height
measurements.
As a minimum, test a base plate and both the 1- and 2-in. measurement blocks. If the equipment
fails the minimum test, then perform the full range of tests to determine system linearity problems,
stand-off problems, or complete system failure.
The owners of the profiler must furnish their own base plate and gauge blocks. Measure the
thickness of the gauge blocks at three different positions on each side of the block with a device
capable of measuring to the nearest 0.001 in. For each block, an average thickness shall be
determined from the measurements made, which shall be used in checking the height sensors as
described in this test. The average thickness shall be marked on each gauge block.
The operator of the profiler should tabulate the measurements and record them in a calibration log.
Determine the difference between each measurement on a gauge block and the average of the ten
measurements on the base plate to get the thickness of the gauge block as measured by the height
sensor. Repeat this calculation for each gauge block. Determine the absolute values of the
differences between the computed block thickness and the known average block thickness. The
absolute difference should be less than or equal to 0.01 in. for each gauge block.
[Link].2. Bounce Test—To perform the bounce test, position the host vehicle on a surface as flat and level
as possible. Power the system and ensure that the profiler has reached operational stability as
specified by the manufacturer. Center a thin, smooth, flat, nonglossy material plate under each
sensor. Using the equipment’s normal data collection software, initiate a data collection run using
a simulated travel speed at the midpoint of the manufacturer’s recommended data collection speed
range. (The only difference between a bounce test and a normal data collection run is that there is
an artificial longitudinal travel signal supplied and the vehicle is not actually traveling along the
road. The bounce test utilizes the same data collection software and routines used during normal
data collection.)
The bounce test will require a profile collection run with a minimum simulated travel distance of
2184 ft. Allow the profiler to collect a minimum of 828 ft of static profile with the host vehicle as
motionless as possible. Next, the sensor(s) should be moved vertically for a total displacement of
approximately 1 to 2 in. (a yardstick may be helpful until the operator gets used to the procedure)
keeping the sensors as close to perpendicular to the surface as possible during this movement. This
movement must continue until a minimum of 528 ft of simulated longitudinal distance has been
covered.
The typical method for full-size, high-speed host vehicles is to push the mounting system
(bumper) down an inch or so and let the vehicle suspension rebound to create the total vertical
travel of 1 to 2 in. The typical method for lightweight, slow-speed host vehicles is to stand toward
the center of the vehicle platform and hop up and down such that all four corners of the vehicle
suspension travel approximately 1 to 2 in. vertically. After a minimum of 528 ft of bounce profile
is collected, allow the profiler to collect an additional minimum of 828 ft of static profile. The
profiles shall then be saved and analyzed using the latest version of ProVAL software to compute
the continuous IRI with a 528-ft base length for each profile collected.
When reviewing the analysis results, ignore the first and last 300 ft of the profile as it is lead-in
and lead-out distance. Ensure the remaining static portions result in an IRI of less than 3 in./mile
and the bounce portion IRI is less than 8 in./mile. The two static IRI portions should be about the
same. This requirement applies to each sensor in the profiler. See Appendix X1 for further details
concerning the bounce test.
Note 2—The Profile Viewer and Analysis (ProVAL) software program developed for the Federal
Highway Administration, which can be used to import, display, and analyze the characteristics of
pavement profiles from many different sources, is available at [Link].
TS-5a R 57-4 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
[Link]. Daily Measurement Control—The operator should select a reasonable method to verify daily that
the inertial profiling system utilized can produce consistent results under normal operating
conditions on pavements typically encountered during data collection. Options include using a
control section as described in Section [Link].1 or checks using the previous day’s data as
described in Section [Link].2. The surface texture of the section used for daily verification should
be similar to that on which measurements are to be obtained for QC/QA purposes.
[Link].1. For profilers used for QC/QA of new construction, control sections can be established by selecting
one or more 0.1-mile sections with a maximum IRI of 120 in./mile that will maintain a consistent
ride profile over the time period when daily checks are needed. For profilers used for network-
level profiling, several control sections having a range of IRI up to 200 in./mile may be selected as
control sections, and profiled at regular intervals. A profiler certified within the past 90 days
should determine the IRI of the control section(s) by making a series of at least five profile
measurements. The average IRI of the measurements shall be used to establish the IRI of the
control section(s), provided that the cross-correlation of the measurements, as determined using
the latest version of ProVAL, is at least 88 percent (value is dependent on the filters used, spectral
content of the measured surface, operator, etc.). Once established, the control section(s) can be
used to validate that an inertial profiler is operating properly on a daily basis for profilers used for
QC/QA of new construction, or on a regular basis for profilers used for network-level data
collection. A control chart should be established and maintained by plotting the runs to determine
profile drift as the section ages. Typically, no single IRI determination should vary more than
5 percent from the original control section IRI. If the distance of the control section is measured
accurately, the control section can also be used to verify the longitudinal calibration of the profiler.
[Link].2. Checks using the previous day’s data can be used when data are being collected on a fairly
continuous basis such as on a construction project or during annual network-level analyses. Under
these circumstances, it is possible to rerun short sections of pavement that were measured on the
previous day for comparison purposes. The current day’s IRI value should not differ by more than
6 percent from the previous day’s IRI value.
[Link].3. Calibration Verification Log—Maintain a log to be kept with the inertial profiler in accordance
with Section [Link] R 56.
6. TEST PROCEDURE
6.1. Table 1 outlines the steps for measuring the pavement profile using an inertial profiler for QC/QA
testing of smoothness of new construction.
TS-5a R 57-5 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
Table 1—Measuring Pavement Profile and Reporting Smoothness Indices for QC/QA of Smoothness of New
Construction
Step Action
1 Clean the roadway path to be measured of all debris and other loose material. Collect data on dry pavement. Collecting data on damp
pavement without any standing water may be acceptable if approved by a representative of the Owner-Agency.
2 When measuring the pavement profile, operate the inertial profiler at a relatively constant speed within the certified range.
All data collected above or below the certified speed range are not acceptable.a
Remeasure any pavement segment where the travel speed of the profiler is less than or exceeds the manufacturer’s recommended
operational speed at any point during data collection.
3 A lead-in length of roadway of up to 450 ft may be required to stabilize the inertial profiler’s filters and achieve the same accuracy in
the first 0.1 mile that is achieved through the rest of the job. The presection length is dependent on the filter type, the grade change on
entering the test segment, and the accuracy required of the first 0.1 mile of measured pavement. Typically, this presection shall be at
least 300 ft in length and located immediately before the section of pavement to be tested. Shorter sections have been used when the
physical constraints of the project required it and the other project conditions made it acceptable.
Take the inertial profiler measurements on one or more longitudinal lines as specified by the Owner-Agency.
If two longitudinal traces are specified, set the sensor path spacing to the values specified by the Owner-Agency. If such is not
specified, sensor path spacing shall be between 65 and 71 in.
If two longitudinal traces are specified, but the profiling system collects profile data only in one longitudinal path at a time, the
longitudinal traces shall be taken in each wheelpath of the lane independently and in the same direction of travel.
4 Preferably collect measurements in the direction of traffic. If this is not practical and data are collected in the other direction, make a
note. Repeat measurements if operator and/or equipment errors are encountered. All runs shall be provided to the agency. When using
an inertial profiler that collects a single trace per pass, take care to ensure that the measurements from each trace in a travel lane start
and stop at the same longitudinal locations. After completing data collection, check roughness indices for reasonableness based on the
operator’s experience. Unreasonable roughness values could be an indication of equipment problems.
5 Data Collection (QA data collection is meant to be performed at the end of the paving operation or staged as prescribed by the Owner-
Agency).
Submit a table to the Owner-Agency that identifies the lane(s), trace(s), and distance location(s) tested for each file created during the
QA testing.
6 Present the profile elevation data to the Owner-Agency in an approved electronic file format. The format must be readable by ProVAL.
If the submitted profile data files are created from raw profile data using an export option in the manufacturer’s software where filter
settings for export can be specified, ensure that the filter settings used to create the data files are similar to the settings that were used to
create data files during certification.b
7 Compute a summary roughness statistic. The recommended index to compute is the continuous IRI reported as a histogram that
presents the percentage of the job that falls within IRI categories selected by the agency.
a
Failure to maintain the minimum certified speed will cause the inertial referencing subsystem to “droop” and the pavement profile elevations will not be usable.
b
The Owner-Agency will calculate the IRI values and any associated pay factors.
Note 3—ProVAL has the ability to calculate the continuous IRI histogram.
6.2. Table 2 outlines procedures to follow when collecting network-level pavement profile data using
an inertial profiler.
Table 2—Measuring Pavement Profile for Network-Level Data Collection
Step Action
1 Warm up the equipment following manufacturer’s recommendations before collecting profile data.
2 Make sure that the lateral spacing between the sensors is set according to agency guidelines.
3 Enter all header information before starting profile data collection.
4 Verify data collection is initiated at the correct location.
5 Drive on the center of the lane.
6 Collect data on dry pavement. Collecting data on a damp pavement without any standing water may be
acceptable based on agency guidelines.
7 Try and maintain a constant speed. Operate the inertial profiler within the speed range recommended by
the manufacturer.
8 Avoid rapid acceleration/deceleration.
9 Mark data collected below the minimum operating speed of the profiler (e.g., due to traffic signals,
traffic congestion), and do not use that data when computing the roughness index. If speed was below
the minimum speed because of traffic congestion, follow agency guidelines regarding reprofiling of that
pavement segment.
10 Monitor for system warnings indicating problems with the height sensor or the accelerometer.
11 Monitor roughness values to see if they are reasonable based on the operator’s experience.
Unreasonable roughness values could be due to an equipment problem. Use recommendations in
Section [Link].1 and [Link].2 to verify the accuracy of collected data
TS-5a R 57-6 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
6.3. To ensure the collection of quality profiles, the host vehicle may need to be operated at the lower
end of the certified speed range when profiling on vertical and horizontal curves and
superelevations for QC/QA testing for smoothness of new construction. Following such a method
is not recommended for network-level profiling. Higher speeds may cause excessive pitch and roll
in the host vehicle, which will adversely affect the integrity of the measurements. In addition,
rapid longitudinal accelerations and decelerations (for example, speed changes such as heavy
braking) may adversely affect the profile measurements.
6.4. Test Data Description and Format—In addition to proprietary file formats, provision shall be
made for export of profile data in the format as per ASTM E 2560 or other ProVAL-compatible
formats.
7. KEYWORDS
7.1. Inertial profiler; International Roughness Index; IRI; ride quality; road profile; roughness;
smoothness.
8. REFERENCES
8.1. NHI Training Course 131100. Pavement Smoothness: Use of Inertial Profiler Measurements for
Construction Quality Control.
8.2. Sayers, M. W., and S. M. Karamihas. The Little Book of Profiling. University of Michigan
Transportation Institute, Ann Arbor, MI, 1998.
APPENDIX
(Nonmandatory Information)
X1. BOUNCE TEST PURPOSE, REASONING, AND ERROR
X1.1. A bounce test is a sanity check on the profiler to ensure that the data from the height sensor and
accelerometer are legitimate and being properly combined to compute the longitudinal profile.
While the host vehicle is stationary, an artificial signal is supplied to the profiler to simulate a
travel speed within the manufacturer’s recommended range (preferably midrange) for collecting
road profile data. A routine data collection run is initiated, after which the suspension on the host
vehicle is excited to cause the sensor(s) to move vertically. Since the sensor footprint is hovering
over the same location, the end result, in theory, should be a perfectly flat and level road profile,
the IRI of which would equal zero if all is working correctly. In practice, a perfectly flat profile
with a zero IRI is not possible. Sources of error in a bounce test include:
Signal noise of the height sensor, accelerometer, and respective circuitry;
Inability to position the host vehicle in such a manner that the sensors are perfectly
perpendicular to a known flat and level surface;
Inability to bounce the host vehicle such that sensors maintain a perpendicular orientation to
the flat and level surface (sensor footprint moves laterally or fore and aft due to pitch and roll
of the host vehicle while bouncing);
Equipment malfunction;
Error in signal and data processing;
TS-5a R 57-7 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.
Accelerometer is mounted such that the axis of the accelerometer is not vertical;
Height sensor and/or accelerometer are not properly secured; and
Height sensor is not mounted at the proper elevation, which may result in the position of the
sensor relative to the pavement surface exceeding the sensor’s measurement range.
X1.2. There are two levels of acceptable error in both static (nonbounce) and bounce profiles in which a
minimum of 528 ft of simulated longitudinal distance is traveled. The limits of IRI of those
profiles are in 3 in./mile and 8 in./mile, respectively.
1
Formerly AASHTO Provisional Standard PP 50. First published as a full standard in 2010.
TS-5a R 57-8 AASHTO
© 2014 by the American Association of State Highway and Transportation Officials.
All rights reserved. Duplication is a violation of applicable law.