Citation
Adaptation of the international roughness index for use on urban roadways in the city and county of Denver, Colorado

Material Information

Title:
Adaptation of the international roughness index for use on urban roadways in the city and county of Denver, Colorado
Creator:
Staley, Brian Joseph
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xii, 83 leaves : ; 28 cm

Thesis/Dissertation Information

Degree:
Master's ( Master of Science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Civil Engineering, CU Denver
Degree Disciplines:
Civil Engineering
Committee Chair:
Rens, Kevin L.
Committee Members:
Janson, Bruce
Kononov, Jake

Subjects

Subjects / Keywords:
Surface roughness -- Measurement ( lcsh )
Roads -- Colorado -- Denver ( lcsh )
Roads ( fast )
Surface roughness -- Measurement ( fast )
Colorado -- Denver ( fast )
International Roughness Index
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 82-83).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Brian Joseph Staley.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
436921364 ( OCLC )
ocn436921364
Classification:
LD1193.E53 2009m S72 ( lcc )

Full Text
ADAPTATION OF THE INTERNATIONAL ROUGHNESS
INDEX FOR USE ON URBAN ROADWAYS IN THE
CITY AND COUNTY OF DENVER, COLORADO
by
Brian Joseph Staley
B.A., University of Colorado Boulder, 2004
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Civil Engineering
2009


This thesis for the Master of Science
Degree by
Brian Joseph Staley
Has been approved
By
-------p-*----
Bruce Janson
7-


Staley, Brian, J. (Master of Science, Civil Engineering)
Adaptation of the International Roughness Index for use on Urban
Roadways in the City and County of Denver, Colorado
Thesis directed by Kevin L. Rens, PhD PE
ABSTRACT
The International Roughness Index (IRI) was developed in 1986 to
generate a longitudinal roadway profile analysis tool for asphalt surfaces.
The IRI is typically applied to freeway and rural highway facilities where
uninterrupted spans of asphalt are expected. This research examined the
application of the IRI to urban roadway facilities. Elements of urban
roadways that influence IRI values include drainage infrastructure, utility
access, and cross traffic approaches. This thesis details the equipment
and methodology used to address challenges in evaluating urban
roadways using the IRI. Alternative road roughness measures are
examined as they compare to the IRI. Repeatability studies were
completed using five different inspectors involved in the data collection
activities. A before repair and after repair study was completed on newly


resurfaced roadways around the CCD in an attempt to establish a set of
baseline values for the application of the IRI to urban roadways. Two
evaluation standards are asserted for use in future roadway evaluations, a
minimum percent-improvement standard, and a three-tiered scaling
system based on straight IRI values. The IRI values for a sample of 33
test cases showed a weighted average post-paving improvement of
36.3%. A minimum standard expectation for percent improvement in
urban asphalt resurfacing is recommended at 20%. Based on IRI data
collected in the context of this study on streets in the CCD recommended
urban IRI (UIRI) values were developed. It is suggested that roadway
surfaces rated as good should show an IRI value of <150 in/mile (2367.44
mm/km) and surfaces rated as acceptable should show an IRI value of
<220 in/mi (3472.24 mm/km).
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
evin L. Rens


ACKNOWLEDGMENT
I would like to thank the City and County of Denver for the funding and
resources supporting this project. The Street Maintenance Division
including William (pat) Kennedy, Brian Roecker, Clayton Goodwin,
Lindsey Van Cleave, and Angie Hager provided valuable assistance in
data collection and input for the project. Rachelle Urso and Darren
Weldon, of the University of Colorado Denver, are also appreciated for
their contributions to the project. Jim Barwick of the City and County of
Denvers Public Works division is also acknowledged as the over all
manager of the CCD UCD research program.
A special thanks to my advisor, Kevin L. Rens, for his support, drive, and
assistance throughout the project. My thanks, also, to my committee
members for each of their contributions to my education; Bruce Janson,
who motivated and guided me throughout my program, and Jake
Kononov, whose thoughtful discussions and academic challenges pushed
me to excel.
I would also like to thank my friends and family who have provided support
and encouragement throughout my education.


TABLE OF CONTENTS
Figures..............................................................x
Tables...............................................................xii
Chapter
1. Introduction ....................................................1
2. Literature Review................................................4
2.1 Historical Context of Profilometric Indexing Systems.............4
2.2 Common Usage of Roughness Indexing in the United States..........5
2.3 Use of IRI Analysis to Differentiate Between Construction Methods ..6
2.4 Current Research on Urban Application of the International
Roughness Index..................................................7
3. IRI Theory ......................................................9
3.1 The Quarter-Car Profile Model....................................9
3.2 The IRI Equation................................................11
3.3 Common IRI Metrics Used on Freeways and Two Lane Rural
Highways .......................................................12
3.4 Alternative Profilometric Indices...............................15
VI


3.4.1 The Half-Car Ride Index..........................................17
3.4.2 The Michigan Ride Quality Index.................................17
3.4.3 The CalPro Simulation Model.....................................18
3.4.4 The Straightedge and Rolling Straightedge Measures..............18
3.4.5 The Ride Number.................................................19
3.4.6 The Performance Index...........................................21
3.5 Limitations of the IRI...........................................21
4. Current Profilometric Practice for the Colorado Department of
Transportation....................................................23
4.1 CDOT Pavement Incentive/Disincentive Provision for Hot Mix Asphalt
Facilities .......................................................24
5. Equipment Used in Assessing the IRI of CCD Roadways..............28
5.1 Profiler Calibration and Performance Standards...................31
5.1.1 Calibration Options Offered by the Ames Engineering Software ...31
5.1.1.1 Reduction Settings............................................32
5.1.1.2 Bump Detection Settings.......................................34
5.1.1.3 Filter Settings...............................................34
5.1.1.4 Rolling Straightedge Simulation Settings......................35
5.1.1.5 CalPro Simulation Settings....................................35
5.2 International Roughness Index Data Output........................35
6. Data Collection Methods..........................................38
vii


6.1 General Methodology...........................................38
6.1.1 Procedural Application of the LISA Model 6200...............39
6.2 Data Repeatability Methodology................................43
6.3 Before and After Comparison Methodology.......................43
7. General Results for Urban International Roughness Index Review of
City and County of Denver Streets.............................45
7.1 Results of the Repeatability Analysis for Segment Data Collection..45
7.2 Results of the Before Repaving and After Repaving Segment
Analysis .....................................................48
7.3 Repaving Methods Used on Test Sites...........................51
7.3.1 IRI Results by Resurfacing Procedure Type...................52
8. Discussion ...................................................55
8.1 Comparison of Acceptable IRI Values for Freeways and Rural
Highways and Acceptable CCD Urban IRI Values..................55
8.1.1 Recommendation for Urban International Roughness Index
Threshold Values............................................57
8.1.2 Analysis of UIRI Percent Improvement Values.................58
8.2 Comparison of Resurfacing Methods Based on Urban Profiling
Results ......................................................59
viii


8.3 Real Impact of Individual Influencing Elements on the Urban
International Roughness Index...................................60
8.4 Using the High Pass Filter Setting to Control Elevation Variation
Influence in IRI Output ........................................62
8.5 Effects of Segment Length and Starting/Stopping Motion on the
Urban IRI ......................................................65
9. Conclusions ....................................................67
Appendix
A. City and County of Denver Urban IRI Segment Data Table.........69
B. Before and After Segment Repaving Data.........................78
Bibliography ........................................................82
IX


FIGURES
Figure
3.1 Quarter-Car Profile Model........................................10
5.1 Pavement Profiler Unit...........................................28
5.2 Detail View of Profiler Laser Track Unit.........................29
5.3 Model 6500 LISA Profiler with Following Safety Vehicle...........30
5.4 Analysis Settings for the Profiling Software Version 5.3.6.......32
5.5 Sample IRI Output Segments From Profiler System Software
Version 5.3.6....................................................33
5.6 Graphical Output of a Direct Profile Measure.....................36
5.7 Direct Profile Output Separated by Laser Track With Overlaid Speed
Profile .........................................................37
6.1 LISA Model 6200 Printer with Sample Output Scroll................40
6.2 Profiler System Software Version 5.3.6 Home Screen...............41
6.3 Project File Selection Screen in the Profiler System Software Version
5.3.6 42
7.1 Graphical Comparison of Before and After Data Locations 1-17..49
7.2 Graphical Comparison of Before and After Data Locations 18-33....50


8.1 Data View Output of Profiler Software Version 5.3.6 With Ignored
Manhole Cover Sections Displayed....................................61
8.2 Detail View of a Disparity Between Data Collection Tracks............62
XI


TABLES
Table
3.1 IRI-based state DOT incentive/disincentive threshold values........14
3.2 Preferred roughness indices by state...............................16
4.1 CDOT incentive/disincentive payment schedule based on HRI
Values ...........................................................24
4.2 CDOT disincentive payment schedule based on percent
improvement.......................................................26
7.1 Repeatability trial data results...................................47
7.2 Weighted average IRI values delineated by resurfacing method......53
7.3 Weighted average IRI values for segments with both before
resurfacing and after resurfacing data, delineated by resurfacing
method ......................................................54
8.1 FHWA scale delineation of mileage for before repair and after repair
conditions........................................................56
8.2 UIRI scale delineation of mileage for before repair and after repair
conditions........................................................58
xii


1. Introduction
The International Roughness Index (IRI) provides a useful tool for the
assessment of pavement roughness on freeways and rural two lane
highways, its use in conjunction with urban roadways, however, is a
concept that has only recently been examined. Addressing the challenges
of collecting pavement roughness data in urban settings requires an
understanding of the factors which influence roughness. These factors
include drainage infrastructure such as cross pans and manhole covers,
and utility access points such as electrical access panels, in-pavement
traffic control devices, and other utility infrastructure inset on the surface of
the road.
In addition to exploring the physical factors influencing pavement
roughness, this thesis will examine the methodological differences between
IRI assessment on highways and urban roadways. These factors include
segment length, start and stop impact, speed of data collection operations,
equipment differences, and threshold limits for good, acceptable, and poor
pavement roughness ratings.
IRI data was collected in the City and County of Denver (CCD) along
urban roadway segments targeted for repaving. The researchers collected
1


data both before and after repaving at several sites in order to begin
establishing a set of standard expectations for urban roadway roughness on
newly repaved surfaces. The methodology and findings of the before and
after study are presented. The CCD was interested in developing a set of
standard expectations for construction contractors through quantifiable IRI
values. They were also interested in identifying potential methodological
improvements in the pavement construction process. A recommendation
was made for establishing a new standard for expected improvement in
pavement roughness after repaving.
The methodologies used by the CCD for resurfacing roadways may
have an impact on pavement roughness. The methods used include a mill
and overlay procedure, a hot in-place recycling process, and a complete
facility reconstruction. Each method should result in a smooth surface
level. This study will compare the IRI results associated with resurfaced
roadways using each of the three methods.
Repeatability exercises were completed to ensure data quality
throughout the project. Researchers completed no fewer than two runs per
lane per segment using different profiler operators for each of the runs.
The well established IRI measurement tool was targeted for use in this
project to assess the quality and roughness of roadways in the Denver
urban environments. The goal of this work was to establish a viable
2


pavement evaluation tool for the CCD for use on its urban roadways to
assist in resource allocation and construction planning practices. The
various influencing elements of urban roadways on the IRI as well as the
methodologies used to control or account for those factors were highlighted.
Various filtration tools were employed to examine particular aspects of
urban factors in pavement roughness analysis.
3


2. Literature Review
The IRI is a pavement monitoring data analysis tool that is both widely
used and widely researched in the engineering field. Its application to
highway facilities is extremely well defined and supported by a prolific body
of research. The application of the IRI to urban environments, however,
has not been widely examined. This potential use for the IRI represents a
relatively new innovation in the application of the IRI by regional and local
transportation authorities.
2.1 Historical Context of Profilometric Indexing Systems
The IRI was developed in 1986 out of a need for a unified analysis tool
for pavement surfaces. Research was commissioned in the 1970s after
several states and countries cited the importance of developing a better
measure for roadway quality (Sayers & Karamihas, 1998). The World Bank
sponsored several studies designed to examine which is more costly: lower
quality, cheaper roads, or higher quality, more expensive roads. The IRI
resulted from this research and is widely regarded as a highly portable and
accurate measure of roadway quality. It is based on correlational analysis
with subjective ridability studies (1998). Juang and Amirkhanian (1992)
4


postulated before its inception on the necessity for a more efficient
analytical means of pavement analysis citing the growing need to assess
and maintain aging highway structures in the United States.
2.2 Common Usage of Roughness Indexing In the United States
Extensive research has been completed on the accuracy and
comparability of the IRI. In the United States of America, several states
offer contractor incentives based on the IRI values assigned to newly paved
or repaved roadway surfaces (Wilde, 2007). These incentives include per
mile, or per unit measured, contract bonuses for new pavement projects
that meet or exceed agreed upon IRI metrics. This practice shows a very
strong confidence in the accuracy and application of the IRI to highways.
Research has been completed on the accuracy and usefulness of the
IRI, showing it to be a strong and viable modeling tool for resource
allocation and planning (Ruotoistenmaki et. al, 2006). By analyzing data
over the course of three years the Ruotoistenmaki team was able to
highlight wasteful spending on unnecessary repaving jobs, this helped
facilitate a more directed approach to infrastructure maintenance.
Additional research has been completed on combining technical pavement
roughness indices with visual pavement condition evaluations in order to
better prioritize pavement maintenance (Bandura and Guaratne, 2001).
5


The Bandura and Guaratne (2001) study focused on generating a predictive
model using fuzzy probability assessments for future pavement conditions
based on current profilometric data. Jiang and Shuo (2005) studied the
incorporation of neural network analysis into predictive modeling of
pavement roughness. They examined the inclusion of traffic capacity and
volumes, weather patterns, and initial IRI data into a cohesive analysis
methodology. Such predictive modeling would allow municipalities to
stream-line maintenance scheduling practices and generate proactive
preventative maintenance program.
Pavement roughness also impacts adjacent research fields. One
study completed in South Africa analyzed the impact of pavement
roughness on vehicle-pavement interaction (Styen, 2001). The real impacts
of varying levels of roughness on vehicle performance for the roadway
surface were assessed. The stated goal of the Styen (2001) study was to
test vehicle-pavement interaction standards to ensure their accuracy with
modern building materials and practices.
2.3 Use of IRI Analysis to Differentiate Between Construction
Methods
A recent study out of Canada compared commonly used pavement
crack treatments (Soleymani, et. al, 2008). The IRI metric was used as the
6


measuring device to assess both spray patches and asphalt mix-patches.
The work sought to not only assess the longevity and quality of individual
crack treatments, but also to analyze which method had the lesser impact
on final IRI values.
2.4 Current Research on Urban Applications of the International
Roughness Index
A large body of research was not discovered regarding the use of the
IRI or other pavement roughness indices in urban settings. Only recently
have transportation researchers focused on the potential usefulness of the
IRI in reviewing urban roadway structures. One such study was completed
recently on urban roadways in Canada. This study focused on the factors
that affect the use of the IRI metric on urban roadways (Reggin, et. al,
2008). Some of the major considerations identified in that research include
false grades introduced to facilitate urban drainage, railway crossings, and
short data collection segments (Reggin, et. al, 2008). The study focused
specifically on element analysis, which may be completed on roadway
features in order to better understand the real impact of individual features
on the final IRI. Intersection rutting was also explored in the Reggin et. al
study relative to its impact as part of a cross street element in the data
collection run. The researchers developed a control equation for railroad
7


crossings based on making an estimate of the influence of the crossing.
This equation is called the Network IRI and was adapted from an example
given in the Reggin, et. al study and given in Equation 2.1 (2008):
Network IRI = (IRI)(L)-(1(S m/kmKd)(n)
(2.1)
Where:
IRI = International Roughness Index (m/km)
L = Length of Segment (km)
d = Average Length of Railroad Crossings
n = Number of Railroad Crossings in Segment
Equation 2.1 is used to adjust urban roadway IRI results to more closely
resemble conventional highway IRI results. The discussion of short
segment lengths impact on final IRI values cites the beginning and end of
data collection runs as problematic. It is asserted that, for longer data
collection segments, the impact of the start and stop spikes in the IRI
values would be reduced as they were averaged across the span of the
data run (Reggin, et. al, 2008). On shorter segments, the length across
which the start and stop spikes were averaged more severely impacted.
8


3. IRI Theory
The IRI was developed in the 1980s as a result of the need for a
unified and repeatable pavement roughness evaluation tool. The National
Cooperative Highway Research Program (NCHRP) was initially responsible
for researching a practical solution and creating such a tool. The World
Bank commissioned research on the topic to develop a tool to assess
roadway quality in developing nations and those countries requiring
infrastructure development assistance (Sayers & Karmihas, 1998). This
research investment was based on the assertion that poor road quality was
more costly to these developing nations as a whole. This measure of cost
was based not only on initial road construction investment, but also on fuel
efficiency, vehicular damage, and other user costs. Representatives of the
World Bank recognized the usefulness of a portable and universal
roughness measure and the final product of the research into such a tool
was presented in the form of the IRI (Sayers & Karmihas, 1998).
3.1 The Quarter-Car Profile Model
The IRI equation was based on the quarter-car profile model, meaning
that it may be derived from a single point, or single wheel, profilometer
9


(Awashthi, 2003). The quarter-car profile model operates by measuring the
movement of the vehicle, or sprung mass, relative to the vehicle tire, or
unsprung mass, along a differential system shown in Figure 3.1.
Figure 3.1 Quarter-Car Profile Model (Awashthi, 2003)
It is more commonly applied with a half-car profile model, which uses two
tracks of data collection, one along each wheel path, and provides an
averaged result of the two when reporting the final segment IRI (Awashthi,
2003). The IRI is gauged from the vertical motion of the data collection
device rather than a direct profile of the road surface.
10


3.2 The IRI Equation
The IRI is calculated exclusively through the use of computer software.
The integral equation examines 0.1 in (0.25 mm) segments in series to
generate a contiguous longitudinal profile of the roadway.
The standard IRI equation is given in Equation 3.1 (Souza, et. al,
2007):
Where:
IRI = International Roughness Index (in/mi or mm/km).
L = length of the section (ft or m).
V = speed of the quarter car model (ft/sec or m/s).
X = longitudinal distance of segment (ft or m).
(3.1)
zu vertical speed of the sprung mass in the quarter-car model
diagram.
zs = vertical speed of the unsprung mass in the quarter-car model
diagram.
dt = the time increment.
11


Equation 3.2 helps illustrate the distinction between a direct profilometric
measure, such as the rolling straightedge measure, and an indirect
profilometric measure. Indirect measures, like the IRI, measure the profiler
response to the pavement surface rather than the pavement itself.
3.3 Common IRI Metrics Used on Freeways and Two Lane Rural
Highways
The IRI is widely used as a measuring tool which gauges performance
quality of contractors responsible for laying asphalt and concrete surfaces.
The IRI metric sets perfectly smooth surfaces at 0 in/mi (0 mm/km),
therefore, the higher the IRI value, the rougher the road. The Federal
Highway Administration (FHWA) classifies IRI values of less than 95 in/mile
(1506.9 mm/km) as good, and IRI values of less than 170 in/mile (2696.5
mm/km) as acceptable for highway installations (FHWA, 2006). Each state
department of transportation (DOT) has a unique scaling system to provide
contractor incentive/disincentive pay scales based on the IRI and other
longitudinal profiling measures. Most of the incentive scaling systems,
based on the IRI, hold contractors to a slightly more rigorous scale with
disincentives starting at an average IRI value of 76 in/mi (1200 mm/km)
(Nemmers, et. al, 2006). Table 3.1 shows a listing of states currently using
12


the IRI as an incentive/disincentive measurement device and the threshold
ranges used. The Colorado Department of Transportation (CDOT)
contractor incentive/disincentive program will be discussed in Chapter 4.
13


Table 3.1 IRI-based state DOT incentive/disincentive threshold values (Nemmers, et. Al, 2006).
State Testing Interval Incentive Pa /ment Range Full Pa\ Range Penalty Assesment Range Correction Work Req.
mi km in/mi mm/km in/mi mm/km in/mi mm/km in/mi mm/km
CT 0.1 0.16 <60 <950 60-80 950-1260 80.1-120 1261-1894 >120 >1894
GA 1 1.6 - - <47.5 <750 - - >47.5 >750
LA - - - - - - - - - -
ME 0.12 0.2 <60 <945 60.1-70 946-1105 70.1-80 1106-1260 >80 >1260
MA 0.12 0.2 - - <95 <1500 - - - -
SD 0.1 0.16 <55 <868 55.1-70 869-1105 70.1-80 1106-1262 >80 >1262
VT 0.2 0.32 <60 <950 60-69 950-1090 70-95 1091-1500 >95 >1500
VA 0.1 0.16 <55 <869 55.1-70 896-1105 70.1-100 1106-1578 >100 >1578
WA 0.1 0.16 <60 <946 60.1-95 947-1500 95.1-115 1501-1815 >115 >1815
WY 0.1 0.16 - -- 55-70 868-1105 - -- -- -


3.4 Alternative Profilometric Indices
While the IRI is widely regarded as the most universal pavement
roughness index, several other methods for longitudinal pavement analysis
exist. The IRI tool bears many similarities to these measures. This section
will review some of the existing measures and discuss the advantages and
disadvantages of their use. The University of Colorado Denver (UCD) and
CCD project ultimately used the IRI metric in order to maintain a level of
international universality such that the results may be used and expanded
upon easily in future studies. Table 3.2 shows a breakdown of the United
States and the profiling method or methods used by their respective DOTs
15


Table 3.2 Preferred roughness indices by state (FHWA, 2006)
State Profiling Index
AL PI
AK
AZ MRN
AR PI
CA PI
CO PI
CT IRI
DE
FL
GA IRI
HI
ID PI
IL PI
IN PI
IA PI
KS PI
KY PI
LA IRI
ME IRI
MD PI
MA IRI
Ml PI/RQI
MN PI
MS PI
MO PI
State Profiling Index
MT
NE PI
NV PI
NH RN
NJ
NM PI
NC CSI
ND
OH PI
OK PI
OR PI
PA PI
PR PI
Rl
SC MRN
SD IRI
TN MRN
TX PI
UT PI
VT IRI
VA IRI
WA IRI
WV MRN
Wl PI
WY IRI
PI: Profile Index
MRN: Mean Roughness Index
IRI: International Roughness Index
RQI: Ride Quality Index
CSI: Cumulative Straightedge Index
RN: Ride Number
16


3.4.1 The Half-Car Ride Index
The half-car ride index (HRI) is a measure closely related to the IRI.
The two are often referred to synonymously. The State of Colorado
provides for the use of this measure as an alternative to the profile index.
The reference, half-car in the title refers to the fact that two points of data
collection are used and averaged to yield a final roughness output. Each
laser track on the profiler used in this study represents half of a vehicle.
The HRI is reported in in/mi (mm/km) just like the IRI. The IRI, however,
may be derived from either a single point of contact or a dual point of
contact system and is not dependent on an average of the two tracks.
While the IRI is not dependent upon the multiple track data collection, using
a method which incorporated both wheel tracks may provide a more
representative analysis of the roadway.
3.4.2 The Michigan Ride Quality Index
The Michigan Ride Quality Index (RQI) produces an output in in/mi
(mm/km) and is intended to quantify user perception of road roughness
(Smith, et. al, 2002). This index provides a similar output to the standard
IRI with differences laying in the direct equations used to derive the final
output. Its lack of common use outside of the United States limits its
17


usefulness in comparing results between studies without the use of a
conversion factor (Smith et. al, 2002). International indices are important
when comparing roads between countries. Uniform data collection
methodologies are also aid researchers in advancing the science of
roughness analysis.
3.4.3 The CalPro Simulation Model
The CalPro Simulation Model has been in use since the 1940s (Song,
et. al, 2008). It was developed for airfield and highway pavements and was
originally limited by its operation speed as the hand-held device was
designed to be pushed while walking. This made data collection extremely
time consuming. Adjustments have since been made to the data collection
procedure and the CalPro Simulation Model results may be compiled by
high speed profilers today. The results are presented in in/mi (mm/km) and
are highly correlated to the PI and IRI. As such it is still widely used by the
Federal Aviation Administration (FAA) and by some state DOTs as a viable
pavement analysis tool (Smith, et. al, 2002).
3.4.4 The Straightedge and Rolling Straightedge Measures
In early pavement roughness analysis, a straight edge was used by
pavement inspectors to measure slopes and to generate a direct
18


longitudinal profile of the roadway (Ksaibati & Mahmood, 2002). This
method was time consuming and required a great deal of precision to
ensure the accuracy of the data. The rolling straight edge (RSE) devices
are wheel-mounted 8-16 ft (2.44-4.88 m) pillars which measure bumps and
dips as variations above and below a zero line (Ksaibati & Mahmood
,2002). This method was among the first measures of pavement roughness
used. The Laser Inertial Surface Analyzer (LISA) 6500 pavement profiler
used in the UCD-CCD study is able to produce a tabular output for the RSE
summary by applying the same measures used in gathering data for the
IRI. This output shows height measures for each individual deviation from
the mean referred to as bumps and dips. A graph shows station-based
location for the measured segment and specification limit lines above and
below the established zero line. The specifications may be set by the
measuring agency. This rolling straightedge test is useful in establishing
location data for individual stresses but does not provide a clear summary
of the overall condition of the roadway and was therefore ruled out for use
in this study.
3.4.5 The Ride Number
The ride number (RN) is a non-linearly proportional scaling system
based on the historical measure of present serviceability index (PSI)
19


(Perera, et. al, 2005). The RN is a 0-5 scale index where 0 is perfectly
smooth and 5 is the highest scale of roughness. The RN differs from the
PSI in the direction of the scaling; in the PSI, a rating of 5 is perfectly
smooth and a rating of 0 is completely rough. The RN was developed to
provide an easily understood rating system for comparisons between
roadways that were measured using the same data collection method. The
RN may be calculated directly from the PI using Equation 3.2 (Awasthi, et.
al, 2003):
RN=5e'160xPI
(3.2)
Where:
RN = Ride Number
PI = Performance Index
This scale is very useful in communicating roughness values to the general
public or a lay audience. It is less useful than the IRI in providing a detailed
analytical view of a roadway segment. The AMES Engineering software
used in the UCD-CCD study reported the RN in conjunction with the IRI
values in each summary report to three significant digits.
20


3.4.6 The Performance Index
The performance index (PI) is a widely use pavement profiling index
which is derived from a specialized rolling straightedge called the California
Profilograph. The California Profilograph device measures roughness
based on a 25 ft (7.62 m) reference plane (Nemmers, et. al, 2006). The
measurements are taken by a wheel mounted at the center of the device
which is free to move vertically. The PI differs from the IRI in that it
represents a direct measure of the pavement surface while the IRI
measures the vertical movement of the vehicle based on a quarter-car
simulation model (Nemmers, et. al, 2006).
3.5 Limitations of the IRI
One of the largest complaints about the IRI and all other pavement
roughness indices on highways is the lack of attention paid to rutting
(Reggin, et. al, 2008). Pavement roughness is gauged using a longitudinal
path, running along the length of the roadway. Rutting may impact a
roughness rating if it intersects the laser path for some portion of the data
collection process. Rutting generally occurs in wheel-paths which are also
longitudinal to the roadway, and therefore the impact on roughness data is
minimal. Methods to assess rutting severity have been generated for use
on traditional highway segments, but there is not currently a single
21


accepted method of analysis. In urban roadway analysis, a longitudinal path
on the main-street segment shows a transverse analysis of any intersecting
cross-streets (Reggin, et. al, 2008).
22


4. Current Profilometric Practice for the Colorado Department of
Transportation
CDOT, like many other states, has a very particular set of standards
by which it measures pavement roughness. CDOT uses the PI, a direct
profile evaluation metric (FHWA, 2006). The Colorado Procedure 74-07a
outlines profiler operations and evaluation practices for contractors working
for the state (CDOT, CP 74). This procedural standard identifies the
FHWAs ProVAL software as the standard for compiling pavement
roughness reports. ProVAL Version 02.60.0009 is free software available
for downloading from the FHWA website. Provisions are also listed for all
of the minutia of pavement profiling standards including calibration tests,
operating speeds, segment lengths, testing procedures, and control
operations. These standards are established for high speed profilers and
geared for use on the Colorado highway system (CP 74).
23


4.1 CDOT Pavement Incentive/Disincentive Provision for Hot Mix
Asphalt Facilities
Specialized standards were developed for use on hot mix asphalt
surfaces in Colorado (CDOT, 2007). These provisions account for
profilometric data collection on various surface treatments and are reported
by the HRI metric in in/mi (mm/meter). CDOT provides contractors with an
incentive/disincentive provision based on the Half-car ride index which is
allocated according to threshold values (CDOT, 2007). Table 4.1 shows
the delineation of the incentive/disincentive standards used.
Table 4.1 CDOT incentive/disincentive payment schedule based on HRI
values (CDOT, 2007)
Pavement Resurfacing Type Incentive Payment 'i' ($/sqyd) No Incentive/ Disincentive Disincentive Payment 'i' ($/sqyd) Corrective Work Required
Reconstructed Surfaces HRI < 50.0 in/mi I = $0.32 63.0 in/mi £ HRI £ 72.0 in/mi I = $0.00 72.0 in/mi £ HRI £ 85.0 in/mi I = (2304 -32 x HRI)/1300 HRI > 90
50.0 in/mi S HRI £ 63.0 in/mi I = (2016-32 x HRI) /1300 HRI > 85.0 I = -$0.32
Overlaid Surfaces HRI <45.0 I = $0.32 58.0 in/mi £ HRI £ 67.0 in/mi I = $0.00 67.0 in/mi £ HRI £ 80.0 in/mi I = (2144 32 x HRI) / 1300 HRI > 85
45.0 in/mi £ HRI £ 58.0 in/mi l = (2016 32 X HRI)/1300 HRI > 80.0 I = -$0.32
A working example may be presented by drawing from the specifications
listed in Table 4.1. A contractor is responsible for a project to resurface a
roadway classified as a reconstructed surface. If this surface measures 20
yd (18.29 m) in width by 1760 yards (1609.34 m) in length. The total
24


surface area would be 35200 sq yd (29,434.83 sq m). If the sample area
produces an HRI of 46 in/mi (726 mm/km), the contractor would be entitled
to a bonus of $0.32 per sq yd or $11264.00. This value represents the
contractor bonus for one mile of roadway for a facility manufactured for the
State of Colorado.
For the same one mile segment, if the HRI produced is 80 in/mi
(1262.63 mm/km), the contractor would be charged a disincentive fee. The
fee is calculated using Equation 4.1.
I = (2304 -32 x HRI)/ 1300 (4.1)
Where:
I = Incentive ($)
HRI = Half-Car Ride Index (in/mi)
In this example, the disincentive charge would be $0.20 per sq yd, or
$7,040.00 for the one mile stretch.
CDOT also allows for the use of a percent improvement scale for
financial incentive/disincentive to be included in contractor agreements
(CDOT, 2007). Percent improvement is calculated using a simple percent
difference formula. The percent improvement incentive program is
25


delineated by rural and urban facilities. Urban facilities are identified as
those highways having a high number of intersections or utility boxes
(CDOT, 2007). Table 4.2 outlines the pay scale used for each facility.
Table 4.2 CDOT incentive/ disincentive payment schedule based on
percent improvement (CDOT, 2007)
Pavement Resurfacing Type Incentive Payment ($/sqyd) No Incentive/ Disincentive Disincentive Payment 'i' ($/sqyd) Corrective Work Required
Rural Highways %l > 60.0 1 = $0.32 40.0% < %l < 45.0% I = $0.00 25.0% 5 %l £ 40.0% I = (32 x %l 1280)/1500 %l < 20.0%
45.0 < %l < 60.0 1 = (32 x %l 1440) / 1500 %l < 25.0% I = -$0.32
Urban Highways %l > 50.0 I = $0.32 -5.0% < %l < 5.0% I = $0.00 -5.0% < %l < 5.0% I = (32 x %l + 160)/ 1500 %l < -25.0%
5.0 < %l < 50.0 I = (32 x %l 160) / 4500 %l < -20.0% I = -$0.32
Referring, again, to a situational example; if a contractor produces a
rural highway that shows a percent improvement of less than 20% the state
will require that the contractor redo the facility until results may be
presented at a higher percent improvement level. If a facility shows an
improvement of 30%, the contractor will not be required to conduct
corrective work on the facility, but they will be subjected to a disincentive
charge as calculated in Equation 4.2:
26


I = (32 x %l 1280) / 1500
(4.2)
Where:
I = Incentive ($)
%l = Percent Improvement (%)
The acknowledgement of the difference in rural and urban facilities in this
disincentive program speaks to the question addressed in this project.
Forming a contiguous measure for urban pavement roughness will allow not
only state agencies, but also city and county municipalities to take
advantage of the IRI metric in determining new pavement quality.
27


5. Equipment Used in Assessing the IRI of CCD Roadways
The UCD-CCD research team used a data collection device produced
by Ames Engineering called the Model 6200 LISA. This dual laser system
is deployed using a Panasonic Toughbook laptop computer and is mounted
to a John Deere brand Gator Utility Vehicle as shown in Figures 5.1 and
5.2.
Figure 5.1 Pavement Profiler Unit
28


Collectively the devices are referred to as the Pavement Profiler. The
proprietary Profiling System Software Version 5.3.6 provided real time
roadway review and a variety of analysis and output options. These
analysis options include the California Profilograph, the HRI, the RN
statistic, the RQI statistic, and the IRI. The UCD-CCD study used the IRI
statistic for contiguity and comparison purposes. Both the Model 6200 LISA
29


and the Panasonic Toughbook were powered by the battery in the John
Deere Gator.
Data collection was completed using two escort vehicles positioned
before and after the profiler to ensure the safety of the driver and passenger
in the John Deere Gator. Figure 5.3 shows the following safety vehicle
during a data collection run.
Figure 5.3 Model 6500 LISA Profiler with Following Safety Vehicle
Inspectors included CCD street maintenance employees and UCD civil
engineering students and faculty. Roadways were analyzed using the
30


pavement profiler at least two times in each direction of travel and in each
lane of interest. Multiple drivers and Panasonic Toughbook operators were
used in order to gauge repeatability between researchers.
5.1 Profiler Calibration and Performance Standards
The American Association of State Highway and Transportation
Officials (AASHTO) have established standards for the estimation of
roadway roughness. The Pavement Profiler used in this study meets the
American Society for Testing and Materials (ASTM) requirements for a
class I road profile measuring device as outlined in the ASTM E950-98
(American Society for Testing and Materials, 2004). The term roughness in
the IRI refers specifically to the average deviation measured in vehicle
motion from a planar surface. This deviation is presented in inches per mile
(mm/km) and is compiled by each 0.1 mile (0.16 km) to produce the final IRI
statistic as given by Eq. 3.1.
5.1.1 Calibration Options Offered by the Ames Engineering Software
The Ames Engineering software program, Profiler Software Version
5.3.6, offers several calibration settings to fine tune roadway analysis
shown in Figure 5.3. Each of these settings was initially set with a
nationally accepted default value. All data runs in the study were collected
31


using the default settings shown in Figure 5.4. Output may be listed in
either English or metric units for all test selections.


Report Options Analysis Setup
[-Reduction Settings-------------
Reduction Length (528
Short Segment Length |250
F" Static Reduction Segments
Bump Detection Settings-
Bump Width
25
Bump Height jo.3
17 B ump D election E nabled
|7 Dip Detection Enabled
Filter Settings-
High Pass Filter Cutoff jO
Low Pass Filter Cutoff ]2
f~ Moving Average Filter Enabled
17 Debris Filter Enabled
feet
feet
feet
inches
feet
feet
Profiler Setup
Rolling Straightedge Simulation-
feet
Straightedge Length |10
Specification Limit j0.25
inches
CalPro Simulation-
Blanking Band Width (0
Minimum Scallop Height jo.03
Minimum Scallop Width |2
Scallop Rounding JO.CH
17 Count Scallops Once
inches
inches
feet
inches
Units--------
<* English
r Metric
CI Documents and Sei!ings\AS Users \Applicalbrt
Data \AmesProS afup. xml
Save
Cancel
Figure 5.4 Analysis Settings for the Profiling Software Version 5.3.6
5.1.1.1 Reduction Settings
The reduction length refers to the length used by the analysis software
to break the profile into segments through which it calculates the profile
32


metrics. The software standard in English units is 528 ft (160.93 m) and in
metric units is 100 m (328.1 ft). Figure 5.5 shows an example of the 528 ft
(160.93 m) analysis sections in the segment summary output.
<- IRI/RN Summary Track 1 ->
From To Disc lRl(in/mi) RN
1+00 6+28 11+56 16+84 6+28 11+56 16+S4 17+96 528 528 528 112 213.42 211.53 171.59 437.86 1.90 2.51 2.76 1.12
Total 1696 214.63 2. 31
<- From IRI/RN To summary Track 2 Dist IRI(in/mi) > RN
1+00 6+28 11+56 16+84 6+28 11+56 16+54 17+96 528 528 528 112 170.57 203.99 204.83 580.51 2. 30 2.63 2.15 0. 59
Total 1696 218.?1 2.25
<- From IRI/RN To summary Average Dist IRI(in/mi) > RN
1+00 6+28 11+56 16+84 6+28 11+56 16+54 17+96 528 528 528 112 192.CO 207.76 188.21 509.18 2.10 2. 57 2.46 0.86
Total 1696 216. 67 2.28

<- Type Burap/Dip From Locations Peak Track 1 -> To Height(in)
Dip 1+17 1+18 1-19 0. 29
Dip 2+97 2+97 2+98 0. 07
Sump 2+98 2+99 3+00 0.07
Sump 3+83 3+S3 3+84 0.04
Dip 3+97 3+98 4-01 0. 21
Sump 4+13 4+23 4+30 0.43
Dip 4+87 4+88 4+89 0. 06
Dip 5+41 5+41 5-41 0.03
Bump 7+74 7+79 7+89 0.23
Dip S+10 8+18 8+24 0. 27
Sump 8+91 8+92 8+92 0. 05
Sump 9+S8 9+90 10+01 0.16
Dip 10+01 10+06 10+15 0. 59
Sump 10+21 10+29 10-31 0. 08
Dip 15+03 15+13 15-23 0. 57
sump 15+96 16+02 16+06 0.18
Di p 17+38 17+41 17+44 0. 09
sump 17+59 17+61 17+75 0. 72
<- Burap/Dip Locations Track 2 ->
Type From Peak to Height(in)
Sump 4+16 4+24 4-30 0.14
Dip 4+75 4+75 4-75 0.01
Dip 6+62 6+66 6-70 0. 09
Sump 6+77 6+81 6-87 0.16
Di p 7+24 7+29 7-35 0. 20
Dip S+23 8+25 8-33 0.08
Bump S+42 8+42 8-42 0. 02
Sump 9+05 9+06 9-07 0.04
Dip 9+44 9+51 9-53 0.13
sump 9+89 9+90 9+98 0.10
Dip 10+10 10+10 10-10 0.03
sump 10+27 10+33 10-43 0. 39
n-t n 11 11 11.73 nil
Figure 5.5 Sample IRI Output Segments From Profiler System Software
Version 5.3.6
The short segment length is a second variable in this category. It only
influences the CalPro Simulation Model output, and the default values are
not pertinent to this study.
33


5.1.1.2 Bump Detection Settings
The bump detection option applies to the CalPro Simulation Model and
the direct, or true, profile. The variables are bump width and bump height
and allow the software output to shade or tag individual bump locations.
This direct profile is not used in assessing the final IRI statistic. The default
setting for bump width is 25 ft (7.62 m), and bump height is 0.3 in (7.62
mm).
5.1.1.3 Filter Settings
There are two filter settings offered in the Profiler System Software
Version 5.3.6: the high pass filter cutoff and the low pass filter cutoff. The
high pass filter serves to temper sharp horizontal and vertical curves that
impact the placement of the planar surface in the final profile. The default
value for the high pass filter cutoff is 0 ft (0 m) indicating that it is not active
at all. The low pass filter applies only to the CalPro Simulation Model and
the straightedge results. It is intended to quiet unwanted noise in the
profile, and produce a smother data run. As it does not apply to the IRI
results, the standard of 2 ft (0.61 m) does not impact this study.
34


5.1.1.4 Rolling Straightedge Simulation Settings
This setting option allows the assignment of variables for consideration
in the rolling straight edge test; straightedge length and a specification limit.
As the dual-track laser system does not emulate the shape of the standard
rolling straight edge equipment, it is able to yield highly accurate
approximations given these variable assignments. The default settings are
10 ft (3.05 m) for the length and 0.25 in (6.35 mm) for the specification limit.
5.1.1.5 CalPro Simulation Settings
The settings pertinent to the CalPro Simulation Model output include
blanking band width, minimum scallop height, minimum scallop width, and
scallop rounding. The default settings are 0 in (0 mm), 0.03 in (0.76 mm), 2
ft (0.61 m), and 0.01 in (0.254 mm) respectively. While the CalPro
Simulation Model ultimately provides a longitudinal profile output similar to
that produced by the IRI, the derivation and data collection process relies
on a dissimilar analytical system.
5.2 International Roughness Index Data Output
The Profiler System Software Version 5.3.6 was set to produce an
output based on the IRI for use in this study. On site, the profiler produces
a paper copy with summary information about the segment and direct
35


profile listings of bumps and dips from the segment analyzed. In addition to
a tabular output, the Profiler Software Version 5.3.6 is able to produce a
graphical output. This graphical output shows the longitudinal profile of the
roadway as collected from the laser track units mounted at the center of the
profiler and lined up with each wheel path. The data is presented with
distance plotted along the x-axis and the planar surface and bump/dip
heights recorded on the y-axis. Figure 5.6 shows a sample of the
softwares graphical output. Figure 5.7 shows a standard view of the laser
tracks in separate screens with an overlaid speed data profile.
: Ames Data Viewer C:\Jobs\I25\8th_avenuel.attf -1C
Fie Tools Windows

JoJxJ
I ops 1 odsTI

ij
[ QOS 11RI: 209.23 (in/mi) | OPS 2IRI: 222,39 flnftnQ | Average Speed : 10.77 mph 0000+00.000 to 0022+55.358 [22S5.36ft.]
Figure 5.6 Graphical Output of a Direct Profile Measure
36


Figure 5.7 Direct Profile Output Separated by Laser Track With Overlaid
Speed Profile
37


6. Data Collection Methods
Careful data collection procedures were followed throughout the study
to ensure the recording of relevant and repeatable data runs. The data
collection practices were agreed upon and understood by all contributing
parties before the start of the study. If a data run was completed that did
not meet the specified standards for the study it was repeated.
6.1 General Methodology
The procedure followed to complete data collection on a segment of
urban roadway involved using two CCD vehicles equipped with emergency
vehicle lights and the Pavement Profiler. For safety, the team would
position the vehicles with emergency lighting in leading and following
positions to the Pavement Profiler as shown in Figure 5.3.
Drivers traveled at approximately 10 miles per hour (16.01 km per
hour) along the segments being analyzed. Data was collected using the
Model 6200 LISA profile with both laser tracks active.
The data collection teams were made up of four people: three drivers,
and one computer operator. Roadway segments ranged from 0.29 miles
(0.47 km) to 1.47 miles (2.37 km) in length. Operators started and stopped
38


the data run manually using the Panasonic Toughbook computer which is
equipped with a touch screen and easily identified start and stop buttons.
The computer is mounted in the profiler on the passenger side and is
connected directly to a printer which produced hard-copies of each data run
output.
The operator would start a data collection run when the laser tracks,
mounted at the longitudinal center of the profiler, intersected the cross-
street curb line at the beginning of the roadway segment. Likewise, the
operator would terminate the data collection run when the laser tracks
intersected the cross-street curb line at the end of the roadway segment.
Data collection was completed on dry surfaces only and typically 3 to
5 segments were completed per outing. For safety, higher traffic volume
roadway data collection was completed during off-peak hours and
occasionally during evening and nighttime hours.
6.1.1 Procedural Application of the LISA Model 6200
When preparing to collect data on a segment of CCD rural roadway
the data collection team followed strict procedural protocol. On site, the
data collection team would power up the profiler unit including the
Panasonic Toughbook computer, shown in Figure 5.1, and the LISA Model
6200 Printer, shown in Figure 6.1.
39


Figure 6.1 LISA Model 6200 Printer with Sample Output Scroll.
The proprietary interface software home screen, shown in Figure 6.2,
allows the user to begin an analysis run or conduct an analytical review of
previous data runs.
40


n.
/J Engineering*
Profile 1------
Height -.-in.
| Status:
Proffer System Software
Version $.3.6
.....sseeWLL^.m.....
Profile 2-
Height-.-in.
Status:
, GPS-------------------------------------
Latitude:
Speed:
System...................................
Cycle Court 4
Battery Voltage: V
System Status: Connection Error (Switch Pod)

Diagnostics
Ha-
Switch Pod
Data: No Link
Status:
Longitude:
Link Quality: No Connection
fa Time/Date
5:09:55 PM
Sunday, March 15,2009
F1 Setup
F2 Calibrate
F3-Start
F5 Analyze
Jal*!
F6 View
Start
Analyze
Calibrate
Setup
View
Quit
F10-Quit
Figure 6.2 Profiler System Software Version 5.3.6 Home Screen
The Start feature on the home screen allows the user to begin a new data
collection run when the integrated system is active. Each run was assigned
a file name that specified the location, boundaries, direction of travel, lane,
and order of collection. These files were maintained both on the Panasonic
Toughbook computer and on a backup drive.
After a data collection run, the printer would produce a hard copy of
the summary results. Sample segments of these hard-copy results are
shown in Figure 5.5. The Analyze feature on the software home screen
41


was used after the data collection runs to review data, and apply filters as
needed. The raw data of each run was saved for future review using
multiple analysis types. Figure 6.3 shows the file selection screen
prompted by the Analyze feature.
| FUePsih i Col Start : Collect ionDir Refer enceUnls ; Ana Start i Ana End Ana Dir \ Texas f|
C ;C.Uob;\i25MSthavewbadf jlOO Positive Feet 1100 ]1647 Forward j Stliavew
pOn* Selected Fite -
' r-z-ami <
' ^
Clear
Clear All
Analyze Files
Select Files
Save Project
Load Project
Setup
Exit
F1 Select
F2 Save
F3 Load
F4 Setup
F5 Quit
Figure 6.3 Project File Selection Screen in the Profiler System Software
Version 5.3.6.
The Setup feature, shown on the home screen in Figure 6.2, allowed the
user access to the features discussed in Section 5.1. The setup screen is
shown in Figure 5.4.
42


6.2 Data Repeatability Methodology
To ensure the quality of the data collected, as well as the
repeatability of this study, the research team exercised strict methodology
when completing a data collection run. For each roadway segment a
minimum of two drivers per direction would operate the Pavement Profiler.
The profiler was equipped with a mechanical positioning arm
mounted on the front of the vehicle. This positioning arm was oriented
perpendicular to the direction of travel and extended toward the center of
the road allowing the profiler driver to align the vehicle with the street center
line, or striping line associated with the edge of the lane of travel as shown
in Figure 5.3. If no striping was available researchers discussed other
points of reverence, such as mid-street seams, in order to maintain
similarity in transverse roadway position between data collection runs.
An initial repeatability trial was completed on two roadway segments
using five separate drivers in each direction of travel. The results of each
data run from each driver were then compared individually and as a group
to find the percent difference within a data collection run.
6.3 Before and After Comparison Methodology
Several of the roadway segments analyzed in this research were
chosen because they were scheduled for repaving maintenance. Ninety-
43


five total collection sites were analyzed in the before-repair state. Of those
ninety-five, repaving has been completed at thirty-three locations. The
thirty-three locations with before repair and after repair data span 27.15
lane-mi (43.69 lane-km). The before and after repair data was reviewed
both individually by location and collectively by weighted average. The
sites spanned multiple development characteristics from residential areas
with low heavy vehicle presence, to commercial areas with a high heavy
vehicle presence. The infrastructure factors influencing IRI values for each
collection site varied widely as well. Some sites had no drainage features
in the wheel path at all, but a high number of cross street access points.
Average values based on segment length are reported as well as a
thorough examination of urban factors influence on the final IRI values.
44


7. General Results for Urban International Roughness Index Review
of City and County of Denver Streets
This study analyzed roughness data collected on 66.9 lane-mi (107.67
lane-km) of urban roadway in the CCD. The average recorded IRI value for
roadway segments, weighted by length, collected before repaving was
278.81 in/mi (4400.4 mm/km). There were 79 before repaving condition
segments considered totaling 55.4 lane-mi (89.16 lane-km) in length. The
average recorded IRI value for roadway segments, weighted by length,
collected after repaving was 175.65 in/mi (2772.25 mm/km). In addition to
the 33 after repaving condition segments included in the before and after
study, 22 additional segments were analyzed after repaving with no
associated before repaving data. There were a total of 55 after repaving
condition segments totaling 39.33 lane-mi (63.30 lane-km). Appendix A
shows the data table for all segments analyzed.
7.1 Results of the Repeatability Analysis for Segment Data Collection
Two segments were selected for use in the repeatability analysis
study. On each of these segments, five profiler operators collected data
consecutively and on the same day. Table 7.1 shows the data table for
45


both segments. In the first segment analyzed, the lowest average IRI value
collected by an operator was 177.51 in/mi (2801.6 mm/km). The highest
value for this first review segment was 186.54 in/mi (2944.12 mm/km). The
percent difference of this segment, between operators, was 2.1%. The
lowest IRI value collected from the second segment considered was 221.87
in/mi (3501.72 mm/km). The highest IRI value recorded on this segment
was 244.01 in/mi (3851.16 mm/km). The percent difference for this
segment, between operators, was 1.1%.
46


Table 7.1 Repeatability trial data results.
Segment Length Track Driver 1 Driver 2 Driver 3 Driver 4 Driver 5 Average IRI
Street From / To (mi) (km) Track (in/mi) (mm/km) (in/mi) (mm/km) (in/mi) (mm/km) (in/mi) (mm/km) (in/mi) (mm/km) (in/mi) (mm/km) Difference
E. 8th Ave. Steele St. Harrison St. 0.40 0.64 ODS1 232.50 3669.53 231.67 3656.43 227.62 3592.51 219.06 3457.41 238.81 3769.12 237.42 3747.182 l.l%
ODS2 247.00 3898.38 236.48 3732.35 242.11 3821.20 260.11 4105.30 238.81 3769.12
AVG 239.75 3783.96 234.08 3694.39 234.87 3706.86 239.59 3781.35 238.81 3769.12
ODS1 154.33 2435.78 162.73 2568.36 163.77 2584.77 162.34 2562.20 159.38 2515.48
E. 8th Ave. Downing St. York St. 0.68 1.09 ODS2 200.69 3167.48 210.35 3319.94 206.46 3258.54 197.28 3113.66 208.42 3289.48 182.57 2881.489 2.1%
AVG 177.51 2801,63 186.54 2944.15 185.12 2921.66 179.81 2837.93 183.90 2902.48


7.2 Results of the Before Repaving and After Repaving Segment
Analysis
Comparisons were made on 33 test trials for which before repaving
and after repaving data was collected. The before and after study
segments reviewed totaled 27.15 lane-mi (43.69 lane-km). The weighted
average of the IRI values for the before repaving trials, within the before
and after study only, was 375.28 in/mi (5922.97 mm/km). The weighted
average of the IRI values for the after repaving trials, within the before and
after study only, was 170.53 in/mi (2691.44 mm/km). The average percent
improvement, weighted by segment length, of all of the before and after trial
segments was 36.3%. Appendix B shows a full listing of the before and
after repaving segment results. Figures 7.1 and 7.2 show a graphical
comparison of the before and after data by location.
48


IRI Values (in/mi)
Figure 7.1 Graphical Comparison of Before and After Data Locations 1-17


IRI Value (in/mi)
CD
400.00
350 00
300.00-
250 00-
200.00-
150 00-
100.00-
50.00
0.00-*


f y y y y y y y y y
* f rsssss'jfjrj
y j* y y
y y .y y y
y y
*//////
Figure 7.2 Graphical Comparison of Before and After Data Locations 18-33


7.3 Repaving Methods Used on Test Sites
There are three methods employed for resurfacing the roadways
examined in this study. The first is a mill and overlay (M&O) resurfacing
technique. This is a process where the outer portions of both sides of the
street are subjected to a 1.5 in (38.1 mm) mill in order to align the gutter
flow lines. A bonding agent is applied to the existing pavement surface
followed by an overlay of 1.5-2 in (38.1-50.9 mm) of asphalt. These are
applied to the entire road surface in order to create a smooth crown
structure.
The hot in-place recycling (HIPR) system is a resurfacing process that
makes use of the existing pavement on a roadway. The road surface is
milled to a depth of 1-3 in (25.4-76.2 mm). The milled product is heated
and mixed with a rejuvenating agent onsite. The heated material is then
reapplied to the road surface. This repaving method minimizes the new
materials cost in resurfacing roadways and corrects surface distress on
asphalt streets.
A complete reconstruction is the final alternative used on the test sites
for resurfacing roadways. A reconstruction effort involves the complete
removal of the old surface material and the laying of all new asphalt. This
may be desirable if corrections must be made to the road foundation.
51


7.3.1 IRI Results by Resurfacing Procedure Type
The weighted average IRI for those roadways resurfaced using the
M&O process was 188.84 in/mi (2980.58 mm/km). The weighted average
percent improvement for segments with both before resurfacing and after
resurfacing data, and resurfaced using the M&O process was 32.2%. The
weighted average IRI for roadways resurfaced using the HI PR method was
128.47 in/mi (2027.62 mm/km). The average percent improvement for
segments, with both before resurfacing and after resurfacing data,
resurfaced using the HIPR method was 46.6%. The weighted average IRI
for roadways that underwent complete reconstruction was 202.16 in/mi
(3190.65) mm/km). The average percent improvement of reconstructed
facilities with both before and after data available was 28%. Table 7.2
summarizes the results of the after resurfacing comparison study. Table
7.3 shows the results of the comparison made on segments with both
before resurfacing and after resurfacing data available.
52


Table 7.2 Weighted average IRI values delineated by resurfacing method
Resurfacing Method Total Length Tested (mi) / (km) Number of Segments Tested Weighted Average IRI (in/mi) / (mm/km)
Mill & Overlay 27.61 44.43 39 188.85 2980.58
HIPR 9.16 14.74 8 128.47 2027.62
Reconstruction 2.56 4.12 8 202.16 3190.65
53


Table 7.3 Weighted average IRI values for segments with both before resurfacing and after resurfacing
data, delineated by resurfacing method.
Resurfacing Method After Repaving Condition Total Length Tested (mi) / (km) Number of Segments Tested Before Repav Weighted (in/mi) / ing Condition Werage IRI mm/km) After Repavi Weighted / (in/mi) / ng Condition Werage IRI mm/km) Percent Improvement
Mill & Overlay 16.71 26.89 21 280.27 4423.45 190.06 2999.68 32.2%
HIPR 9.16 14.74 8 240.55 3796.55 128.48 2027.77 46.6%
Reconstruction 1.28 2.06 4 300.65 4745.10 216.61 3418.71 28.0%


8. Discussion
The IRI results for segments reviewed in this study have provided a
body of work from which assertions can be made about the threshold limits
and expectations for roadways in the CCD. In the context of this study the
target sites represent a good cross section of street types, neighborhoods,
and development characters within the CCD. The CCD have expressed
interest in basing future pavement roughness expectations on the results of
this study. As such, comparisons were made to the national standards for
freeways and rural highways in an attempt to create a new set of threshold
values for the application of an urban IRI.
8.1 Comparison of Acceptable IRI Values for Freeways and Rural
Highways and Acceptable CCD Urban IRI Values
The Federal Highway Administration (FHWA) classifies IRI values of
less than 95 in/mile (1506.9 mm/km) as good, and IRI values of less than
170 in/mile (2696.5 mm/km) as acceptable for highway installations (FHWA,
2006).
55


Table 8.1 shows the percentage of the 33 segments mileage that
would be classified as good, acceptable, and not acceptable for both before
repaving and after repaving conditions using current FHWA metrics.
Table 8.1 FHWA scale delineation of mileage for before repair and after
repair conditions _____________________________________________
Pre-Repair Post-Repair
Good 0 0
Acceptable 0 55.14%
Not Acceptable 100% 44.86%
Prior to repairs, all 33 segments had IRI values greater than 200 and,
therefore, did not meet the current IRI threshold level of 150 for an
acceptable rating prior to repairs. After repaving, all 33 segments had IRI
values greater than 95 and therefore did not meet the criteria to achieve the
rating of good. Only 55.14% of the repaired segments met the metric for
an acceptable ride. Forty-four percent of the repaved roads, therefore, did
not meet the FHWAs acceptable ride criteria and 100% did not meet
metrics for good criteria. The unacceptable segments were further
inspected and subjectively analyzed for ride. All of the unacceptable
segments exhibited a ride quality that was normal. Therefore, new
reasonable urban IRI metrics needed to be developed.
56


The IRI values for CCD roadways after repaving spanned a
considerable range. High IRI values of 315 in/mi (4975 mm/km) were
observed for W. 13th Ave between Kalamath St. and the Light Rail tracks in
the eastbound direction, and 377 in/mi (5950 mm/km) for the same
segment in the westbound direction. Subjectively the roadway segment
appears to be well constructed and observers have noted a smooth ride.
There is a significant elevation change at this site, paired with a rolling
terrain and several cross street intersections. Each of these conditions
influences the final IRI values at the site.
Other roadways within the CCD show a relatively good IRI rating
relative to urban environments. The test site on E. Yale Ave. between S.
Colorado Blvd. and Interstate 25 showed an IRI value of 99.71 in/mi
(1573.7 mm/km) in the Eastbound direction, and 99.51 in/mi (1570.55
mm/km) in the Westbound direction.
8.1.1 Recommendation for Urban International Roughness Index
Threshold Values
Data gathered in this study suggests that reasonable UIRI values may
be reclassified as follows: UIRI values less than 150 in/mile (2367.44
mm/km) are good, and UIRI values less than 220 in/mi (3472.24 mm/km)
are acceptable. Using this criteria, 8.8% of the before repair mileage was
57


rated as acceptable while 91.2% was unacceptable as shown in Table 8.2.
The after repaving ratings were showed 31.6% rated as good and 55.4% as
acceptable. 6.7% of the after repair mileage did not meet the acceptable
rating using revised metric criteria.
Table 8.2 UIRI scale delineation of mileage for before repair and after
repair conditions
Pre-Repair Post-Repair
Good 0 31.64%
Acceptable 8.84% 61.62%
Not Acceptable 91.16% 6.74%
8.1.2 Analysis of UIRI Percent Improvement Values
An examination of percent improvement has also been a good way to
gauge the effectiveness of a repaving project as measured by roughness.
Some state agencies, including the Colorado DOT, used percent
improvement expectations on freeways and rural highways as a viable
measure in providing contract incentives.
Although some segments still rated poorly relative to the rest of the
data segments after repairs, the IRI values did improve by 10 to 15 percent
in these areas. Revisiting the example location at W. 13th Ave between
Kalamath St. and the Light Rail tracks, the high values for both the before
and after repaving data collection runs still yielded a 14% improvement in
58


the eastbound direction and an 11% improvement in the westbound
direction after repaving was completed. These percent improvements by
direction, however, are still 1.2 and 1.4 standard deviations below the mean
improvement values in the before and after study for the eastbound and
westbound directions respectively. In the current study, the direct average,
by segment, of percent improvement was 35.1%.
8.2 Comparison of Resurfacing Methods Based on Urban Profiling
Results
A comparison between the viability of various reconstruction methods
may be made based on profiling index results. The relatively low number of
after-repaving condition test sites, however, limits the viability of such a
comparison. The reconstructed facilities, for instance, only account for 8
test sites and 2.56 mi (4.12 km). There are 39 test sites at locations which
were repaved using the M&O resurfacing method, accounting for 27.61 mi
(44.43 km), and 8 test sites at locations that used the HIPR method for
resurfacing, accounting for 9.16 mi (4.12 km). As more data is gathered by
the CCD street maintenance team on the various resurfacing methods a
more meaningful argument may be formed to suggest that one resurfacing
method is superior to another. Generally, however, the resurfacing method
59


chosen must be based on multiple factors including, but not limited to,
resources available, contractor capabilities, and cost.
8.3 Real Impact of Individual Influencing Elements on the Urban
International Roughness Index
A comparison study was completed on a test surface that had been
recently resurfaced along Lowell Boulevard between Florida Avenue and
Evans Avenue to assess the real impact of individual attributes on the final
IRI. The attributes analyzed were manhole covers and cross-streets. The
roadway was analyzed and produced a final average IRI of 172.13 in/mi
(2716.72 mm/km). This value was confirmed with multiple runs and
established as the baseline IRI. During these initial runs the research team
used a marking tool in the Profiling software that allowed them to note the
locations of manhole covers and cross-streets along the collection area.
Subsequently, researchers triggered the ignore function whenever one or
both of the Profilers laser tracks made contact with a manhole cover or
cross street. Figure 8.1 shows a data collection run where manhole covers
only were removed, with gray columns demarking the ignored sections.
60


I--- OPS 1 --- OPS 2 I
Figure 8.1 Data View Output of Profiler Software Version 5.3.6 With
Ignored Manhole Cover Sections Displayed
The lines indicate the individual laser readings as measured above and
below a planer surface. The resulting IRI, ignoring manhole covers, was
164.60 in/mi (2597.87 mm/km). Figure 8.2 shows a close up view of the
laser track data as one laser crosses a manhole cover and the other laser
does not. This disparity between the tracks causes an increase in the
averaged IRI value. Next, a data collection run was completed in which
both manhole covers and cross streets were ignored which resulted in an
61


average IRI value of 150.35 in/mi (2372.96 mm/km) which is an
improvement over the initial value by 21.78 in/mi (343.75 mm/km) or
approximately 13%. Limiting the number of influencing factors on an urban
roadway predictably aides the final IRI value.
|--- ops 1 --- odsTI
Figure 8.2 Detail View of a Disparity Between Data Collection Tracks
8.4 Using High Pass Filter Settings to Control Elevation Variation
Influence in IRI Output
The high pass filter (HPF) is a tool incorporated into the Ames
Engineering Profiler System Software Version 5.3.6 used to reduce the
impact of elevation changes in the road surface on the IRI. It was observed
62


that cross-street intersection points along a segment of urban roadway tend
to negatively impact the resultant IRI values. The HPF tool was employed
to help reduce the impact of cross-street intersection points on the IRI
values for a roadway segment.
Long wavelength bumps or dips in a traditional application of the IRI
on a freeway or highway typically measure 120 feet or greater in length.
The longer lengths of these bumps or dips have very little impact on the
roughness or drivability of a high speed roadway and should therefore be
eliminated from consideration in the IRI. The HPF tool eliminates the
impact of these bumps or dips on the final reported IRI value.
The application of the IRI to urban settings requires a reassessment of
when and how filtering tools may be applied. Cross-street intersections do
impact the drivability of a roadway in an urban environment. The
usefulness of this tool is in assessing the quality of newly repaved surfaces
while excluding extraneous factors such as elevation shifts caused by cross
street intersections. Such influences should not be included in an IRI
review of a roadway in which contractor performance is being evaluated.
The HPF may be used as an effective tool in reducing or eliminating the
excess noise in a roadway segment in order to better assess the
smoothness of the surface being examined.
63


The sample group of urban roadways used in this study showed an
average width of 70 feet. This value was used in the HPF tool to reassess
196 data sets collected on 49 roadway segments around the CCD. The
application of the HPF tool showed an average improvement over unfiltered
IRI values in the before repaving condition of 4%, and an average
improvement in the after repair condition of more than 6%. The higher rate
of improvement in the after repaving condition shows that the cross-street
intersections tend to account for a larger percentage of the roughness on
smoother surfaces.
A test site with a very high number of cross-street intersections
observed is 13th Ave. between Kalamath St. and Light Rail Tracks. This
site reported a high rate of improvement after the application of the HPF;
over 10% in the westbound direction and nearly 14% in the eastbound
direction when the HPF tool was applied to the after repaving condition.
This tool appears to be effective in reducing the impact of cross-street
intersections on the final IRI values collected in urban settings. The tool is
useful when assessing before and after repaving conditions to ensure that
the resulting IRI values reflect only the roughness or smoothness of the
repaved surface rather than the uncontrollable factors of cross-street
intersections.
64


8.5 Effects of Segment Length and Starting/Stopping Motion on the
Urban IRI
Some methodological factors influencing IRI data collection include the
length of the segments and variability of speed due to traffic control devices
and congestion. The length of urban area data collection sites are short
compared to typical highway segments. The start and stop process in a
data collection runs tends to increase the final IRI value. On longer
segments this becomes less relevant because the pertinent data is
averaged over a larger mileage span. Along shorter distances the effect
becomes more pronounced. A span of 20 ft (6.1 m) was selected at the
beginning of the data collection segments and 20 ft (6.1 m) at the end data
collection segments for removal to quantify this trend. IRI values for data
runs which included the first and last 20 ft (6.1 m) tended to be higher by 2-
3% than data runs analyzed without the first and last 20 ft (6.1 m) of the
segment.
Variability of speed is controlled for in the analysis process by the data
collection software. For contiguity and good research practice maintaining
a single operating speed was used. On highway facilities conducting data
collection during off-peak hours allows for fairly constant operating speeds.
In an urban environment traffic control devices and pedestrian and
65


vehicular traffic are difficult to avoid. Therefore, variability of data collection
speeds is a necessary factor to consider in urban IRI data collection.
The traveling speed of the profiler also effects the hardware
operation of the profiling equipment. Standard high-speed profilers used on
highways have a design operating speed of 25-60 mi/hr (40.2-96.6 km/hr).
This allows a sampling rate of approximately 16,000 samples per second.
The Model 6200 LISA profiler is rated for collection speeds ranging from 5-
15.5 mi/hr (8.0-24.9 km/hr). A target data collection speed was chosen at
10 mi/hr (16.1 km/hr) for use in the CCD. This slower operating speed
allows the laser track data collectors to run a continuous sampling rate for a
highly accurate and detailed IRI rating.
66


9. Conclusions
This study was designed to begin establishing a set of standards and
practices in the application of the IRI to urban roadways. The project has
completed review of more than 95 segments of urban roadways which total
to more than 60 miles (97 km). Thirty-three of these segments were
resurfaced during the study and analyzed for IRI improvements. These
initial results help provide a knowledge base for expected values on various
roadways within the CCD. The elements influencing IRI values which are
unique to urban environments were identified. These elements represent
conditions and infrastructures which inform all aspects of urban roadway
development. As such, comparisons between urban roadways and
controlled access freeways or rural highways must be made while
considering all of the influencing factors. Comparisons between various
segments within urban areas, however, can be made while accounting for
influencing factors. The current results have shown that the equipment and
users have a very strong repeatability level between trials. The before and
after study results show that the data collected was both useful and
applicable to assessing improvement in urban roadway treatments. For
CCD roadways, it was determined that UIRI values that are less than
67


150in/mi (1658 mm/km) can be rated as good, and UIRI values that are less
than 220 in/mi (2842 mm/km) may be rated as acceptable.
68


APPENDIX A. CITY AND COUNTY OF DENVER URBAN IRI SEGMENT
DATA TABLE
1 g / a* a* a* g at i at a* g a* $ V* at
s 2 S s § | ; i | 5 * St i 1 3 3 S S
2 9 J | 3 3 8 i S
1 3 s T S "5 3 * 8 i g T |
£ i X i 5 S g 2 2 i S E 3 "S' 8 2 S 8 J i S J S
§ f A s A i J $ "5 s Ol 5 A i A s ? 8 If S J d "5 T 2 I 3 g 8 g
c g 2 R g i sj g g s fi f 8 i 5j
.* I 8 8 8 8 | $ 8 s' 3 "5 8 "S' R ffj 8 5 8 K 8 s "5 E 8 8 s | 5 7 g 8 2 p to 2 "5 i i 8 g i ¥ $ 8 $ 3 3
1 8 o s o s t < 8 o 8 0 8. i < 8 0 8 0 i < s o 8 o 1 < 8 0 8 0 2 | 8 0 8 0 r < 8 o 8 0 1 f < 8 0 8 o ? < 8 o 8 0 r < 8 o 8 o 1 < i o 8 o f < 8 o 8 o & e $ < 8 o 8 o & J < 8 o 8 o r < 8 0 8 0 & ? 4 8 o 8 o i < 8 0 8 o 4
3 K i 1 i 1 i i 5 i 3 5 1 5 3 3 3 § i
5 | | 1 E e E e E <3 E 4 E e E <3 E * E 3 E £ E 6 E 4 E 3 E 4 E <6 E
1 E : s 5 a $ 3 5
5 1 a c * 1 i 1 i £ £ c 1 i 5 ?
e -= t t c t c r c r r c - = r r
1 Z a s (9 a > o K s 1 £ o | 1 £ b S | 1 b S | i a Is 1 e 1 4 *i 3 3 > ID 3 > 6 2 CD 3 3 > a i z i a 5 3 z 3
I | | 3 6 E 1 z u 1 1 z o ? ? 0 j 3 h h 1 3 O i i u | 1 3 5 | 11 S
1 3 I z z | 5 1 5 3 S 1 5 K £ £ s I 1 £ 1 3 l - 5
69


E. 8th Ave. Downing St. York St. rt wb 0.68 m&o Before ODS1 154 33 162.73 16377 162 34 159 38 182.57 2.1%
ODS2 200 69 21035 206.46 197 26 208 42
Average 177.51 186 54 185.11 179.81 183 90
After ODS1 225.62 228 15 237.36 0.4%
ODS2 25027 245.27
Average 238.05 236 71
E, Bth Ave. Steele St. Harrison St rt wb 0.40 m&o Before ODS1 224.75 215.71 220.33 220.69 206 99 227.77 4.1%
ODS2 263 26 232.17 224 70 232.30 236 75
Average 244.01 223.94 222.52 226.50 221 87
After ODS1 201 86 203.82 207.78 1.1%
ODS2 210.34 215.08
Average 206 10 20945
Irving SL W. 17th Ave. W. 32nd Ave. rt sb 1.24 m&o Before ODS1 337.70 338.71 337.82 348.20 0.1%
ODS2 358 14 357.82 358.98
Average 347.92 348 27 346.40
After ODS1 201 18 207.06 204.81 0.2%
ODS2 207.99 203.00
Average 204 59 205.03
Irving St W. 17th Ave. W. 32nd Ave. rt nb 1.25 m&o Before ODS1 344.59 357.40 388.59 2.2%
ODS2 420.50 431.86
Average 382.55 394.63
After ODS1 232.10 229.22 262.88 3.7%
ODS2 307.60 26265
Average 269.85 255.94
S. Zuni St. W. Alameda Ave. W. Exposition Ave. rt sb 0.49 m&o Before ODS1 230.12 226.29 230.14 1.8%
ODS2 236.13 228.02
Average 233.13 227.16
After ODS1 159.83 156 36 165.50 0.9%
ODS2 173.23 172.56
Average 166.53 164.46
S. Zuni St. W. Alameda Ave. W. Exposition Ave. rt nb 0.49 m&o Before ODS1 268.27 297.87 278.56 1.9%
ODS2 261.31 266.78
Average 274.79 262.33
After ODS1 166.55 17270 164.55 0.6%
OOS2 161.15 157 79
Average 163.85 165.25
W. Exposition Ave. S. Llpan St. S. Tejon St rt eb 0.49 m&o Before ODS1 230.55 236.64 231.91 0.6%
ODS2 231.25 229.18
Average 230.90 232.91
After ODS1 153.79 155.68 160.74 0.8%
ODS2 165.96 167 52
Average 159 68 161.60
W. Exposition Ave. S. Lipan St. S. Tejon St. rt wb 0.49 m&o Before ODS1 262.66 257.87 279.46 0.6%
ODS2 293.74 303 38
Average 278.30 280.63
After OOS1 170.95 169.47 170.62 0.2%
ODS2 170.69 171.35
Average 170.82 170.41


W. Kentucky Ave. S. Federal Blvd. Morrison Rd. rt eb 1.20 m&o Before OOS1 277 99 273.82 262.80 0.9%
OOS2 244.34 25506
Average 261 17 264.44
After ODS1 16526 178.58 178.93 0.9%
OOS2 174 78 177.09
Average 160 03 177.84
W. Kentucky Ave. S. Federal Blvd. Morrison Rd. rt wb 1.20 m&o Before ODS1 287.11 284.37 270.95 0.5%
ODS2 256.74 255 59
Average 271.93 269 98
After ODS1 150.23 147 27 163.77 0.6%
ODS2 17873 178 85
Average 164.48 163 06
S. Irving St W. Mississippi Ave. W. Florida Ave. rt sb 0.49 m&o Before OOS1 320.11 327 90 310.00 1.0%
ODS2 295.43 296.54
Average 307.77 31222
After OOS1 130.60 132 57 147.10 1.6%
ODS2 160.20 165 04
Average 145.40 148.81
S. Irving St W. Mississippi Ave. W. Florida Ave. rt nb 0.49 m&o Before ODS1 322.13 31830 305.02 0.6%
ODS2 290.49 289 16
Average 306 31 303.73
After OOS1 128.04 123.92 160.21 3.2%
OOS2 199.67 169.22
Average 163 86 156.57
S. Lowell Blvd. W. Florida Ave. W. Evans Ave. rt sb 0.73 m&o Before ODS1 250.55 2S0.49 236.71 0.0%
ODS2 222.73 223.05
Average 236.64 236.77
After OOS1 152 63 145.36 162.02 1.5%
ODS2 174.89 175.18
Average 163.76 160.27
S. Lowell Blvd. W. Florida Ave. W. Evans Ave. rt nb 0.73 m&o Before oost 268.74 287.54 205.45 0.8%
OOS2 305.39 300.12
Average 297.07 293.83
After OOS1 145 88 148.81 168.97 1.3%
ODS2 189.03 192.16
Average 167.46 170.49
W. Louisiana Ave. S. Huron St. S. Zuni St. rt eb 0.96 m&o Before ODS1 300.39 310.28 306.79 3.3%
ODS2 298.91 317.57
Average 299.65 313.93
After ODS1 220.61 223.71 214.66 0.0%
ODS2 208.62 205.69
Average 214.62 214 70
W. Louisiana Ave. S. Huron St. S. Zuni St. rt wb 0.98 m&o Before ODS1 321.65 324.89 320.30 0.0%
ODS2 316.90 31575
Average 320.28 320.32
After ODS1 219.15 216.32 212.00 0.6%
OOS2 206 77 205.76
Average 212 96 211 04


E. Evans Ave. S. Holly St. S. Quebec St. rt eb 1.02 m&o Before ODS1 228 46 226 89 235.82 1.5%
OOS2 248.16 239.73
Average 238.32 233.31
After OOS1 185.33 190.59 214.37 1.8%
OOS2 237.87 24368
Average 211.60 217.14
E. Evans Ave. S. Holly St. S. Quebec St It eb 1.02 m&o Before OOS1 231.06 226 01 226.38 0.3%
ODS2 220.64 227.82
Average 225.85 226.92
After ODS1 162.44 166.99 172.15 2.8%
O0S2 174 94 164 22
Average 168.69 175.61
E. Evans Ave. S. Holly St. S. Quebec St. rt wb 1.02 m&o Before OOS1 266.07 271.36 257.01 0.5%
O0S2 246 23 244.38
Average 256.15 257.87
After O0S1 189.83 189 63 202.95 1.2%
OOS2 219.60 212.74
Average 204.72 201 19
E. Evans Ave. S. Holly St. S. Quebec St It wb 1.02 m&o Before OOS1 284.46 280.52 262.56 1.2%
ODS2 236.33 248.90
Average 260.41 264.71
After ODS1 180.61 179.19 176.55 0.1%
ODS2 172.13 174.27
Average 176.37 176.73
W. 13th Ave. Kalamath St. Light rail rt eb 0.29 reconstruct Before OOS1 333.64 345.80 349.06 0.1%
OOS2 364.76 352.05
Average 349.20 348.93
After ODS1 289.76 270.74 278.47 4.7%
ODS2 285.75 267.63
Average 287.76 269.19
W. 13th Ave. Kalamath St. Light rail rt wb 0.29 reconstruct Before ODS1 350.77 358.29 373.40 406.03 2.1%
ODS2 466.76 434.59 452.39
Average 408.77 396.44 412.90
After ODS1 317.73 326.30 338.89 1.7%
ODS2 368.04 343.49
Average 342.89 334.90
W. 13th Ave. Kalamath St. Light rail ctr eb 0.29 reconstruct After ODS1 334.49 322.48 N/A
ODS2 310.46
Average 322.48
W. 13th Ave. Kalamath St. Light rail ctr wb 0.29 reconstruct After OOS1 308.35 303.16 N/A
ODS2 297.97
Average 303.16
Tennyson St. W. 26th Ave. W. 32nd Ave. rt sb 0.49 reconstruct Before ODS1 240.40 238 53 276.18 2.2%
ODS2 320.46 30532
Average 280.43 271 93
Tennyson St W. 26th Ave. W. 32nd Ave. rt nb 0.49 reconstruct Before ODS1 236 17 232.30 254.13 4.3%
ODS2 287.40 26065
Average 261.78 246.47


ODS1 269.79 276.49
Before OOS2 226.41 230.76 251.87 1.6%
0.35 Average 249.10 254.63
ODS1 121.20 117.83 112.54
After OOS2 205.07 197 27 191.56 157.58 3.5%
Average 163 14 157.55 152.06
OOS1 240.68 238.23
Before ODS2 203.30 205.61 222.01 0.1%
wb 0.35 Average 222.09 221.92
ODS1 105 34 95.32 96.40
After ODS2 142 67 153.79 144.79 123.05 1.7%
Average 124 01 124.55 120.60
ODS1 86 57 87.36
W. 1st Ave. Federal Blvd. Knox Ct. ctr s side 0.35 reconstruct After ODS2 85.31 80.27 84.88 1.8%
Average 85 94 83.62
ODS1 68.33 79.49
W. IstAve. Federal Blvd. Knox Ct. ctr n side 0.35 reconstruct After ODS2 76.27 88.76 83.21 1.5%
Average 82.30 84.12
ODS1 340 25 326.44
E. 31st Ave. Williams Ave. York St. rt eb 0.32 reconstruct Before OOS2 320 57 306.70 324.02 2.8%
Average 330.41 317.62
ODS1 214.63 21256
E_ 31st Ave. Williams Ave. York St. It eb 0.32 reconstruct Before ODS2 218.71 235 63 220.43 2.4%
Average 216.67 224.19
ODS1 460.90 458.66
W. 46th Ave. Jason SL Pecos St. rt eb 0.36 reconstruct Before ODS2 379.33 381.48 420.09 0.0%
Average 420.11 420.07
ODS1 303.67 319.16
W. 46th Ave. Jason St. Pecos St. rt wb 0.36 reconstruct Before ODS2 300.66 343.37 316.74 6.5%
Average 302.17 331.31
0DS1 364.80 366.60
Before ODS2 367.14 362.31 365.76 0.1%
S. DTC Pkwy. E. Belleview sb 0.66 panel Average 365.97 365.55
Ave. and grind ODS1 241.14 24269
Progress ODS2 250.41 246.07 245.08 0.4%
Average 245.78 244.38
ODS1 312.07 317.14
Before ODS2 345.63 338.23 328.27 0.3%
S. OTC Pkwy. E. Belleview It 0.66 panel replacement and grind Average 328.65 327.68
Ave. ODS1 236 06 231.47
Progress ODS2 232.71 229 46 232.43 1.2%
Average 234.36 230.47
E. Belleview Ave. panel O0S1 266.57 267.81
S. DTC Pkwy. 1-225 rt nb 0.66 replacement Before ODS2 267.20 269.93 267.88 0.5%
and grind Average 266.68 268.67
E. Belleview Ave. panel OOS1 279.10 280.38
S. DTC Pkwy. 1-225 It nb 0.66 replacement Before ODS2 283.57 262.48 281.39 0.0%
and grind Average 281.34 281 43


E. 3rd Ave. Steele St. Colorado Blvd. rt eb 0.47 reconstruct Before ODS1 326 74 327 02 354.09 0.2%
ODS2 360.57 382.00
Average 353.66 354.51
E. 3rd Ave. Steele St. Jackson St. rt wb 0.38 reconstruct Before ODS1 47092 479.21 405.08 1.0%
ODS2 333.28 336 92
Average 402.10 406 06
W. 38th Ave. Federal Blvd. Sheridan Blvd. rt eb 1.47 HIPR Before ODS1 250.95 249.06 235.10 0.1%
ODS2 218.86 221.51
Average 234.90 235.29
After ODS1 12027 121.09 124.95 0.3%
ODS2 12917 129.26
Average 124.72 125.17
W. 38th Ave. Federal Blvd. Sheridan Blvd. rt wb 1.47 HIPR Before ODS1 23297 237.02 229.84 1.0%
ODS2 223.33 226.04
Average 228.15 231.53
After ODS1 111.46 120.76 122.81 3.0%
ODS2 128.90 130.09
Average 120.19 125.42
Te)on St. W. 32nd Ave. W. 44th Ave. rt sb 0.99 HIPR Before ODS1 206.39 197.92 202.24 1.2%
OOS2 192.80 209 84
Average 200.59 203.86
After ODS1 115.80 115.51 116.55 0.3%
ODS2 116.83 118.05
Average 116.32 116.78
Tejon St W. 32nd Ave. W. 44th Ave. rt nb 0.99 HIPR Before ODS1 237.94 230.27 259.90 1.8%
ODS2 27530 296.10
Average 256.62 263.18
After ODS1 125.61 123.64 127.51 0.4%
ODS2 130.06 130.72
Average 127.84 127.18
Pecos St W. 38th Ave. W. 52nd Ave. rt nb 1.41 HIPR Before ODS1 273.20 267.09 291.24 1.7%
ODS2 316.10 308.55
Average 294.65 287.82
After ODS1 153.35 151.06 158.41 0.0%
ODS2 163.39 165.85
Average 158.37 158.45
Pecos St W. 38th Ave. W. 52nd Ave. rt sb 1.41 HIPR Before ODS1 211.73 213.72 218.01 0.8%
ODS2 226.65 219 74
Average 21929 216.73
After ODS1 141.32 151 40 147.25 4.0%
ODS2 144.84 151.43
Average 143.08 151 42
Downing St. E. 26th Ave. E. MLK Blvd. rt nb 0.50 HIPR Before ODS1 293.04 300 02 317.39 0.9%
ODS2 337.59 338 90
Average 315.31 319.46
Downing St. E. 26th Ave. E. MLK Blvd. rt sb 0.50 HIPR Before ODS1 271.55 252.26 265.64 4.2%
ODS2 275.52 263.24
Average 273.53 257 75


E. 26th Ave. Downing St. York St. rt eb 0.71 HIPR Before OOS1 244.26 245.93 243.61 0.2%
ODS2 243.56 240.64
Average 24392 243.29
E. 26th Ave. Downing St. York St. rt wb 0.71 HIPR Before ODS1 264.66 271.44 268.09 1.8%
ODS2 264.83 271.44
Average 264.74 271.44
Williams St. E 26th Ave. E. MLK Blvd. rt nb 0.50 HIPR Before ODS1 335.12 329.54 327.58 1.3%
OOS2 325 92 319.74
Average 330.52 324.64
Williams St. E 26th Ave. E. MIK Blvd. rt sb 0.50 HIPR Before ODS1 290.96 293.93 304.42 0.0%
ODS2 318.02 314.76
Average 304.50 304.34
E. 29th Ave. Downing St. Williams St. rt wb 0.36 HIPR Before O0S1 369 93 31609 334.36 9.7%
ODS2 344.71 306.70
Average 357.32 311.40
E. 29th Ave. Downing St. Williams St. rt eb 0.36 HIPR Before ODS1 267.30 269.61 300.71 3.2%
ODS2 320.56 345.33
Average 293.94 307.47
E. 31st Ave. Downing St. Williams St. rt eb 0.36 HIPR Before ODS1 186 64 179.15 210.20 3.8%
OOS2 245.15 229.87
Average 215.69 204.51
E. MLKBlvd. Downing St. Williams St. rt wb 0.36 HIPR Before ODS1 262.66 262.98 274.24 0.4%
ODS2 284.19 287.12
Average 273.43 275.05
E. Yale Ave. S. Colorado Blvd. 1*25 rt wb 0.71 HIPR Before ODS1 206.71 203.51 252.99 0.4%
ODS2 297.66 304.06
Average 252.19 253.79
After ODS1 97.28 99.65 99.51 3.7%
ODS2 96.51 104.56
Average 96.90 102.12
E. Yale Ave. S. Colorado Blvd. 1-25 rt eb 0.71 HIPR Before OOS1 238.88 235.76 232.15 2.6%
ODS2 233.96 219.96
Average 236.43 227.86
After ODS1 106.67 105.79 97.71 1.6%
ODS2 90.97 87.60
Average 98.82 96.60
E. Albrook Dr. Peoria St. Dillon St. rt b 1.01 HIPR Before ODS1 228.69 234.71 238.97 1.1%
ODS2 245.46 247.01
Average 237.07 240.86
E. Albrook Dr. Peoria St. Dillon St. rt eb 1.01 HIPR Before ODS1 261.07 255.82 268.19 0.8%
ODS2 278.37 277.50
Average 269.72 266.66
S. Syracuse Way E. Yale Ave. S. Yarned St. rt nwb 0.52 panel replacement Before ODS1 196.20 195.08 182.08 0.5%
ODS2 169.23 167 79
Average 182.71 161.44
S. Syracuse Way E. Yale Ave. S. Yarned St. rt seb 0.52 panel replacement Before ODS1 180.1S 179.35 176.93 0.1%
ODS2 173.42 174.81
Average 176.78 177.08


E. Colfax Avc. Colorado Blvd. Quebec St. rt eb 1.96 CDOT m&o Before ODS1 198 34 197 09 204.76 0.2%
OOS2 21065 213.05
Average 204.49 20507
E. Colfax Ave. Colorado Blvd. Quebec St. rt wb 1.96 CDOT m&o Before ODS1 168.86 177.21 175.71 1.8%
ODS2 178.14 178.63
Average 173.50 177.92
E. Quincy Ave. S. Wadsworth Blvd. City Limit rt wb 0.49 m&o Before ODS1 244.70 245.23 244.30 1.8%
OOS2 237.82 249.44
Average 241.26 247.34
After ODS1 182.52 166.11 172.77 2.1%
OOS2 157.77 164.68
Average 170.15 175.40
E. Quincy Ave. S. Wadsworth Blvd. City Limit rt eb 0.49 m&o Before ODS1 221.92 220.69 222.07 0.5%
ODS2 223.75 221.92
Average 222.84 221.31
After ODS1 211.73 208 19 201.03 0.2%
ODS2 189.68 194.52
Average 200.71 201.36
E. 13th Ave. Yorks St. Colorado Blvd. rt wb 1.00 reconstruct Before ODS1 244.92 245.79 256.94 1.5%
ODS2 274.29 262.76
Average 259.61 254 27
E. 13th Ave. Yorks St. Colorado Blvd. It wb 1.00 reconstruct Before ODS1 378.70 321.08
ODS2 263.47
Average 321.06
W. Wagontrail Dr. Irving St. Perry St. rt wb 0.44 m&o Before OOS1 436 89 42666 457.22 1.2%
OOS2 485.23 478.10
Average 461.06 453.36
W. Wagontrail Dr. Irving St. Perry St. rt eb 0.44 m&o Before ODS1 331.51 312.11 302.49 3.2%
ODS2 287.35 276.99
Average 309.43 295.55
S. Perry St. Tufts St. Beltewood Ave. rt nb 0.37 m&o Before ODS1 396.58 417.23 397.41 1.9%
ODS2 385.44 388.40
Average 392.01 402.61
S. Perry St Tufts St. Bellewood Ave. rt sb 0.37 m&o Before ODS1 378.95 382.18 372.86 2.9%
ODS2 351.60 378.72
Average 365.27 380.45
W. Tufts Ave. Newton St. Sheridan blvd. rt eb 0.82 m&o Before ODS1 304.35 26529 301.39 0.5%
ODS2 300.43 315.48
Average 302.39 300.38
W. Tufts Ave. Newton St. Sheridan blvd. rt wb 0.82 m&o Before ODS1 291.04 298.94 298.37 1.5%
O0S2 299.55 303.95
Average 295.29 301.45
S. Irving St. Wagontrail Dr. Chenango Ave. rt sb 0.24 m&o Before ODS1 413 65 417.69 410.54 1.6%
ODS2 398.15 412 64
Average 405.90 415.17
S. Irving St Wagontrail Dr. Chenango Ave. rt nb 0.24 m&o Before ODS1 435.26 426.01 416.24 0.6%
ODS2 401.92 401.74
Average 418.59 413.68


S. Irving St. Union Ave. Wagontrall Or. rt sb 0.11 m&o Before OOS1 555 75 548.58 537.86 2.9%
OOS2 49613 54698
Average 526.94 54876
S. Irving St. Union Ave. Wagontrall Dr. rt nb 0.11 m&o Before ODS1 518.71 509 10 527.82 0.3%
ODS2 534.75 548.70
Average 526.73 S26.90
E. 11th Ave. Quebec St. Yosemite St. rt wb 0.98 m&o Before ODS1 293.34 289.53 290.18 283.69 0.4%
OOS2 271.40 278.28 279.43
Average 282.37 283 90 284.80
Park Ave. West E. Colfax Ave. Welton St. rt sb 0.71 m&o Before OOS1 28383 295 62 425.54 2.8%
O0S2 550.40 572.12
Average 417.11 433.97
Park Ave. West E. Colfax Ave. Welton St. It sb 0.71 m&o Before OOS1 233.70 238.91 251.66 4.4%
OOS2 254.03 279 98
Average 243.87 25944
Park Ave. West E. Colfax Ave. Welton St. rt nb 0.71 m&o Before ODS1 317.92 316.26 459.94 1.4%
ODS2 611.34 594 25
Average 464.63 455.25
Park Ave. West E. Colfax Ave. Welton St. It nb 0.71 m&o Before OOS1 227 89 235.12 246.28 2.4%
OOS2 272 89 24920
Average 25039 242.16


APPENDIX B. BEFORE AND AFTER REPAVING SEGMENT DATA
RESULTS.
Length IRI
Street From To Lane Dir (mi) (km) Treatment Run (in/mi) (mm/km)
E. 8th Ave. Steele St. Harrison St. rt wb 0.40 0.64 m&o Before 227.77 3594.77
After 207.78 3279.27
Irving St. W. 17th Ave. W. 32nd Ave. rt sb 1.24 2.00 m&o Before 348.20 5495.49
After 204.81 3232.44
Irving St. W. 17th Ave. W. 32nd Ave. rt nb 1.25 2.01 m&o Before 388.59 6133.00
After 262.89 4149.18
S. Zuni St. W. Alameda Ave. W. Exposition Ave. rt sb 0.49 0.79 m&o Before 230.14 3632.25
After 165.50 2611.97
S. Zuni St. W. Alameda Ave. W. Exposition Ave. rt nb 0.49 0.79 m&o Before 278.56 4396.42
After 164.55 2597.02
W. Exposition Ave. S. Lipan St. S. Tejon St. rt eb 0.49 0.79 m&o Before 231.91 3660.11
After 160.74 2536.89
W. Exposition Ave. S. Lipan St. S. Tejon St. rt wb 0.49 0.79 m&o Before 279.46 4410.70
After 170.62 2692.78
78


w. Kentucky Ave. S. Federal Blvd. Morrison Rd. rt eb 1.20 1.93 m&o Before 262.80 4147.76
After 178.93 2824.06
W. Kentucky Ave. S. Federal Blvd. Morrison Rd. rt wb 1.20 1.93 m&o Before 270.95 4276.39
After 163.77 2584.75
S. Irving St. W. Mississippi Ave. W. Florida Ave. rt sb 0.49 0.79 m&o Before 310.00 4892.59
After 147.10 2321.69
S. Irving St. W. Mississippi Ave. W. Florida Ave. rt nb 0.49 0.79 m&o Before 305.02 4814.07
After 160.21 2528.60
S. Lowell Blvd. W. Florida Ave. W. Evans Ave. rt sb 0.73 1.17 m&o Before 236.71 3735.87
After 162.02 2557.05
S. Lowell Blvd. W. Florida Ave. W.Evans Ave. rt nb 0.73 1.17 m&o Before 295.45 4662.99
After 168.97 2666.82
W. Louisiana Ave. S. Huron St. S. Zuni St. rt eb 0.98 1.58 m&o Before 306.79 4841.97
After 214.66 3387.90
W. Louisiana Ave. S. Huron St. S. Zuni St. rt wb 0.98 1.58 m&o Before 320.30 5055.19
After 212.00 3345.95
E. Evans Ave. S. Holly St. S. Quebec St. rt eb 1.02 1.64 m&o Before 235.82 3721.82
After 214.37 3383.32
79


E. Evans Ave. S. Holly St. S. Quebec St. It eb 1.02 1.64 m&o Before 226.38 3572.95
After 172.15 2716.97
E. Evans Ave. S. Holly St. S. Quebec St. rt wb 1.02 1.64 m&o Before 257.01 4056.34
After 202.95 3203.12
E. Evans Ave. S. Holly St. S. Quebec St. it wb 1.02 1.64 m&o Before 262.56 4143.89
After 176.55 2786.45
W. 13th Ave. Kalamath St. Light rail rt eb 0.29 0.47 reconstruct Before 349.06 5509.18
After 278.47 4395.04
W. 13th Ave. Kalamath St. Light rail rt wb 0.29 0.47 reconstruct Before 406.03 6408.34
After 338.89 5348.63
W. 1st Ave. Federal Blvd. Knox Ct. rt eb 0.35 0.56 reconstruct Before 251.87 3975.13
After 157.58 2487.11
W. 1st Ave. Federal Blvd. Knox Ct. rt wb 0.35 0.56 reconstruct Before 222.01 3503.86
After 123.05 1942.13
W. 38th Ave. Federal Blvd. Sheridan Blvd. rt eb 1.47 2.37 HIPR Before 235.10 3710.46
After 124.95 1971.98
80


W. 38th Ave. Federal Blvd. Sheridan Blvd. rt wb 1.47 2.37 HIPR Before 229.84 3627.52
After 122.81 1938.21
Tejon St. W. 32nd Ave. W. 44th Ave. rt sb 0.99 1.59 HIPR Before 202.24 3191.83
After 116.55 1839.49
Tejon St. W. 32nd Ave. W. 44th Ave. rt nb 0.99 1.59 HIPR Before 259.90 4101.95
After 127.51 2012.46
Pecos St. W. 38th Ave. W. 52nd Ave. rt nb 1.41 2.27 HIPR Before 291.24 4596.50
After 158.41 2500.15
Pecos St. W. 38th Ave. W. 52nd Ave. rt sb 1.41 2.27 HIPR Before 218.01 3440.81
After 147.25 2324.02
E. Yale Ave. S. Colorado Blvd. I-25 rt wb 0.71 1.14 HIPR Before 252.99 3992.89
After 99.51 1570.55
E. Yale Ave. S. Colorado Blvd. I-25 rt eb 0.71 1.14 HIPR Before 232.15 3663.90
After 97.71 1542.14
E. Quincy Ave. S. Wadsworth Blvd. City Limit rt wb 0.49 0.79 m8iO Before 244.30 3855.70
After 172.77 2726.79
E. Quincy Ave. S. Wadsworth Blvd. City Limit rt eb 0.49 0.79 m8iO Before 222.07 3504.89
After 201.03 3172.82
81


BIBLIOGRAPHY
American Society for Testing and Materials. (2004). ASTM E950 -
98(2004) Standard Test Method for Measuring the Longitudinal Profile
of Traveled Surfaces with an Accelerometer Established Inertial
Profiling Reference. .
Awatashi, G., Sigh, T. and Das, A. (2003). On Pavement Roughness
Indices. IE (!) Journal-CV. 84 (May), 33-37.
Bandura, N. and Gunaratne, M. (2001). Current and Future Pavement
Maintenance Prioritization Based on Rapid Visual Condition
Evaluation. J. Transp. Eng., Ml .2 (March/April), 116-123.
Colorado Department of Transportation. (2007). Standard Specifications
for Road and Bridge Construction Revision Notice. Colorado, USA.
Colorado Department of Transportation. Standards Practice for Operating
Profilers and Evaluating Pavement Profiles. Colorado Procedure 74-
07a. Colorado, USA.
Federal Highway Administration. (2006). 2006 Status of the Nation's
Highways, Bridges, and Transit: Conditions and Performance.

Jiang, Y. and Li, S. (2005). Gray System Model for Estimating the
Pavement International Roughness Index. J. of Performance of
Constructed Facilities. (February), 62-68.
Juang, C., Amirkhanian, S. (1992). Unified Pavement Distress Index for
Managing Flexible Pavements. J. Transp. Eng.. 118.5
(September/October), 682-699.
Model 6200 Laser Inertial Surface Analyzer (LISA) User Manual. Ames
Engineering, Ames, IA.
Nemmers, C. J., Gagarin, N., Mekemson, J. R. (2006). Assessing IRI vs.
PI as a Measure of Pavement Smoothness. Organizational Results
Research Report. ORO6.017.
82


Perera, R. W., Kohn, S. D. and Tayabji, S. (2004). Achieving a High Level
of Smoothness in Concrete Pavements Without Sacrificing Long-Term
Performance. FHWA. FHWA-HRT-05-068.
Reggin, A., Shalaby, A., Emanuels, R., and Michel, G. (2008) Urban
Considerations for Using Road Roughness to Manage Road
Networks. Paper presented at the 2008 International Conference on
Managing Pavement Assets, Alberta, Canada.
Ruotoistenmaki, A., Seppala, T., Kanto, A. (2006). Comparison of
Modeling and Measurement Accuracy of Road Condition Data. T
Transp. Eng., 132.9 (September), 715-721.
Sayers, M. W. and Karamihas, S. M. (1998). Little Book of Profiling; Basic
Information About Measuring and Interpreting Road Profiles. The
Regent of the University of Michigan. Ml, USA.
Smith, K. L., Titus-Glover, L., and Evans, L. D. (2002). Pavement
Smoothness Index Relationships, Final Report. U. S. Department of
Transportation. FHWA-RD-02-057.
Soleymani, H. R., Palsat, D., Mesher, D. and Campbell, P. (2008). The
Effect of Pavement Crack Treatments on IRI and Surface Profile A
Case Study in Alberta. Paper presented at the 2008 International
Conference on Managing Pavement Assets, Alberta, Canada.
Song, I., Teubert, C., Hayhoe, G. F. (2008). "California Type Profilograph
Comparisons. Transportation Systems 2008 Workshop. Pheonix, AZ.
Souza, R. O., Neto, S. D., and Farias, M. M. (2007). Study of Long-Term
Pavement Performance (LTPP) Pavement Deflections.

Steyn, W. (2001). Consideration of Pavement Roughness Effects on
Vehicle-Pavement Interaction. Paper presented at the 2001 SATC,
Pretoria, South Africa.
Wilde, W. (2007). Implementation of an International Roughness Index for
Mn/DOT. Mn/DQT Final Report, Minnesota State University, Mankato.
83