r.
:.
on.

..
g V as ,
”mm... ,
.um .5» Envy»... .2.
.4: .
o. rug".
.5... ,
E!

.8 La..." , w
fiaa$

.51.?
v .

.r (
. If: x
.91..

.z

3

 

 

 

 

u ) 51.104,

‘ .
.
aria»? .4:
an.»
. r z

.
. . . v . ‘ p )‘C:
a , . .l “4 .
v...‘ I p Q . v , v I
t r
. bfycuthfl ‘ A H!) V w 3

$2“?
. i...
1

3m“...
um? $sz

up. n. . 1.2..
nudnfluhud

. . ‘
in pa) . a
.v‘QX‘31 i. It.“
.99. .3i521.)
A n at. {3.

In.

x Hurt;

1!.

: v.1“

117..-?! ;
A .z .1.
A? x \ 7

FE. X .. , $2 .
2 ES. .. in???

 

 

 

 

 

 

,i.

. . 3.» .crLJwa.:

 

as?" ', LIBRARY

b. Michigan State
"7/ University
, 0 ')

This is to certify that the
thesis entitled

QUANTIFYING THE RELATIVE EFFECTS OF DESIGN AND
SITE FACTORS ON THE PERFORMANCE OF RIGID
PAVEMENTS USING SPS-2 LTPP DATA

presented by

ASWANI SASAANKA PULIPAKA

has been accepted towards fulfillment
of the requirements for the

M. 8. degree in Civil Engineering

 

 

ILL-Mammfessor’s Signature
3/?1/ as"

Date

MSU is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c2/CIRC/DateDuep65-p. 15

QUANTIFYING THE RELATIVE EFFECTS OF DESIGN AND SITE
FACTORS ON THE PERFORMANCE OF RIGID PAVEMENTS
USING SPS-Z LTPP DATA

By

Aswani Sasaanka Pulipaka

A THESIS
Submitted to
Michigan State University

in panial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Civil and Environmental Engineering

2005

 

ABSTRACT

QUANTIFYING TI-HZ RELATIVE EFFECTS OF DESIGN AND SITE FACTORS
ON THE PERFORMANCE OF RIGID PAVEMENTS USING SPS-2 LTPP DATA

By

Aswani Sasaanka Pulipaka
The results of a study conducted to evaluate the relative effects of various design and site
factors on the performance of jointed plain concrete (JPC) pavements are presented here.
Data related to the Specific Pavement Study-2 (SPS-Z) experiment of the Long Term
Pavement Performance (LTPP) program were used in the study. The experiment was
designed to investigate the effects of concrete slab thickness, base type, drainage,
concrete flexural strength and slab width on the performance of J PC pavements. Based on
an assessment of performance using published standards, a majority of the SPS—2 sections
are exhibiting “good” performance. Data were analyzed by employing appropriate
frequency- and magnitude-based statistical methods. On the whole, base type was found
to be the most critical design factor affecting performance, in terms of cracking and
roughness (IRI). Pavement sections with permeable asphalt-treated base and in-pavement
drainage performed better than those with a dense-graded aggregate base or a lean
concrete base. PCC slab thickness also played an important role in determining the
cracking performance. PCC flexural strength and slab width only have marginal effects

on performance, at this point in time.

 

To my parents, Syamala and Satyanarayana,
to my brother Anirudh,

and to my other kith and kin.

iii

 

ACKNOWLEDGEMENTS

I would first like to thank my academic advisor Dr. Neeraj Buch for his support
and guidance throughout my graduate program. I would also like to thank the National
Cooperative Highway Research Program (NCHRP) for sponsoring this research.

I would like to express thanks to Dr. Karim Chatti and Dr. Gilbert Baladi for
being a part of the thesis committee. The guidance provided by Dr. Denis Gilliland and
Dr. Richard Lyles has played an important role in the completion of this research study.

Heartfelt appreciation is due to my friend Dr. Syed W. Haider (former graduate
research assistant) for his invaluable guidance during this research. Mr. Praveen
Desaraju’s (former graduate assistant) help in getting me acquainted with the research is
appreciated. I also appreciate Mr. Phani R. Mullapudi’s help during final formatting of
my thesis.

Finally, I would like to express my gratitude to my alma mater Michigan State

University for everything it has given to me.

iv

TABLE OF CONTENTS

LIST OF TABLES ...........................................................................
LIST OF FIGURES ..........................................................................
CHAPTER 1 — INTRODUCTION ........................................................
1.1 BACKGROUND ......................................................................
1.2 RESEARCH GOAL AND OBJECTIVES .........................................
1.3 SCOPE OF THE STUDY ............................................................
1.4 LAYOUT OF THE THESIS .........................................................
CHAPTER 2 - AN OVERVIEW OF THE SPS-2 EXPERIMENT .................
2.1 INTRODUCTION ....................................................................
2.2 STRATEGIC STUDY OF STRUCTURAL FACTORS FOR RIGID
PAVEMENTS — SPS-2 EXPERIMENT ..................................................
2.2.1 Objective ......................................................................
2.2.2 Experiment Design ...........................................................
2.2.3 Current Status of the Experiment ..........................................
2.2.4 Monitoring Requirements ...................................................
CHAPTER 3- LITERATURE REVIEW ................................................
3.1 SIGNIFICANCE OF DESIGN FACTORS .......................................
3.2 FINDINGS FROM PREVIOUS STUDIES ON SPS-2 .........................

3.3 EFFECTS OF EXPERIMENTAL FACTORS ON RIGID PAVEMENT

PERFORMANCE ..............................................................................

3.3.1 Effect of structural factors ...................................................

3.3.2 Effects of site factors .........................................................
CHAPTER 4- DATA AVAILABILITY AND EXPERIMENT REVIEW ........
4.1. INTRODUCTION ....................................................................
4.2. ASSESSMENT OF DATA AVAILABILITY .....................................

4.2.1. General Site Information ....................................................
4.3. EXPERIMENT REVIEW ............................................................

4.3.1. Construction Deviations based on Construction Reports ...............

4.3.2. Design versus Construction Review based on Data .....................
CHAPTER 5- EVALUATION OF KEY PERFORMANCE TRENDS ............
5.1. GENERAL ASSESSMENT OF PERFORMANCE ..............................
5.2. TRANSVERSE CRACKING .......................................................
5.3. LONGITUDINAL CRACKING ....................................................
5.4. JOINT FAULTING ...................................................................
5.5. ROUGHNESS (IRI) ..................................................................
CHAPTER 6- ANALYSIS METHODS ..................................................
6.1. PERFORMANCE INDICATORS ..................................................

viii

X

1
1
2
3
3

4
4

4
4
4
8
ll

12
12
13

15
15
24

27
27
27
28
35
35
36

46
46
48
51
52
55

59
59

 

6.2. SITE-LEVEL ANALYSIS METHODS ........................................... 61

6.3. OVERALL ANALYSIS METHODS .............................................. 63
6.3.1. Linear Discriminant Analysis (LDA) ...................................... 63
6.3.2. Binary Logistic Regression (BLR) ......................................... 66
CHAPTER 7- EFFECTS OF EXPERIMENTAL FACTORS ON PAVEMENT
PERFORMANCE ............................................................................ 68
7.1. EFFECT OF CONSTRUCTION ON PAVEMENT PERFORMANCE ...... 68
7.1.1. Transverse and Longitudinal Cracking .................................... 69
7.1.2. Roughness ..................................................................... 71
7.1.3. Drainage Issues ................................................................ 72
7.2. SITE-LEVEL ANALYSES ........................................................... 73
7.2.1. Effect of design features on performance- Comparisons at level—A 76
7.2.2. Effect of design features- Paired Comparisons at level-B .............. 81
7.3. OVERALL ANALYSIS ............................................................. 85
7.3.1. Frequency-based Methods .................................................. 87
7.3.2. Analysis of Variance ......................................................... 96
7.3.3. Comparison of Designs ...................................................... 103
7.4. SYNTHESIS OF RESULTS FROM ANALYSES .............................. 111
7.4.1. Effects of structural factors for rigid pavements — SPS-2
experiment ............................................................................. 102

CHAPTER 8- ENVIRONMENTAL AND LOAD EFFECTS ON RIGID

PAVENIENTS ................................................................................. 1 18
8.1. INTRODUCTION ..................................................................... 118
8.2. DESCRIPTION OF THE SPS-8 EXPERIMENT ................................. 118
8.3. CURRENT STATUS OF SPS-8 RIGID PAVEMENTS ........................ 120
8.4. DATA AVAILABILITY ............................................................ 120
8.4.1. Construction Reports ......................................................... 120
8.4.2. Climate Data .................................................................. 121
8.4.3. Traffic Data ................................................................... 121
8.5. DESIGN VERSUS ACTUAL CONSTRUCTION REVIEW .................. 121
8.6. DISTRESS OCCURRENCE ........................................................ 122
8.7. ANALYSES AND RESULTS ...................................................... 126
8.7.1. Site-Level Analysis .......................................................... 128
8.7.2. Overall Analysis .............................................................. 128
8.7.3. Comparison between Similar Designs of SPS-8 and SPS-2
Experiments ............................................................................ 133
8.8. SUMMARY OF RESULTS ......................................................... 135
CHAPTER 9- CONCLUSIONS .......................................................... 136
9.1. EFFECTS OF STRUCTURAL FACTORS FOR RIGID PAVEMENTS —
SPS-2 EXPERIMENT ................................................................... 138
9.2 ENVIRONMENTAL AND LOAD EFFECTS ON RIGID PAVEMENT
PERFORMANCE ........................................................................ 141
9.3 LIMITATIONS OF THE EXPERIMENTS AND ANALYSES .................. 142

vi

9.4 RECOMIVIENDATION S FOR FUTURE DATA COLLECTION AND

RESEARCH .............................................................................. 144
APPENDIX .................................................................................... 147
REFERENCES ................................................................................

vii

LIST OF TABLES

Table 2- 1 SPS-2 Experiment Design Matrix ..................................................................... 5
Table 2- 2 Status of the SPS-2 design factorial ................................................................ 10
Table 2- 3 Monitoring Requirements ............................................................. 11
Table 3-1 Summary of PCC Pavement Base Types (Smith et al. 1997) .......................... 21
Table 4- 1 Summary of traffic data availability ................................................................ 30
Table 4- 2 Summary of data availability (percent of sections) for PCC properties .......... 32
Table 4- 3 Summary of availability (percent of sections) PCC flexural strength data ..... 32
Table 4- 4 CTE data obtained from FHWA 33
Table 4- 5 Summary of data availability for SPS-2 experiment ....................................... 34
Table 4- 6 Summary of deviation in thickness from design ............................................. 40
Table 6- 1 Example calculation of relative performance for the effect of slab thickness on
longitudinal cracking ................................................................................................ 63
Table 7- 1 Subjective ratings of drainage functioning for SPS-2 test sections based on
video inspection results (Hall and Correa 2002) ....................................................... 74
Table 7-2 Effects of drainage*, based on Level-A analysis ............................................. 77
Table 7- 3 Effects of base type based on Level-A analysis .............................................. 78
Table 7- 4 Effects of slab thickness, based on Level-A analysis ...................................... 79
Table 7- 5 Effects of flexural strength, based on Level-A analysis .................................. 81
Table 7- 6 Effects of slab width based on Level-A analysis ............................................. 82
Table 7- 7 Effect of drainage*, based on Level-B analysis .............................................. 83
Table 7- 8 Effect of base type on cracking, based on Level B analysis ........................... 84
Table 7- 9 Effect of slab thickness on cracking, based on Level B analysis .................... 85
Table 7— 10 Summary of results from LDA on transverse cracking- Overall .................. 90
Table 7— 11 Results from LDA on transverse cracking, WF zone .................................... 90
Table 7- 12 Summary of results from LDA on longitudinal cracking .............................. 91
Table 7- 13 Results from LDA on Longitudinal cracking, in WF zone ........................... 91
Table 7- 14 Summary of results from LDA on initial roughness ..................................... 91
Table 7- 15 Summary of results from LDA on initial roughness, in WF zone ................. 92
Table 7- 16 Summary of results from LDA on roughness ................................................ 92
Table 7- 17 Summary of results from LDA on roughness, in WF zone ........................... 92
Table 7- 18 Summary of p-values from BLR for determining the effect of experimental
factors on pavement performance measures- Overall ............................................... 95
Table 7- 19 Summary of p-values from BLR for determining the effect of experimental
factors on pavement performance measures- WF Zone ........................................... 95
Table 7- 20 Summary of p-values from ANOVA for determining the effects of design
factors on pavement performance— Overall ............................................................ 98
Table 7- 21 Summary of marginal means from ANOVA for determining the effect of
main design factors on pavement performance measures—Overall ......................... 98
Table 7- 22 Summary of p-values from ANOVA for determining the effects of design
factors on pavement performance— WF Zone ......................................................... 99
Table 7- 23 Summary of marginal means from ANOVA for determining the effect of
main design factors on pavement performance measures—WP Zone ..................... 99
Table 7- 24 Summary of p-values from one-way ANOVA for determining the effect of
site factors on pavement performance measures .................................................... 100

viii

Table 7- 25 Summary of marginal means from one-way ANOVA for determining the

effect of site factors on pavement performance measures ...................................... 100
Table 7- 26 Summary of p-values for comparisons of standard deviates— Change in

roughness ................................................................................................................ 106
Table 7- 27 Summary of mean change in roughness ...................................................... 106
Table 7- 28 Design features of designs 0201 through 0212 ........................................... 107
Table 7- 29 Design features of designs 0213 through 0224 ........................................... 107
Table 7- 30 Summary of effects of design and site factors for rigid pavements ............ 117
Table 8- 1 SPS-8 Experiment Design ................................................ ' ........... 119
Table 8- 2 Distribution of rigid pavement sections in the SPS-8 experiment ............. 120
Table 8— 3 Summary of data availability for SPS-8 experiment —Rigid pavements ...... 123
Table 8- 4 Summary of available traffic data .................................................. 124
Table 8- 5 Summary of traffic data available from the construction reports .............. 124
Table 8- 6 Distresses in SPS-8 sections ........................................................ 126

ix

LIST OF FIGURES

Figure 2- 1 Geographical distribution of the SPS-2 sites ................................................... 8
Figure 4- 1 Distribution of SPS-2 sites with respect to climate ........................................ 37
Figure 4- 2 Actual thickness of sections with target thickness of 203 mm ...................... 39
Figure 4- 3 Actual thickness of sections with target thickness of 279 mm ...................... 39
Figure 4- 4 Actual base thickness of sections on DGAB .................................................. 40
Figure 4- 5 14-day flexural strength of sections with target strength of 3.8 MPa ............ 41
Figure 4- 6 28-day flexural strength of sections with target strength of 3.8 MPa ............ 42
Figure 4- 7 365-day flexural strength of sections with target strength of 3.8 MPa .......... 42
Figure 4- 8 4-day flexural strength of sections with target strength of 6.2 MPa .............. 43
Figure 4- 9 28-day flexural strength of sections with target strength of 6.2 MPa ............ 43
Figure 4- 10 365-day flexural strength of sections with target strength of 6.2 MPa ........ 44
Figure 4— 11 Age distribution in the SPS-2 sites .............................................................. 45
Figure 4- 12 Variation in traffic across SPS-2 sites .......................................................... 45
Figure 5-1 Distribution of transverse cracking (percent slabs cracked) ........................... 48
Figure 5-2 Site-wise occurrence of transverse cracking for SPS-2 test sections .............. 50
Figure 5— 3 Distribution of longitudinal cracking, m ........................................................ 52
Figure 5- 4 Site-wise occurrence of longitudinal cracking in SPS-2 test sections ........... 52
Figure 5- 5 Occurrence of transverse cracking by site factor ........................................... 53
Figure 5- 6 Occurrence of transverse cracking by design factor ...................................... 53
Figure 5- 7 Extent of longitudinal cracking by site factors .............................................. 53
Figure 5- 8 Extent of longitudinal cracking by design factors .......................................... 54
Figure 5- 9 Site-wise occurrence of faulting in SPS-2 test sections ................................. 54
Figure 5— 10 Percent of joints that faulted >= 2.0 mm ...................................................... 55
Figure 5- ll Site-wise occurrence of final roughness values for SPS-2 test sections ...... 56
Figure 5- 12 Distribution of roughness, IRI/ km .............................................................. 57
Figure 5- 13 Extent of faulting >=1.0 mm in site factors ................................................. 57
Figure 5- 14 Extent of faulting >=1.0 mm in design factors ............................................ 58
Figure 5- 15 Extent of roughness in site factors ............................................................... 58
Figure 5- l6 Extent of roughness in design factors .......................................................... 58
Figure 6- 1 Correlation between PI and AREA/age ........................................... 60
Figure 6- 2 Example of violation of residual assumptions for ANOVA ..................... 65
Figure 7- 1 Transverse cracking with time - All sections ................................................. 70
Figure 7- 2 Longitudinal cracking with time - All sections .............................................. 70
Figure 7- 3 Transverse cracking with time - Selected sections (without Nevada) ........... 71
Figure 7- 4 Longitudinal cracking with time - Selected sections (without Nevada) ........ 71
Figure 7- 5 IRI with time - Selected sections (without Nevada) ...................................... 72
Figure 7- 6 Effect of design factors on change in IRI ..................................................... 105
Figure 7- 7 Comparison among designs 0210 through 0212 .......................................... 108
Figure 7- 8 Comparison among designs 0213 through 0224 .......................................... 108
Figure 7- 9 Comparison of designs 0201 through 0212 ................................................. 109
Figure 7- 10 Comparison of designs 0213 through 0224 ............................................... 109
Figure 7- 11 Comparison of designs 0201 through 0212 ............................................... 110
Figure 7- 12 Comparison of designs 0213 through 0224 ............................................... 110

Figure 8- 1 Thickness deviations in sections with target PCC thickness of 203 mm.....124

Figure 8- 2 Thickness deviations in sections with target PCC thickness of 279 mm . . ...125

Figure 8- 3 Deviation from target base thickness of 152 mm ................................ 125
Figure 8— 4 Average l4-day flexural strength of PCC ........................................ 125
Figure 8- 5 Current IRI in SPS-8 sections ...................................................... 126
Figure 8- 6 Average transverse cracking by experimental factors .......................... 130
Figure 8- 7 Average transverse cracking by subgrade type .................................. 130
Figure 8- 8 Average long. spalling by experimental factors ................................. 131
Figure 8- 9 Average long. spalling by subgrade type ......................................... 131
Figure 8- 10 Average no. of joints having faulting > 2% by experimental factors ....... 131
Figure 8- 11 No. of joints having faulting > 2% by subgrade type ........................ 132
Figure 8- 12 Average IRI by experimental factors ............................................ 132
Figure 8- 13 Average IRI by subgrade type .................................................... 132

xi

CHAPTER 1 - INTRODUCTION
1.1 BACKGROUND

Pavement design and construction have a major influence on pavement performance,
which in turn has a significant impact on the management of highway infrastructure.
Each design feature affects the performance of a pavement in different ways, depending
on site conditions. Highway agencies currently use a wide variety of alternative design
features that have evolved over time based on experience and/or research studies.
However, for most of the highway agencies, the design features are often a part of the
agency “standar ” that remain unchanged over years (Gharaibeh and Datter, 2001).

The Long-Term Pavement Performance (LTPP) program was initiated in 1987 as a
part of the Strategic Highway Research Program (SHRP) to address the needs of the
highway agencies by means of studying in-service pavements. This program comprises of
many full-scale field experiments that address a variety of issues aimed at improving the
design and management of highways. The objective of LTPP program is to “Increase
pavement life by investigation of the long-term performance of various designs of
pavement structures and rehabilitated pavement structures, using different materials and
under different loads, environments, subgrade soils, and maintenance practices” (SHRP
1986). From this broad objective, six specific objectives were established for the LTPP
program (Elkins et a]. 2003):

0 Evaluate existing design methods.

0 Develop improved design methodologies and strategies for the rehabilitation of

existing pavements.

0 Develop improved design equations for new and reconstructed pavements.

0 Determine the effects of loading, environment, material properties and
variability, construction quality, and maintenance level on pavement distress
and performance.

0 Determine the effects of specific design features on pavement performance.

0 Establish a national long-term pavement database to support SHRP objectives

and needs.

In order to meet these objectives, the program has two complementary sets of studies—
General Pavement Studies (GPS) and Specific Pavement Studies (SP8). The GPS use
existing pavements while the SPS usually use newly constructed pavements. The SPS-2

experiment is aimed at the fifth of the above said objectives of the LTPP program.
1.2 RESEARCH GOAL AND OBJECTIVES

For specific site conditions, the performance of pavements will depend not only on
thickness of each pavement layer and material properties, but also on other design
features such as slab width, base type, and in-pavement drainage. The performance of a
pavement system also depends on the interaction effects between these features.

This research focuses on the effects of design features on the performance of rigid
pavements, based on an evaluation of the test sections from the ongoing Specific
Pavement Study-2 (SPS-2) experiment of the LTPP. The SPS-2 experiment, titled as
“Strategic Study of Structural Factors for Rigid Pavements”, provides an opportunity for
the study of the effects of PCC slab thickness, in-pavement drainage, base type, slab
width and PCC flexural strength, in various site conditions (soil-climate combinations).

In addition to this, rigid pavement sections of the SPS-8 experiment were used to

investigate the effects of environment and traffic on rigid pavement performance. The
SPS-8 experiment is aimed at studying the effects of environment, in the absence of

heavy loads, on pavement performance.
1.3 SCOPE OF THE STUDY

The findings presented in this thesis are based on the assessment of test sections included
in the SPS-2 and SPS-8 experiments of the LTPP. The analysis is limited to the use of
relevant LTPP data (i.e. DataPave) available from Release 17 of National Information

Management System (NIMS) database.
1.4 LAYOUT OF THE THESIS

The thesis is divided into 9 chapters including this introductory chapter. A description of
the experiment design and the current status of the SPS-2 experiment are presented in
Chapter 2. A summary of findings from a review of relevant literature is presented in
Chapter 3. A summary of data availability and findings from a review of the experiment
are presented in Chapter 4. Chapter 5 contains results of an evaluation of performance
trends of the SPS—2 sections. Based on the extent and occurrence of distresses, different
methods of analysis were employed. A brief description of these methods is presented in
Chapter 6. The results from statistical analyses conducted on performance of the SPS-2
test sections are presented in Chapter 7. An evaluation of the performance of rigid
pavement sections in SPS-8 experiment is presented in Chapter 8. Chapter 9 contains the

conclusions and recommendations for future research.

CHAPTER 2 - AN OVERVIEW OF THE SPS-2 EXPERIMENT
2.1 INTRODUCTION
This chapter presents an overview of the SPS-2 experiment in terms of its goals,
experiment design, and associated experimental factors. The chapter also contains a brief

description of the current status of the experiment.

2.2 STRATEGIC STUDY OF STRUCTURAL FACTORS FOR RIGID
PAVEMENTS — SPS-2 EXPERIMENT

2.2.1 Objective
The Specific Pavement Study-2 (SPS-2) experiment, entitled Strategic Study of

Structural F actors for Rigid Pavements, was designed as a controlled field experiment to
study the relative influence and long-term effectiveness of design features, in different
site conditions, on the performance of doweled jointed plain concrete pavements (JPCP)
with uniform transverse joint (perpendicular) spacing (Hanna et al. 1994; SHRP 1986;
SHRP 1990). As the test sections in the experiment are monitored since inception, the
experiment provides an opportunity to infer, more precisely, the influence of the

experimental factors on pavement performance.

2.2.2 Experiment Design
The experimental design plan for the SPS-2 experiment was developed by SHRP

in cooperation with state highway agencies. Table 2-1 presents the experimental design
for the SPS-2 experiment. The experimental factors can be grouped into structural factors
that relate to the base and concrete slab, and site factors that relate to climate and

subgrade soil type.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VNN VNN VNN VNN VNN VNN VNN VNN mV .
N_N N_N N_N N_N N_N N_N N_N N_N 5m NO EN
:N :N :N :N :N :N :N :N mV .m
MNN mNN mNN mNN MNN MNN mNN mNN Em w 955 we.»
OMN O~N O_N ON ON ON ON O_N mV .O
NNN NNN NNN NNN NNN NNN NNN NNN 5m N m ON
~NN flNN ON ON ON _NN ~NN _NN mV .
OON OON OON OON OON OON OON OON Em w m
ONN ONN ONN ONN ONN ONN ONN ONN mV .
ON ON ON ON ON ON ON ON Nam N O
ON ON ON ON ON ON ON ON mV . EN
EN EN EN EN EN EN EN EN Em w m O04 oz
OON OON OON OON OON OON OON OON mV .
EN EN EN MEN 2N w_N EN EN 5m NO ON
EN EN :N :N SN EN EN SN mV .m
ON ON ON ON ON ON ON ON Nam w
EN EN EN EN EN EN EN EN mV .
VON VON VON VON VON VON VON VON 5m N O
ON ON ON ON ON «ON ON ON mV w.m EN
EN EN EN EN EN EN EN EN Em m<00 oz
NON NON NON NON NON NON NON NON mV .
VON VON EN EN EN EN EN VHN Em NO mON
3N EN EN EN EN EN EN 2N mV w.m
ON ON ON ON ON ON ON _ON 5m
39:
8:80 0:5 3.300 psi 3580 3E 3.800 05m c5 fiwcobm GE: 2;
52>? EBB—m $05.25. P owafifim
ouooiéz 632m onoocméz 032w nflm EVE 8mm
in 33 8a
09$ =8 03 _ paw .8250 822:5 EoEo>mm

 

 

Ea: cyan sensuaxm 3mm 2 -N 033

 

The structural (design) factors included in the experiment are:
0 drainage -presence or lack of in-pavement drainage,
0 base type -dense-graded aggregate base (DGAB), lean concrete base (LCB), and
permeable asphalt-treated base (PATB),
0 PCC slab thickness — 203 mm (8 in) and 279 mm (11 in),
0 PCC flexural strength- 3.8 MPa (550 psi) and 6.2 MPa (900 psi), at l4-day, and

0 lane (slab) width— 3.66 m (12 ft) and 4.27 m (14 ft).

 

The site factors included in the experiment are:
0 subgrade soil type (fine-grained and coarse-grained),
0 climate (wet freeze, wet no-freeze, dry freeze and dry no-freeze), and

0 traffic (considered as a covariate).

At each site, 6 sections have a target PCC slab thickness of 203 mm (8 in) and the
remaining 6 have a target PCC slab thickness of 279 mm (11 in). The 76 mm (3-in)
difference in thickness was planned to “demonstrate the effect of surface thickness and its
interaction with other factors on performance” (SHRP 1990). The other factors with two
levels (PCC flexural strength and slab width) have 6 test sections corresponding to each
level. Also 4 test sections have dense-graded aggregate base (DGAB) of 153 mm (6 in),
4 sections have lean concrete base (LCB) of 153 mm (6 in), and the other 4 sections have
permeable asphalt treated base (PATB) of 102 mm (4 in) placed over a DGAB of 102
mm (4 in). In-pavement drainage is provided only in sections with PATB.

Certain design features are controlled in the experiment and are common to all SPS-2

sections. These features include (SHRP 1986):

o The monitored part of a test section is 152.4 m (500 ft) long excluding transition
zones of at least 15.2 m (50 ft) each on both sides of the monitoring length, for
material sampling and other destructive testing.

0 A uniform perpendicular joint spacing of 4.6 m (15 ft).

0 All the sections with 203 mm (8 in) as the target PCC slab thickness are built with
dowel bars of 32 mm (1.25 in) diameter. The sections with the target PCC slab
thickness of 279 mm (11 in) are built with dowel bars of 38 mm (1.5 in) diameter.
Also, all the dowels are 457 mm (18 in) long and placed at slab mid-depth with a
center-to-center spacing of 305 mm (1 ft).

0 The HMA or PCC shoulders are not tied to the mainline pavement of the test
sections.

0 Longitudinal joints are tied using 762 mm (30 in) long, No. 5 epoxy-coated
deformed steel bars of grade 40 steel and spaced 762 mm (30 in) center-to-center.

0 All structural repairs are performed on the test sections before opening to traffic.

In addition, all joint sealing is completed prior to opening to traffic.

A SPS-2 test site must have a minimum estimated traffic loading of 200,000 rigid
ESAL per year in the design lane. Traffic will thus vary from site to site and will
therefore be treated as a covariant in the study.

Based on the average annual precipitation and the average annual Freezing Index, the
sites in the experiment have been categorized into different climatic zones using the
thresholds defined in LTPP. While 508 mm (20 in) is the precipitation threshold for wet
and dry climates, 83.3 °C-day (150 °F-day) of freezing index is the threshold for freeze

and no-freeze climates.

In the experiment, the 12 sections at a given site are represented by either XX-
0201 through XX-0212, or XX-0213 through XX—0224, where XX denotes the site code
(abbreviation). The number 02 indicates the SPS experiment number and the last two
digits represent the sequential numbering of the sections.

The SPS-2 experiment requires construction of multiple test sections with similar
details and/ or materials at several sites. Construction uniformity was thus considered
important for the success of the experiment, and guidelines were developed to help
participating highway agencies. Adherence to the criteria ensures that any difference in
performance between test sections constructed with similar experimental parameters at
different locations is mainly due to the site conditions and their interactions with design
factors. Details of the guidelines are presented elsewhere (Hanna et al. 1994).

With respect to monitoring of the test sections for collection of various data (e. g.
inventory, materials and testing, traffic, distress, profile, deflection, climate, maintenance,

etc.), guidelines were developed for the experiment and details are available elsewhere.

2.2.3 Current Status of the Experiment

A total of 14 sites with 167 (NV site has 11 sections) test sections are included in
the experiment according to the data (Release 17 of DataPave). The geographical

distribution of the sites within the SPS-2 experiment is presented in Figure 2-1.

 

 

 

Figure 2- 1 Geographical distribution of the SPS-2 sites

For the seven factors in the experiment if a full-factorial design were
implemented, the experiment will have 384 sections, i.e. 48 designs in each of the 8 soil-
climate combinations. However, in the existing fractional-factorial design (see Table 2-1)
there are 192 factor-level combinations comprising of 8 site-related (subgrade soil type
and climate) combinations and 24 pavement structure combinations (design factors). The
experiment was developed such that 12 sections are built, with only half of the possible
combinations of design factors, at each of the sites.

Originally, it was planned that “48 test sections representing all structural factor
and subgrade type combinations in the basic experiment are to be constructed in each of
the four climatic zones, with 24 test sections to be constructed on fine-grained subgrade
and 24 test sections to be constructed on coarse-grained subgrade” (SHRP 1990).
Moreover, for each climatic zone and soil type combination, 12 sections are to be

constructed at one site and the other 12 sections at the other site. The experiment thus

permits the construction of 12 test sections (0201 through 0212 or 0213 through 0224) at
one site with the complementary 12 test sections to be constructed at another site within
the same climatic region and on a similar subgrade soil type.

The current status of the design factorial is shown in Table 2-2. There are six cells
within the table that are missing from the original factorial. Though the experiment was
designed to have 4 sites in each climatic zone, there are only 2 sites each in Wet No-

Freeze and Dry No—Freeze climatic zones, and 3 sites in the Dry Freeze climatic zone. A

majority of sites (7 of 14) were constructed in the Wet Freeze zone.

Table 2- 2 Status of the SPS-2 design factorial

 

 

 

 

 

 

 

 

 

Subgrade . Wet Dry
Type Desrgns Freeze Non Freeze Freeze Non Freeze Total
KS (20) WA (53)
. 0201-0212 OH (39) NC (37) NV (32).
Fine-
grained MI (26) . 9
0213-0224 IA (19) CO (8)
ND (38)
Coarse- 0201-0212 DE (10) CA (6) 5
_grained 02130224 WI (55) AR (5)‘ AZ (4)
Total 7 2 3 2 14

 

 

 

 

 

 

 

 

Note: ' Two sections in NV and five sections in C0 are coarse-grained while two sections in AR are fine-
grained.

The experiment design also called for half of the sites to be constructed on coarse-
grained soil and the other half to be constructed on fine-grained soil. In addition to this, it
was required that all the sections within a site be constructed on the same type of soil
(coarse or fine). Of the 14 sites, 5 sites were constructed on coarse-grained subgrade soil
(see Table 2-2). In 3 of the 4 climatic zones the number of sites constructed on fine-

grained and coarse-grained soils is not the same.

10

2.2.4 Monitoring Requirements

A summary of data collection guidelines is presented in Table 2-3 (Chatti et al,

2005). Monitoring of sections is to be continued until the LTPP program concludes,

application of rehabilitation construction event, or test section goes out-of—study.

Table 2- 3 Monitoring Requirements

 

Post construction

Long-term MonitorinLFrequency

 

 

 

Data type monitoring In effect before 10/1/99 In effect after 10/1/99
Longitudinal <6 months is Brennrally, but may be
- postponed up to one Annually
Profile permitted year
Deflection <6 months rs Brennrally and Brennrally and

 

 

 

 

 

permitted responsive responsive
Biennially, but may be
Manual distress <3 months postponed up to one Annually
year

 

ll

 

CHAPTER 3- LITERATURE REVIEW

This chapter summarizes the existing understanding of the effects of design features on
pavement performance, based on a literature review that focused on publications (reports

and papers) that documented research relevant to the objectives of the SPS-2 experiment.

3.1 SIGNIFICANCE OF DESIGN FACTORS

The process of designing pavements is more than just the determination of
thickness. Design factors such as base type, drainability, joint spacing, and edge support
conditions play an important role in long-term performance (Smith et al. 1989). A
“complete JPCP design” consists of the following main design features (Gharaibeh and
Darter 2001):

0 Portland cement concrete (PCC) slab,

0 Transverse joints,

0 Longitudinal joints,

0 Base course,

0 Subbase course (if any),

0 Subdrainage (edge drains, permeable layer),
0 Shoulders and curb and gutters, and

0 Subgrade treatment.

The performance of a pavement system depends on the interactions between these
features. Each of these design features is defined by several characteristics that also affect
pavement performance. Though each design feature affects the pavement performance
(and in turn costs) in a different manner, the design features are “often part of the agency

“standards”, which remain unchanged over years” (Gharaibeh and Darter 2001). Many

12

 

forms of functional and structural distress in pavements are direct consequences of
problems in design (Armaghani 1993).

A comprehensive understanding of the effects (main and interaction) of design
and site factors can help engineers design the most cost-effective combination of the
features for given site conditions. The impact of drainage alone can extend the service
life of pavements by 10 years (50 percent) and save about 41% of the costs, not including
user and maintenance costs, according to a study by Forsyth et al (Forsyth et al. 1987).
Provision of doweled joints can increase the life by 55% (reduced joint faulting, spalling,
cracking) according to another study by Smith et al (Smith et a1. 1990). As the magnitude
of the effects vary depending on the site conditions (traffic, subgrade soil type and
climate), a thorough project-level life cycle cost analysis will give a better estimate of the

effects.

3.2 FINDINGS FROM PREVIOUS STUDIES ON SPS-2

Important findings from previous studies on the SPS-2 experiment that are relevant to
this study are presented here.

Desaraju studied the influence of design and construction features on the response
and performance of the SPS-2 test sections (Desaraju 2003). It was found that pavements
with undrained, DGAB or LCB performed poorly than those with PATB and drainage.
Sections with thinner PCC slabs and standard slab width showed higher transverse
cracking and pumping. It was considered “too early” to comment on the occurrence of
faulting because of “insignificant magnitudes of faulting in the test sections”.

Jiang and Darter conducted a “first comprehensive review and evaluation” of the SPS-2
experiment and identified early trends in the performance of the test sections; the details

of the study are presented elsewhere (J iang and Darter 2001). Based on the study,

13

 

positive effects of slab thickness on cracking (transverse and longitudinal) were observed.
Thinner widened slabs were found to have highest levels of cracking. Sections with
PATB showed lowest levels of cracking, faulting and roughness, while sections with
LCB exhibited the highest levels of cracking and sections with DGAB show evidence of
higher faulting. Widened slabs showed a positive effect on faulting. It was also found that
sections constructed on coarse-grained soil were smoother than those built on fine-
grained soil.

Perera and Kohn studied the effects of factors affecting pavement smoothness
based on LTPP data from various experiments including SPS-2 (Perera and Kohn 2001).
The findings related to the SPS-2 experiments are presented here. While the effect of
PCC slab thickness was found to be insignificant, base type was found to have an effect
on early-age roughness. Highest early-age roughness was observed on sections with
LCB. The design with thin slab, normal strength concrete and DGAB was found to be
more susceptible to changes in curvature than the other designs.

The effects of subsurface drainage on pavement performance were studied using
data from SPS-2, by Hall and Correa (Hall and Correa 2003). It was found that un-
drained pavement sections built on DGAB or on LCB may develop roughness, transverse
and longitudinal cracking more rapidly than drained sections built on PATB. The quality
of drainage in the test sections was evaluated by the researchers as a part of the project;
however, the effect of quality of drainage was found only on cracking.

Selezneva et al evaluated the faulting data in LTPP and developed “indices and related
statistical parameters” (Selezneva et al. 2000). A significant positive effect of widened

slabs was found. Also use of drainage was found to improve faulting performance.

14

Pavements with stabilized bases were observed to have significantly lower levels of
faulting compared to those with untreated bases. However, the effect of dowels was
found to overshadow the effects of design features on faulting.

Vongchusiri et al performed a site-level evaluation of the Michigan SPS-2 test
sections (Vongchusiri et al. 2003), while Suthahar et al performed a site-level evaluation
of Colorado SPS-2 test sections (Suthahar et al. 2000). Based on the analysis of Michigan
test sections, it was found that the undrained sections have higher roughness compared to
those with drainage. Early age cracking in some Michigan sections with LCB was related
to shrinkage cracking in LCB. At the time of analysis, the Colorado test sections showed
comparable performance, however based on analysis of the deflection data, use of a

widened slab was “highly recommended”.

3.3 EFFECTS OF EXPERIMENTAL FACTORS ON RIGID PAVEMENT
PERFORMANCE

Findings from studies that dealt with SPS-2 data were presented above. Based on
a review of the literature that is relevant to the objectives of the SPS-2 experiment, the
understanding of the effects of these factors was studied and is presented below.
3.3.1 Effect of structural factors
In a study that was based on analysis of LTPP data from concrete pavement
sections (JPCP), it was found that key design and construction features significantly
affect JPCP performance (Khazanovich et al, 1998). The “general order of impact on
performance” of the design factors (included in SPS-2) is as follows:
1. Subdrainage,
2. Base type and modulus,

3. Slab widening,

15

4. Slab thickness, and

5. Slab modulus/strength.
The effects of each of these factors, based on the literature review, are summarized
below:
Effect of drainage: It is common practice in many states to incorporate positive
subsurface drainage into new pavements (Forsyth et al. 1987; NCHRP 2002). In order to
estimate the cost-benefit aspects of the provision of drainage, an understanding of the
precise effects of drainage on long-term pavement performance is required.

In the SPS-2 experiment, drainage is provided to sections with PATB. In
comparison to sections with PATB are the sections with DGAB, to study the effect of
drainage. The type of drainage system planned for SPS-2 is considered the most positive
drainage system (Forsyth et al. 1987; Hagen and Cochran 1996) with a minimum
laboratory permeability value typically around 300 m/day (Mallela et al. 2000). It can
also be cost—effective in addition to being beneficial to pavement performance according
to research conducted under the project NCHRP 1-34 (NCHRP 2002).

Drainage is considered as one of the important factors that can reduce wide
variability in the overall performance of concrete pavements (Ring HI 1977). In
undrained pavements, all the layers will be subjected to flooding for a minimum of a few
weeks in arid areas and many months in wet climates. Water trapped in a pavement is a
major contributor to poor performance as the action of loads on an inadequately drained
pavement can reduce subgrade support causing pumping, and also redistribute the
subgrade and subbase materials causing faulting (Cedergren 1988; Ring HI 1977). In

well-drained designs all layers above the drainage layer will be moist, not fully saturated,

l6

most of the time thereby decreasing the chances of rapid deterioration due to traffic
(Cedergren 1988). It was found through field experiments that about 40% of rainfall gets
into the concrete pavement and also that spring thaw flows are roughly equivalent to a
major rain event (Hagen and Cochran 1996). The action of in-pavement drainage can
reduce the saturation time under the concrete by a factor of 10 to 100 (Ray 1981b).

It is believed that in many cases one of the best ways to improve performance at a
relatively low additional cost is to provide a drainage system (Cedergren 1988;
Fleckenstein and Allen 1996; Mallela et al. 2000; Ray 1981b). Reduction in pumping,
faulting, cracking, and D-cracking was observed as an effect of drainage in various
studies (Fleckenstein and Allen 1996; Forsyth et al. 1987; Harvey et al. 2000; NCHRP
2002; Peshkin et al. 1989; Smith et al. 1997; Tart Jr. 2000). In addition to having lesser
distress, pavements with edge drains have significantly lesser growth in roughness
(Khazanovich et al. 1998) and higher subgrade strength (Fleckenstein and Allen 1996).
Though drainage systems are known to be beneficial to concrete pavements, drainage is
not a substitute for poor pavement design and for a drainage system to be effective other
design features that help minimize moisture intrusion should be present (Mallela et al.
2000).

Effect of base type: Slabs on different types of bases will have different mechanistic
response under same load (Wu and Qian 1997). The slab-base interface friction,
erodibility, stiffness, and drainability of a base are the main characteristics that influence
the performance of concrete pavements (Smith et al. 1989). While the erodibility of a
base course determines the amount of support available to the slab during wet conditions,

stiffness of the base layer influences stresses and deflections of the slab. Drainability of a

17

pavement determines the amount of time for which water is present at the slab-base
interface and thus a highly drainable base would result in better performance compared to
an undrained one. In light of the wide variation in the types of bases, base type is known
to be an important factor in determining concrete pavement performance (Khazanovich et
al. 1998; Smith et a1. 1997; Wade et al. 1995; Wu 1992), especially in terms of cracking,
faulting and spalling (Smith et al. 1989).

In the SPS-2 experiment, three types of base types are included— DGAB, LCB
and PATB. Table 3-1 is a summary of relative advantages and disadvantages of these
base types, along with design considerations. In comparison with stabilized bases, DGAB
is known to be more erodible and to have lesser stiffness. It is undesirable according to
some experts to design untreated bases, unless it is of high quality and is equipped with
drainage, for pavements with heavy traffic, especially in severe climate. In View of less
erodibility and greater stiffness of stabilized bases, they have been in use for pavements
with heavy traffic (Ray 19813; Ray 1981b; Wade et al. 1995). When a good drainage
system is provided, the stabilized bases are more beneficial. However, stabilized bases
can cause uncontrolled early cracking because of higher slab-base friction and bonding at
the interface (Voi gt 2000). Uncontrolled cracking during early age is often the cause for
poor performance of concrete pavements (Forster 2000). LCB is known to have early
occurrence of cracks owing to shrinkage and some analytical designs have thus suggested
a two stage approach for its design— the first to deal with “primary cracking” and the
second to tackle “secondary cracking” (Brown 1979).

As a result of a major national field study it was found that permeable bases are

“very good”, DGAB is “good” and LCB is a “poor” base (Smith et al. 1989). While LCB

18

was associated with “considerable” pumping, faulting and cracking, PATB sections had
virtually no faulting, cracking or spalling. DGAB was found to be “generally fair to very
good” with open-graded ones performing better than others. Poor drainage of LCB was
considered to be the reason for its poor performance. Similar results were obtained in a
study on LTPP pavements (Khazanovich et al. 1998).

Large curling and warping stresses are also associated with slabs constructed on
LCB, owing to the high slab-base friction offered by LCB. In the US, reflection cracking
that may result from friction between slab and LCB is prevented by breaking the bond
(Smith et al. 1997). However, in Germany the two layers are allowed to remain bonded,
and the LCB layer is notched to coincide with joints in the slab (Springenschmid and
Fleischer 2001). This allows the transverse joints to open up uniformly while allowing for
monolithic action of the slab-base system.

In contrast to LCB, DGAB was found to contribute the least to curling and
warping in slabs. Bases with stiffness between 800 and 1100 psi, such as LCB and
econocrete, can significantly increase stresses in slabs during early ages (Armaghani
1993; Harvey et al. 2000). Separation of the slab from the LCB was also observed from a
study, which could be because the LCB does not conform to the shape of the deformed
slab (Hansen et al. 2001; Wu and Qian 1997). This action results in loss of support for the
slab and affects the slab’s response to loading.

Contrary to the findings mentioned above, some studies have indicated that PATB
may not be cost-effective while LCB can be cost-effective when doweled joints and

shorter joint spacing are designed (Gharaibeh and Darter 2001).

19

Effect of PCC slab thickness: The effect of PCC slab thickness is perhaps the best
understood among the main design features of the SPS-2 experiment. In the experiment,
two levels of PCC slab thickness are present— 203 mm (8-in) and 279 mm (1 l-in). As
mentioned before, the 76 mm (3-in) difference was planned to allow for interaction of
this factor with other factors in the experiment.

Field studies have shown that regardless of the site conditions, slab thickness
plays a significant role in pavement performance with thicker slabs showing less distress,
especially cracking (Smith et al. 1989). The effect of increasing slab thickness, from the
same study, was found to be more noticeable for thinner pavements than for thicker
pavements. However, based on a study of LTPP concrete pavements, it was found that
slab thickness did not show “much significance” in the general data analysis
(Khazanovich et al. 1998).

Though thicker slabs have more fatigue life than thinner slabs for a given traffic
loading, it is to be noted that usually pavements seldom fail in fatigue (Armaghani 1993).

Most failures are direct consequences of design and construction issues.

20

 

 

8080 00 0w8:080m
Ea as: gas:

 

 

 

 

.28: 2:0: :0 20:80 . 80:8 new 5:3 888 33 030800.:
82002 8%: QNQOSV w:0_ :00 0:08:08 880: .32 800 33 00080
00 33 \0 oewa 80 :00000800 o 5800 00 0300083 00:88:00 00 zmem -8:0Q
0.8080: 80.8: €035:
88% 3:60:82 £882 o 00 08800: 80008 83:80 E 80808 003 moon—00¢
828 a as 8:838:08 0:588
8 880m 00 wcfiogoz o 007. 8800.85 80802:: 89,088“
A8035. 83 00 w80:0m o 028 58, :0505 080805 :0800
83:80. 88. 00 w880 o A809 880:3: 00 00:88:80: w:0bm
8.30 :wi w888 3:883:00 :8: 830% 8:00:80 0308 w:0bm 0:80:00
00 83002 08% nan: 0825 o 0:00:00 NN 8008 800 swam 00:58:00 00 ammm :03
A83 888%: 2808:0800
883082080800 o :8: 030% 80800
$33,820 oSmN VN 00 V: 805 800 swam 30 8803899
:0 E05000 .93 Ohm 0580 8:088:00 £82080 888886 00:20:
-OON $8020 3:00:00 :8: 0: 028 >82 08808 8 8:0:00Q $833880:
08:88 00000030. o 08800: 5800 0:0
:88: Sb: 8:838 83:80 0: 52500088 33 083305
swam 08: 0030030200 o 88:30:80: $830805 8080200 80m
00 83002 :9?— :98008 00 :wwm0Q o 8:58:00 00 20055 80¢ :88; m0>080m 050088
20>0A 058B
000:088000m 803802800 :wmmon EM 808380.685 8038030.: 09C. 80m

 

 

 

 

 

 

:32 a so 8ng .25. 88 22:28 00: 0o 8568 E 2.5

21

Effect of PCC flexural strength: In the SPS-2 experiment, PCC flexural strength has two
levels— 3.8 MPa (550 psi) and 6.2 MPa (900 psi), at 14 days. Through the experiment,
the effect of this factor on long—term pavement performance is being studied. Though the
experiment includes flexural strength as the only PCC material property, it is to be noted
that other PCC properties such as transport properties, elastic modulus, coefficient of
thermal expansion, drying shrinkage, and entrained air void system significantly affect
pavement performance (Hansen et al. 2001). However, PCC fatigue is typically related to
flexural strength.

In concrete applications other than in pavements, the use of high strength concrete
is on the rise (Hansen et al. 1998). There is lack of sufficient information on the effects of
high strength concrete on pavement performance and pavement engineers are usually
reluctant to consider using higher than normal (28 MPa) concrete for typical pavement
applications. In case of good design and construction practices normal strength concrete
can produce excellent pavement performance, however, higher strength concrete may be
advantageous to control cracking due to loss of support (Hansen et a1. 2001; Harvey et al.
2000).

Presently, for pavement applications, concrete mixes are designed to develop 17
to 20 MPa (2500 to 3000 psi) of compressive strength at early age (1 to 3 days), for
reducing user costs by opening pavements early to traffic (Armaghani 1993; Harvey et al.
2000). Design compressive strength for high strength concrete is usually 55 to 75 MPa
(8000 to 10900 psi, approximately). Higher strength concrete is known to have the
benefits of higher resistance to physical and chemical deterioration as the permeability of

high strength concrete is lower than normal concrete (Hansen et al. 2001). From a study

22

on LTPP concrete pavements, it was found that the main distress affected by concrete
strength is transverse cracking i.e. concrete with lower strength has more cracks in
comparison to concrete with higher strength (Khazanovich et al. 1998).

This high strength is obtained by increasing the cement content in the concrete
mixture. Higher the cement content higher is the heat development in the mixture during
hydration. Field surveys have shown “strong evidence” of its negative effects, major
problem being early age (within one year) mid-slab transverse shrinkage cracking
(Armaghani 1993). Temperature control during hydration is thus an important issue for
high strength concrete to control drying shrinkage and in turn cracking. Current
specifications, however, neglect many of the early age environmental effects on
performance of PCC pavements (Grater et a1. 1999).

Effect of slab width: Displacement and deformations in the slab play a major role in
creating critical conditions for occurrence of distress (Armaghani 1993). Widened slabs
are usually used in truck lanes to significantly reduce the edge stresses and comer
deflections caused by heavy loads (Ray, 1981), and increase the fatigue life of the
pavement (Harvey et a]. 2000). The design is also cost-effective compared to standard
slab width designs according a study (NCHRP Research Results Digest, 2002). This
design has been in use in Europe since early 1970s (Ray 1981b). In the SPS-2
experiment, slab width, sometimes referred to as lane width, has two levels— 3.7 m (12-
ft) and 4.3 m (14-ft).

Widened slab pavements have shown lesser distress (pumping and comer
cracking) and better ride quality compared to pavements with standard slab width

(Armaghani 1993). Widened slab also reduces both faulting and cracking, irrespective of

23

site conditions, and increases JPCP life by 5 to 13 years (Gharaibeh and Darter 2001).
However, from an early analysis of LTPP data from SPS-Z experiment, it was found that
longitudinal cracking is more prevalent in widened slab sections in comparison with
those with conventional width, among sections with thinner slabs [203 mm (8-in)]
(Gharaibeh and Darter 2001). In contrast to this finding, based on analysis of LTPP SPS-
2 experiment in Colorado, widened lane sections are “highly recommended” for
increasing the load carrying capacity of the pavements (Suthahar et al. 2000).

Widened slab may also have an important role in reducing faulting and pumping.
In a research concerned with pumping hydrodynamics, it was found that the water
ejection rate from leave slab toward approach slab is lower by 50% for pavements with
widened slab (Ray 1981b). Also, the reduction in water ejection rate from leave slab to

shoulder is 30%.

3.3.2 Effects of site factors
As mentioned earlier, the subgrade soil type and climate are the site factors in the

SPS-Z experiment. The effects of site factors on concrete pavement performance are
known to be significant. However, as the designer cannot control these factors,
appropriate modifications should be made in the design based on an understanding of the
effects of site factors. The effects of site factors, based on the literature review, are
summarized below:

Environment is one of the three “best-known causes” of pavement damage, the
others being traffic loads and poor materials (Hudson and Flanagan 1987). All the three
causes interact to cause pavement damage; however, the extent of interaction differs

based on conditions. The most important agents of environment are moisture and

24

temperature. While moisture causes several problems including warping, base/subbase
failure, freeze-thaw effects, and swelling; temperature differentials cause curling stresses
in the slab. Temperature conditions at the time of paving influence the extent of locked-in
gradient at the time of setting of concrete (Rao et al. 2001).

Based on a study on LTPP concrete pavements, it was found that wet-freeze (WF)
climate with fine-grained soil is the “most critical” condition, while warm or dry climate
with coarse—grained soil is the “most favorable” (Khazanovich et al. 1998). From the
same study it was found that joint faulting is prevalent in WF climate while transverse
cracking is critical in dry climates where higher thermal gradients exist.

Though the effects of environment are important, it should be noted that the
environment can only exacerbate the effects due to traffic loadings, as found in a field
study by Hudson and Flanagan (Hudson and Flanagan 1987).

The effect of subgrade soil type on pavement performance is often influenced by
the climatic conditions. However, in general, fine-grained subgrade soil causes faulting
due to greater erodibility and volume changes with moisture (Khazanovich et al. 1998).
Fine-grained soils under severe climatic conditions provide non-uniform support and may
cause corner cracking and pumping.

Most of the studies discussed above are limited in scope. The studies that
analyzed SPS-Z data suffer from limited statistical rigor and/or from lack of enough data.
The other studies that were conducted to study the effects of design features on rigid
pavement performance are observational, in addition to being restricted in scope. A

controlled field experiment was required to determine precisely the main and interactive

25

effects of the strategic design and site features. The SPS-Z experiment was proposed and
constructed in light of this need.

This thesis presents results on the effects of experimental factors based on a
comprehensive analysis of data (contained in Release 17 of DataPave) from the ongoing
SPS-2 experiment. The methodology presented in the thesis would be valid for analysis

of typical in-service pavement data.

26

i‘

CHAPTER 4- DATA AVAILABILITY AND EXPERIMENT REVIEW

4.1. INTRODUCTION

This chapter presents a summary on data availability and a synopsis of a review of the
experiment. While information about data availability is useful in defining the scope of
the study, a review of the experiment with respect to the guidelines, is important in
identifying deviations, if any.

Release 17 of DataPave was the primary data source for this study. Data from
previous releases of DataPave, and other sources were used only to supplement the data
from the Release 17. The construction information for each site was obtained from the
construction reports. The essential data required for this study can be broadly classified
into following categories:

0 Site Information: site information, construction issues, climatic and traffic data.

0 Material Data: Material type and properties for various bound and un-bound
pavement layers.

0 Pavement Structure: Layer type and thickness information and other design features
such as lane width, shoulder type and dowel bar diameter etc.

0 Monitoring Data: Longitudinal and transverse profiles, distress and deflections.

The availability of data grouped under each of the above categories is presented next,

followed by a summary of a review of the experiment using the available data.

4.2. ASSESSMENT OF DATA AVAILABILITY

The variables corresponding to the SPS-2 experiment can be divided broadly into
two categories- dependent variables, and independent variables. The dependent variables

are those used to describe pavement response and performance. A majority of pavement

27

responses are surface deflections from Falling Weight Deflectometer (FWD) testing.
Measures of pavement performance are those that cumulate with time (e. g., cracking).

The independent variables are those that describe the design and construction
factors. These can be divided mainly into main, and exogenous (or confounding)
variables. Main variables are those that are controlled in the experiment, whereas the
variables that have potential impact on pavement response and performance but are not
controlled in the experiment design are exogenous variables. Exogenous variables that
are independent of the main experiment variables are traffic and age. All other exogenous
variables are associated with the main design and construction variables. These include
material properties of the various pavement layers, which constitute the structural factors
in the design matrix, and the climatic factors, which describe the four climatic zones in
the matrix.

After identification of all the relevant variables, a database was developed for the
analysis. For cases where multiple data values were available for a data element, the
values were averaged to obtain a best estimate. For example, IRI values corresponding to
a survey were averaged over several runs. The availability of relevant data elements is

discussed in 'detail in the subsequent sections.

4.2.1. General Site Information

The information that is common to all sections at a test site has been categorized
under this heading.
Construction Reports

The construction reports contain information related to pavement geometry,
construction issues (and deviations, if any), traffic, environmental conditions during

construction, and material quality control data. These reports are available for all the sites

28

 

in the experiment. Details of the construction issues are available elsewhere (Desaraju
2003; Chatti, Buch et al. 2005).
Climatic data

In the LTPP database, climatic data is available from two sources- Virtual
Weather Station (V WS) and Automated Weather Station (AWS). The climatic zones in
LTPP are defined based on two parameters- average annual precipitation and average
annual freezing index. These two variables were used to confirm the climatic
classification of each site. As data for all the sites were not available from Release 17 of
DataPave, DataPave 3.0 data were used to complete the data.
T raflic data

The traffic data available in the LTPP database are presented in three forms-
monitored (M), estimated (E) and axle distribution (A). Traffic data availability is shown
in Table 4-1. Traffic being one of the most important factors that impact pavement
performance, inconsistency in traffic data was compensated, to some extent, by
estimating traffic, called ‘proposed’ traffic (Chatti, Buch et al. 2005), for each of the sites
based on the three sources of traffic data. The ‘proposed’ traffic was used as a covariate
in all statistical analyses to adjust the dependent variable according to the variation in
traffic across sites.

Traffic opening date is the date on which traffic was allowed to pass over the
newly constructed test sections. This data is available in the database for all the sites. The
age of a section was calculated using this date and the corresponding last distress survey

date. In other words, age is defined as the latest age for which distress data are available.

29

Table 4— 1 Summary of traffic data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
4 E M,A M,A M,A E E NO NO NO NO
5 NO M,E,A A M,E,A A A A A
6 NO

8 l NO E ] M,A M,A M,A A A A NO NO
10 NO E NO NO NO NO NO
19 NO E M M,A NO NO A A A
20 E, M E,M,A E, M E E E E E E NO NO
26 M,A M,A M M M M NO A A NO
32 NO E M,A NO A A NO NO
37 M,A M,A E,M E,M,A E,M,A E,M,A A A A
38 NO E E E E E E E NO
39 NO NO M,A NO A A A
53 I NO E E,A M,A A A A NO
55 NO NO NO NO A NO

 

 

Note: ‘E’ indicates availability of estimated traffic data, ‘M’ indicates availability of monitored traffic data,
and ‘A’ indicates availability of Axle-distribution data.

Material data

Data pertaining to materials used in the construction of pavement sections have
been categorized as materials data. The data that were used for this study include the
gradation of subgrade soil (percent passing #200 sieve) and PCC (mix design information
and properties, and PCC strength data), apart from strength testing results of lean

concrete.

Subgrade

Subgrade soil data in the form of percent passing #200 sieve (available for all
sites) were used to classify subgrade soil as either “fine” or “coarse”, based on Unified
Classification system, and compare results with subgrade soil classification data that are
available from other sources in DataPave.

Lean Concrete

The compressive strength data for LCB are available for 96 % of the sections. The
7-day compressive strength was used to compare with the specified target strength of 3.5

MPa.

30

Portland Cement Concrete

All the details of PCC mix design such as cement content, aggregate content
(coarse and fine), water content, and additive type are available for all sites except WI
(55). For most of the sites which have data, two types of mixes were used, one for each of
the two levels of target 14-day flexural strength. In DE (10) more than two types of mixes
were used, and PCC mix design data are available for all the sections. Though not a part
of the experiment design, PCC compressive strength, split tensile strength, and modulus
of elasticity are also reported in the database. The mechanical properties of concrete were
recorded at 7, 14 and 365 days after casting. Compressive strength and split tensile
strength data from testing of core samples are also available. Table 4—2 is a summary of
data availability for PCC mechanical properties (except for flexural strength) from
DataPave Release 17.

Only 52% of the sections in the SPS-2 experiment have 14-day flexural strength
data. Data for sections in CA (6) and ND (3 8) are not available. The 14-day flexural
strength data were used for comparison with specified target strengths. Table 4-3 is a
summary of flexural strength data availability.

Coefficient of thermal expansion (CTE) of PCC is an important requirement for
conducting a thermal analysis. CTE data are unavailable in DataPave of Release 17. Data
were obtained from Portland Cement Concrete Pavements, FHWA. However, CTE data
was available only for 16 sections, precluding any rigorous analysis based on the data.

Table 4-4 below is a summary of CTE data obtained from FHWA.

31

Table 4— 2 Summary of data availability (percent of sections) for PCC properties

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Compressive Strength Tensile Strength Elastic
Srte ID
Core Fresh Core Fresh MOdUhIS
4 83 50 92 50 92
5 0 0 0 0 25
6 25 50 50 0 100
8 92 100 92 100 100
10 50 50 67 25 42
19 83 50 100 50 100
20 0 92 0 83 0
26 42 42 42 42 50
32 92 50 92 50 92
37 0 50 75 0 100
38 25 0 25 0 100
39 92 50 75 42 83
53 92 58 100 58 100
55 0 0 0 0 100

 

 

 

 

 

 

Table 4- 3 Summary of availability (percent of sections) PCC flexural strength data

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site ID % of sections with data
l4-day 28-day 365-day
4 75 75 75
5 58 58 58
6 0 0 0
8 100 100 92
10 50 50 50
19 50 50 50
20 100 100 92
26 42 42 50
32 50 50 50
37 25 33 42
38 0 42 92
39 50 50 50
53 58 58 58
55 58 58 O

 

 

 

 

32

 

 

Table 4- 4 CTE data obtained from FHWA

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site ID Aggegate Type SHRP ID CTE, in/in/°C
5 - 0215 10.2
5 - 0220 11.3
10 Diorite 0205 1 1.6
10 Diorite 0208 9.2
10 Diorite 021 1 9.5
19 Limestone 0224 9.6
20 Limestone 0207 10
20 Limestone 0208 10.65
32 - 0203 10.9
32 - 0208 13.9
32 - 0209 11.1
37 Granite 0203 8.9
37 Granite 0204 11.9
39 Limestone 0204 10.2
55 - 0222 8.8
55 - 0223 9.8

 

Pavement Structure data

All data that relates to the structure (cross-section) of the pavement sections have
been categorized as pavement structure data. This data has been used to compare as-
designed thickness with as-built thickness, for the sections. Information about layer type
and thickness is available for all test sections in the experiment. Information about the
size and spacing of dowel bars is available for all the sections except for 4 sections (0205

through 0208) in WA (53). Though not a part of the experiment design, details about the

shoulders have been obtained.

Monitoring data

All data that are collected during distress surveys and during FWD testing has
been categorized as monitoring data. Longitudinal profile data, distress data, faulting
data, and deflection data fall under this category. These data are available for all the

sections in the experiment. Table 4-5 is the summary of data availability (from Release

17 of DataPave) for all the types of data listed above.

33

 

Table 4- 5 Summary of data availability for SPS-2 experiment

 

 

 

 

 

 

 

 

 

 

Data availability, % of
Data category Data type sections
Construction reports 100
Climatic data
Virtual Weather Station
Annual Temperature 93
Annual Precipitation 93
Automated Weather Station
Site information Monthly Temperature 93
Monthly precipitation 93
Traflic data“
Traffic Open date 100
Monitored 65
Estimated 71
Axle Distribution data 78
Subgrade
Sieve analysis 53
Classification 1 00
Backcalculated moduli 0
Lean Concrete Base
Compressive Strength 96
Materials data Portland Cement Concrete
PCC mix data 100
14-day Flexural Strength 52
Compressive Strength 92
Split tensile Strength 91
Static modulus of Elasticity 78
CTE 0
Unit weight 63
Layer details
Type 100
Representative thickness 100
Dowel bar details
Diameter 98
Pavement structure Length 98
Spacing 98
Shoulder information
Type 93
Width 93
Thickness 93
Profile data (IRI) 100
Distress data 100
. . " Faulting data 100
Monrtonng FWD data
Deflection 100
Temperature during testing 100

 

Note:

* Monitored, Estimated, or Axle Distribution data is considered to be available for a site even if the data
is available only for one year.

** Data is said to be available for a section even if it is available for just one year.

34

4.3. EXPERIMENT REVIEW

As mentioned earlier, guidelines were developed by LTPP for controlling
variability in the experiment so as to allow for manifestation of “pure” effects. If some
deviations occur, it is important that those are addressed in the analysis. So as to identify
deviations, if any, in the SPS-2 experiment, a review of the experiment was conducted
using the available data. The results from this review are presented here. A brief
discussion on the construction issues (from construction reports) will be followed by a

discussion on the deviations in the design and site features.

4.3.1. Construction Deviations based on Construction Reports

Information regarding construction issues was obtained from the construction reports.
A design versus construction review for each site can be found in the appendix. Some of
the major issues in the SPS-2 experiment are below:

0 Shrinkage cracks in LCB were observed soon after construction at sites in AZ (4),
CA (6), DE (10), MI (26), NV (32), NC (37), ND (38), and WA (53).

o PCC mixes that were different from what was stipulated were used at sites in DE
(10) and NV (32). In DE, all the 550-psi PCC has been replaced with the
Delaware DOT Type ‘B’ mix that gives a flexural strength of approximately 650
psi. Some SHRP 900—psi concrete was also replaced (sections 0202 and 0206),
with 900-psi mix with 7.5-bag mix, after breaking and removing cracked PCC. In
NV, The sections that were supposed to have PCC of 550-psi 14-day flexural
strength have a 475-psi mix and the sections with target 14-day flexural strength
of 900—psi have a 750-psi mix.

0 Construction delays occurred due to bad weather at sites in MI (26) and ND (3 8).

35

o Improper size dowel bars were used at sites in CA (6) and NC (3 7) sites. At the
site in CA (6), 32 mm and 38 mm diameter bars were used in both thinner and
thicker slab sections. At the site in NC (37), all the sections were constructed with
38 mm —diameter dowel.

0 Underground structures were present at sites in IA (19) and KS (20) within the
monitoring length of the sections.

0 Repairing (such as Partial depth repairs, full depth repairs, crack sealing, and
shoulder restoration) was done to some sections [20-0201, 32-0201, and all the

sections at the sites in AR (5) and ND (3 8)] after opening the sections to traffic.

4.3.2. Design versus Construction Review based on Data

The experiment review using the available data was done for those factors in the
experiment that are bound by the guidelines. These features are subgrade soil type and
climatic zone, which are site factors, and, layer thickness and flexural strength, which are
design factors. In addition, a review of the status of the experiment in terms of traffic and
age was performed.

Site Factors

Subgrade type: In AR (5), sections 0222 and 0223 were constructed on fine-
grained soils while all the other 10 sections were constructed on coarse-grained soils.
Similarly, in NV (32), sections 0201 and 0205 were built on coarse-grained soils whereas
the other 9 sections were built on fme-grained soils. At the CO (8) site, 5 sections were
constructed on coarse-grained soils while the other 7 were constructed on fine-grained

soils. At other sites, all the sections were constructed on same type of subgrade soil.

36

 

Climatic zone: The climatic data (VWS data) were used to categorize sites into 4
climatic zones according to the LTPP definitions for climatic zones. Figure 4-1 is a
scatter plot showing all the sites in the experiment and LTPP criteria (reference lines at

508 mm and 83.3 °C-day) regarding climatic zones.

 

 

 

 

 

 

1400- 05:
5. 37:
8 1200— M010
.3 E
a 1000- 5 39 19
§ 20 26 a 55
E 800- : a a
3 WNF WF
3 600-1 E 38
a i ---------------------------------------------- °
< a 5 0
200.. G4 E 032
I l l l l I l l
0 200 400 600 800 1000 1200 1400

Average Annual Freezing Index, C-day

Figure 4—1 Distribution of SPS-2 sites with respect to climate
Though the SPS-2 sites were found appropriately classified, it is evident from the figure

that some sites [AR (5), DE (10), NC (37), and ND (38)] are on the fringes of the climatic
zones. Thus the effects of the climatic zones, as defined by LTPP, may not be easily
discernible because wide variation exists within each zone.
Design Factors

PCC slab and DGAB thickness for the test sections were compared with their
respective target thickness. Table 4-5 is a summary of a comparison between as-designed

and as-constructed thickness, for PCC layer and the base layers.

37

 

PCC thickness: The experiment has two levels of PCC thickness- 203 mm and
279 mm. The allowable deviation from the target PCC thickness according to the
guidelines is 6.4 mm (0.25 in) (SHRP, 1986). Among sections with target thickness of
203 mm, only 28 sections (33 %) conform to the allowable deviation of 6.4 mm. The
remainder of the sections (67%) was built either thicker or thinner by more than 6.4 mm.
Figure 4-2 is the cumulative frequency graph of PCC thickness. Among sections with a
target thickness of 279 mm, 44 sections (53 %) conform to the allowable deviation of 6.4
mm. Figure 4-3 is the cumulative frequency graph showing the percent of sections and
number of sections below the corresponding thickness values.

Base thickness: Though there are no guidelines limiting deviation from design
thickness for LCB and PATB, the allowable deviation from target elevation for DGAB is
12.7 mm (0.5 in). Figure 4-4 is a cumulative frequency distribution of actual base
thickness of DGAB sections (target thickness of 152 mm). About 80% of the sections
built on DGAB have thickness that falls within the allowable range (see Table 4-6), as

defined by the guidelines.

38

No. of sections

No. of sections

 

100%
r 90%
r: 80%
.- 70%
60%
r 50%
40%
a 30%
~ 20%
' 10%
~ 0%

Percent sections below

 

 

 

<= 180 180-197 197-209 209-220 220-240 240-260

Thickness range, m
- No. of sections -<>- Percent sections below

Figure 4- 2 Actual thickness of sections with target thickness of 203 mm

 

 

 

50 T! ‘4 100%
45 T 4— 90%
T 4— 80%
35 ~4~ 70%
30 ; -— 60%
25 j -~ 50%
20 -- ~- 40%
15 Jr -- 30%
10 1 —~ 20%
5 - -~ 10%
0 i— +-—l—-— 0%

<273 273-285 285-300 300-315
Thickness range, m

- No. of sections -<>- Percent sections below

Figure 4- 3 Actual thickness of sections with target thickness of 279 mm

39

Percent sections below

 

45 . 100%
40 .- i 90%
35 + 80% 3
+ 70% 2
m 30 «A o
G ..D
8 i 60% E
8 25 “‘ 500/ -s
T o
“5. 20 i .L 400/ g
2° 15 l o 5
30% E,
10 - 20%
5 " 10%
0 _ 0%

 

 

<=140 140—165 165-190 190-215 215-240 240-260
Basethickness,mm

 

- Percent sections below -<>- No. of sections

Figure 4- 4 Actual base thickness of sections on DGAB
Table 4— 6 Summary of deviation in thickness from design

 

 

 

 

 

 

 

Coefficient Below Within Above
Layer Target Count, Mean, 5‘3?de of allowable allowable allowable
Thickness, no. of Devratron, .
Type . mm Variance, range, % range, % range, %
mm sections mm o/ . . .
0 sections sections sections
PCC 203 84 212 12 5 7 33 60
279 83 286 9 3 4 53 43
DGAB 152 56 163 34 21 4 80 16
LCB 152 56 160 10 6 - - -
PATB 102 55 101 14 14 - - -
DGAB
below 102 55 114 36 32 - - -
PATB

 

 

 

 

 

 

 

 

 

 

 

PCC Flexural Strength: At each site, 6 sections have a target 14-day flexural
strength of 3.8 MPa (500 psi) and the other 6 sections have target l4-day flexural strength
of 6.2 MPa (900 psi). Comparisons between the measured flexural strength and target 14-
day strength were made. Figure 4—5 through Figure 4-10 are cumulative histograms for
flexural strength at 14, 28 and 365 days.

It is evident from the plots that at 365 days most of the sections that failed to

reach the target at 14 days have reached their target strengths. Among the sections with

40

target PCC l4-day flexural strength of 3.8 MPa, 7 sections, of the 44 sections for which
data are available, failed to meet the criterion of 3.4 MPa at 14-days. At 28-days, 1 of the
sections failed to meet the criterion and at 365 days all the sections have met the
criterion. Among sections with target 14-day PCC flexural strength of 6.2 MPa flexural
strength data are available for 42 sections. Of these sections, 16 sections failed to meet
the target of 5.6 MPa at 14-days. Eight sections met the criterion at 28 days and 5 more
met the criterion of 5.4 MPa at 365 days.

Based on the flexural strength data available in Release 17 of DataPave, it thus

 

seems that most of the sections met target strength not at 14-days but later. As it is safe to
assume that the long-term performance is more affected by final strengths, these
deviations may not affect the experiment significantly. However, if deviation from target
strength were available for all sections, considering flexural strength deviation as a

covariate in statistical analyses will be an option.

 

   
  
 

 

 

14 l -- 90%
m 12 i ~» 80%
t: _ m
g 10 ~— 4 70% g
g " 60% §
“:3 8 7 - 50% “a
B 64 - 40% 5
§ - _ E
z 4 T 30% A
r 20%
2T — 10%
0 - ~ 0%
00 v-1 V 00 —~ In 00 N In
N' "5 "5 "5 ...- sr 3* v5 V9
V 9°. ~. V. 0°. — w, 99 N
N m M M V V V In
l4—day Flexural strength, MPa

- Number of sections -<>- Percent sections below

Figure 4- 5 14—day flexural strength of sections with target strength of 3.8 MPa

41

Number of sections
Percent of sections

 

 

 

2881
3184
3438
3841
4145
4548
4862
5265
55-59
5962
6266

28-day Flexural strength, MPa
- Number of sections '0' Percent sections below

Figure 4— 6 28-day flexural strength of sections with target strength of 3.8 MPa

 

            

 

 

 

 

8

—1m%
14+ ’ 1~m%
12- ’ ° J~m%

.310“ . ° +~m% E
8 ~~m% E
E 8* . -~w% g
g 6“ --w% E
E 4T . 1-w% :
wzmt
2““ o‘l Ill d-m%
0- .Al .m6

3841
4145
4548
5265
5569
5962

‘0.
"5"
(‘5
o

6.6-6.9

3.4-3
4.8-5.2

365-day Flexural strength, MPa
- Number of sections '0- Percent sections below

Figure 4- 7 365-day flexural strength of sections with target strength of 3.8 MPa

42

Number of sections

Number of sections

 

 

 

 

O
0
00 ~ v 09 -— an 00 “i try as N sq as
N "D "a t? '7 ‘T 5r "9 "P "9 ‘i’ ‘9 ‘9
V w ._. q: oo _ in. 09 or m as or o
N ("1 M M V <1” V V) V) V) \O \O

14—day Flexural Strength, MPa
- Number of sections -0- Percent sections below

Figure 4- 8 14-day flexural strength of sections with target strength of 6.2 MPa

 

 

 

 

O
0
0° N V2 °~ N. *0. or m ~° <3. M e O.
"9 ‘1’ ‘1' ‘1 V? "P "P ‘9 “P '7 ‘T ‘T °.°
'0 09 N. w or N. ~52 0 M <2 G M N
M M V V V V3 V3 '1) \O \O I" I\ l\

28-day Flexural Strength range, MPa
- Number of sections -<>- Percent sections below

Figure 4- 9 28-day flexural strength of sections with target strength of 6.2 MPa

43

Percent of sections

Percent of sections

 

 

 

 

14 — 100%
12 -~ l 90%
m ~ 80%
§ 10 * ~ 70% g
‘3 8 ~ ‘i 60% 8
E ~ 50% ‘8
.8 6 " ~ 40% 8
5 e
2 4 - 30% 3
—~ 20%
2 _.
-r 10%
0 — o . L 0%
vs. as N so as N so O. m N ca
'7 5r '9 V9 '9 *9 e '7 '7 l7 °9
-—: w 99 N w as. N so. o «2 IN.
V- st- <3- In In In so so P- rx l\

365-day F lexural Strength range, MPa
- Number of sections -<>- Percent sections below

Figure 4- 10 365-day flexural strength of sections with target strength of 6.2 MPa

Miscellaneous features

Dowel diameter: It was stipulated in the guidelines (SHRP, 1986) that all the sections
with 203 mm target PCC slab thickness have 32 mm diameter dowels while the sections
with 279 mm have 38 mm diameter dowels. Improper size dowel bars were used at CA
and NC sites. At the CA site, 32 mm and 38 mm diameter bars were used in both thinner
and thicker slab sections. At the NC site, all the sections were constructed with dowels of
diameter 38 mm. At all other sites no deviation was observed.

In addition to the factors discussed above, the status of the experiment in terms of
age of the sites and the traffic at the sites was reviewed. Figure 4-11 shows the
distribution of age among the SPS-2 sites. As mentioned before, age was calculated using
the traffic open date and the latest distress survey. Similarly, Figure 4-12 shows traffic
variation among the SPS-2 sites. From both the figures it is evident that there is

considerable variation in age and traffic across sites.

 

20

Age, years

 

‘3 ltdulhltm

 

AZARCACODEIAKSMINVNCNDOHWAWI

KESAL per year
~— N a b)
8 § 8 §

5

Site
Figure 4- 11 Age distribution in the SPS-2 sites

 

M
8
I I Y I

 

C
111

AZ(4)

 

AAAAAAAAAAAAA

398828888886:

m" "t 0 :11":

UUQSQEEZEOEB
Site

Figure 4- 12 Variation in traffic across SPS-2 sites

45

CHAPTER 5- EVALUATION OF KEY PERFORMANCE TRENDS

The effects of various design and construction factors on performance are manifested in
the form of distresses, which are identified from condition surveys. For the LTPP studies,
the condition of test sections, in terms of a suite of performance measures, is assessed
from time to time and is made available through DataPave. Based on those data, a
performance evaluation of the SPS-2 test sections was conducted and is presented in this
chapter.

A general assessment of performance is followed by a detailed evaluation, based
on “standards” developed by Khazanovich et al (Khazanovich, Darter et al. 1998).
Depending on the extent of distress in the pavement sections, statistical methods were

selected for analyses.

5.1. GENERAL ASSESSMENT OF PERFORMANCE

The overall performance of the SPS-2 test sections is “good”. As mentioned
before the age of the test sections ranges from 5 to 12 years with an average age of about
7.5 years. Given that the sections are “young”, severe distress is not expected and the
actual performance is in line with this expectation. Lack of distress could also be due to
mild climate, over-design, and/or low traffic (Khazanovich, Darter et al. 1998).

Pumping, faulting, and cracking are the “common types of distresses” in jointed
plain concrete pavements (Huang 1993). In addition to these distresses, corner breaks,
blowups, spalling, durability cracking, sealant damage, and scaling are observed for rigid
pavements.

While cracking occurred in about 30% of the sections, pumping occurred only in

about 13% of the sections. Most of the pumping occurred at the site in Arkansas (5)

46

where all sections but for one have length of pumping between 40 and 152 m. It is
important to note that the definition of pumping in the LTPP also includes water bleeding
which may not be as critical as pumping of fines along with water. In general, faulting in
the SPS-2 sections is not significant.

While about 60% of the sections have spalling, only seven sections have comer
breaks. Most of the spalling that was observed is longitudinal spalling, which occurred in
53% of the sections. Longitudinal spalling was found in 80% of the pavement sections
that have transverse spalling. While most of the transverse spalling was low in extent,
longitudinal spalling was about 8 m long on average.

None of the sections have durability cracking or blow-ups. However, scaling was
observed in 13% of the sections and among those sections the average scaling is about 44
m2. While transverse joint sealant damage, mostly of low severity, was observed in all the
sections at almost all the joints, longitudinal sealant damage was observed in about 60%
of the sections. On average, 162 m of sealant damage was observed in sections with
longitudinal sealant damage. Longitudinal sealant damage usually accompanied pumping,
in that 86% of the sections with pumping have sealant damage.

In light of this condition of the SPS-2 pavement sections, key performance
measures were selected for analysis. The criteria for selection were the mechanism of
distress, the relation of the distresses with the experimental factors, and the extent of
occurrence of the distress. As spalling, scaling, and sealant damage are more related to
PCC material and construction practices, they were not selected as key performance
measures. Though pumping is a common distress type, because of low occurrence and

because water bleeding was also measured as pumping, it was not selected. Hence, the

47

main performance measures considered for analyses are cracking (transverse and
longitudinal), faulting, and ride quality (IRI). Details of the performance evaluation

follow.

5.2. TRAN SVERSE CRACKING

Transverse cracking in LTPP is measured and presented in two forms— number
of cracks and length of cracking. In each of these two forms the severity level of cracking
is also measured and presented. For this study, the total number of transverse cracks was
taken as the performance measure.

Among the SPS-2 sections, transverse cracking was observed in 31% of the
sections. Figure 5-1 shows the distribution of cracking in the SPS-2 sections. About 65%
of the cracked sections have medium severity cracking, and 37% of the cracked sections
have high severity cracking. A possible explanation for the “low” occurrence of
transverse cracking lies in the contraction joint spacing of 4.6 m (1 S-fi), which is
typically considered the optimum spacing to avoid cracking (Ray 1981; Smith, Peshkin et

al. 1989; Armaghani 1993; Harvey, Roesler et al. 2000).

 

Cl 0 El 0-25 25-50 I 50-100

Figure 5-1 Distribution of transverse cracking (percent slabs cracked)

The site-wise occurrence of transverse cracking in the SPS-2 test sections is
shown as box plots in Figure 5-2. While the box and whiskers give an idea about

distribution of cracking, the horizontal line in the box is the median. It is evident from the

48

plot that the sections in Nevada [NV (32)] have distinctly higher cracking than sections in
any of the other sites. The data from the sections at this site were excluded from analyses
as severe cracking at the site was premature and was related to construction issues.
According to the construction report (Nichols Consulting Engineers 1998), the PCC
mixes were different from the ones stipulated by the SHRP. The sections with target 14-
day flexural strength of 3.8 MPa (550—psi) have a 3.3 MPa (475-psi) mix and the sections
with target 14-day flexural strength of 6.2 MPa (900-psi) have a 5.2 MPa (750-psi) mix.
Shrinkage cracking also occurred in the LCB and could have contributed to the outlying
performance of the test sections. Excluding sections from NV, cracking was observed in
26% of the sections, with medium severity cracking in 63% of the cracked sections, and
high severity cracking in 23% of the sections. Figure 5-5 and Figure 5-6 show the
occurrence of cracking as a function of design and site factors. The occurrence of
transverse cracking shown in these figures does not include sections from the site in NV
(32). Of all the experimental factors, it seems that base type, PCC slab thickness and

drainage have considerable effects on the occurrence of cracking.

49

 

 

 

 

250-“
£2 0
3200—
3
“>3 150—1
U)
G
8
95 loo—T
‘—
O
.D
E 50—
Z *
*
0_=Afi*;++; ;;é+_
I I I _ I I I I I I j I I I I
4 5 6 8 10 I9 20 26 32 37 38 39 53 55

State ID
Figure 5-2 Site-wise occurrence of transverse cracking for SPS-2 test sections

The transverse cracking performance was compared to the performance criteria
developed by Khazanovich et al (Khazanovich et al, 1998) for JPCP pavements.
According to those criteria, for pavements that are 7.5 years old, which is the average age
of sections in SPS-2, “good” performance is defined as less than 2% slabs cracked,
“normal” performance is defined as 2 to 4% slabs cracked and “poor” performance is
defined as more than 4% slabs cracked. The criteria were developed for transverse
cracking in terms of percent slabs cracked. The percent of slabs cracked is not given
directly in LTPP database. Data from faulting file (MON_DIS_JPCC_FAULT) of the
data were used in conjunction with distress maps to identify crack locations and calculate
the percent slabs cracked.

Based on the criteria, among sections in SPS-2 excluding ones from NV, 72% are
exhibiting “good” performance, 6% are exhibiting “normal” performance while the

remaining 21% are exhibiting “poor” performance.

50

5.3. LONGITUDINAL CRACKING
In the LTPP database, longitudinal cracking of different severity levels is

presented in linear meters. For this study, total length of longitudinal cracking was taken
as the performance measure. Longitudinal cracking was observed in 27% of the test
sections of the SPS-2 experiment and most of the cracking, at this stage, is of low
severity. Figure 5-3 shows the distribution of longitudinal cracking occurrence in all the
SPS-2 sections. The site-wise occurrence of longitudinal cracking in the SPS-2 test
sections is shown as box plots in Figure 5-4. As in the case of transverse cracking, the
outlying performance of sections from NV (32) is evident for longitudinal cracking. In
addition to sections from Nevada (NV), some sections from Arkansas [AR (5)] and
Arizona [AZ (4)] have shown outlying performance. The probable reasons behind
abnormal transverse cracking performance of the Nevada sections also apply to the case
of high longitudinal cracking in those sections. These sections (from AR and AZ) were
not excluded from analyses, as the median performance of these sections is comparable to
the performance of other SPS-2 test sections. In addition, no construction deviations that
could substantiate the outlying performance were documented in the construction reports.
Figure 5-7 and Figure 5-8 show the occurrence of longitudinal cracking as a
function of design and site factors. These figures are based on data from all SPS-2
sections other than ones from NV site. Of all the experimental factors, it seems that base

type and PCC slab thickness have considerable effects on the occurrence of cracking.

51

73% i

 

220/0

5%
G 0-25 E11 25-50 I >50
Figure 5- 3 Distribution of longitudinal cracking, m

 

300-‘

' 250-

ksm

50— * ,,

*

Length of Longitudinal Crac

* *
0- :1 _.____.__-__5.4 ;$__:__

I I I I I I I I I I I I I I
4 5 6 8 10 19 20 26 32 37 38 39 53 55

State ID
Figure 5- 4 Site-wise occurrence of longitudinal cracking in SPS-2 test sections

 

 

 

5.4. JOINT F AULTIN G

The site-wise occurrence of faulting in the test sections is shown as box plots in Figure 5-
9. It is evident from the plot that less than 5 joints per section have faulting greater than
1.0 mm, in a vast majority of. Figure 5-10 shows the distribution of faulting (in terms of
percent of joints that faulted more than 2 mm) in the SPS—2 test sections. Figure 5-13 and
Figure 5-14 show the distribution of faulting by site factors and design factors in sections
other than ones from NV. None of the factors seem to be significantly affecting the

occurrence of faulting.

52

Percent of sections cracked

50% __

40% r

30% r

20% ‘

10% ‘

 

0%i

Percent of sections cracked

40% i

 

50%

30% r
20% a

10%

 

0%2

JIILII

 

DF DNF WF WNF

 

Site factor levels

 

Figure 5- 5 Occurrence of transverse cracking by site factor

 

 

8 8 a 2 2 a 8 5 9

‘3 °‘ 8 E 2:: 5:;
Design factor levels

Figure 5- 6 Occurrence of transverse cracking by design factor

B 45% ._ 5-.-“, w w- ~-~-—---~—T

i4, 40% —

g 35% —

E’ 30% r

g 25% -

8 20% r

“5 15%

E 10% —

E 5%-

°‘ 0% ~

DF DNF WF WNF

Site factor levels

Figure 5- 7 Extent of longitudinal cracking by site factors

53

3.7 m

4.3 m

 

 

 

 

 

 

 

 

 

 

 

E 50%
8
:3 40% r
2
.2 30% ‘
E
,3 20% -
O
S 10% a
a: 0% -
m m E a o w =9 E E
E; E3 :2 I5 EI 25 E; E: r- rfi
D 9* 8 Si 00 N M <1-
rfl cu en so
Design factor levels
Figure 5- 8 Extent of longitudinal cracking by design factors
25-
E 0
g 20—
7C
00
5 15—
é o
5
'E 1o—J o
a
.5
5’.
d 5— o g *m
z o
101
26 0 * 155 159
0—1 é-*ET_'Ié.—5‘%I:I 1&1 &* *
I I I I I I I I I I I I I I

4 5 6 8 10 19 20 26 32 37 38 39 53 55
StateID

Figure 5- 9 Site-wise occurrence of faulting in SPS-2 test sections

Based on the performance criteria by Khazanovich et al (Khazanovich et al,
1998), pavements of age 7.5 years are said to be exhibiting “good” performance if the
average joint faulting is less than 1.5 mm, “normal” performance if the average joint
faulting is between 1.5 and 3.0 mm, and “poor” performance if the average faulting

exceeds 3.0 mm.

54

 

27%

 

E 0 El 0-20 I 20-40 I >40
Figure 5- 10 Percent of joints that faulted >= 2.0 mm

According to these thresholds, all the SPS-2 sections except one are showing
“good” performance. As mentioned earlier, the 203 mm and 279 mm test sections are
provided with 32 mm (1.25-in) or 38 mm (1 .5-in) dowels, respectively, and the
contraction joint spacing is 4.6 m (15-ft). As the joints are doweled and the joint spacing
is not large, it seems that the performance of the test sections is as expected.

In general, the extent and magnitude of faulting, as mentioned before, is “low”
indicating “good” performance of a majority of the test sections. In addition, for LTPP
data collection the Georgia Faultmeter, which has a least count of 1.0 mm, is used to
measure faulting at joints and cracks. This least count is “large” (Selezneva et al, 2000),
as typical faulting values in SPS-2 pavements are close to 1.0 mm. In light of these
issues, it is too early to make reliable conclusions regarding the effects of design and/or
site factors on faulting. Thus results from analysis on faulting are not presented in this
thesis.

5.5. ROUGHNESS (IRI)
The site-wise status of current roughness in the test sections is shown as box plots in
Figure 5-11. It is evident from the plot that for most of the sections the current roughness

is less than 1.8 m/km, i.e. Present Serviceability Index (PSI) of 3.5 based on equations

55

proposed by Hall and Correa (Hall and Correa Mufioz 1999). Also, sections from the site
in Michigan [MI (26)] have higher roughness than most of the sections in the experiment.
Sections 0214, 0215, 0217 and 0218 were observed to have roughness ranging between
3.2 to 4.5 m/km. Sections 0214 and 0218 were built with a roughness of 1.5 and 1.9
m/km, respectively. The sections 0215, 0217, and 0218 were decommissioned from the
experiment by year 2000, due to reasons not mentioned in DataPave. However, data
available in the database till the sections were decommissioned was used in this research.
The status of current roughness in the test sections is given in Figure 5-12. Figure 5-15
and Figure 5-16 show the distribution of roughness by site factors and design factors.

Drainage and subgrade soil type seems to be having considerable effects on roughness.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

g 4"

2

CL 3—

$

0)

i=3

2?

a 2‘ - o

E ‘— Q T lfll $#

0 l '

a 1— é 1 ¢$'i'
l I I F I I I I I I I I I
4 5 6 8 10 19 20 26 37 38 39 53 55

StateID

Figure 5- 11 Site-wise occurrence of final roughness values for SPS-2 test sections

56

 

6% 3% 3%
I<l D 14.5 D 1.5-2 132—2.5 I>2.5

Figure 5- 12 Distribution of roughness, IRI m/ km
By Khazanovich et a1 (Khazanovich et al, 1998), a pavement at an age of 7.5 years, is
“good” if roughness is less than 1.0 m/km and “poor” if roughness is greater than 2.0
m/km. Based on these criteria, the initial roughness of 54% of the sections is “poor” and
that of the other sections is “normal”. From the current roughness data, 88% of the
sections are exhibiting “normal” performance and 6% of the sections each are exhibiting

“good” or “poor” performance.

 

10%
8% ‘

6%~

4%“
0% l L W,

DF DNF WF WNF

Percent of joints

 

 

Site factor levels

Figure 5- l3 Extent of faulting >=l.0 mm in site factors

57

Percent of joints

m/km

Averager rsoughnes,

 

 

 

 

 

 

 

   

 

 

 

 

10%
8% s
6% d
4% 4
2% a I
0% s
m m m E E Q <6 N E 8
(<5 3 g e e E g g 14 m,
D ”-1 8 3 00 N m V
N N m‘ ,5
Design factor levels
Figure 5- l4 Extent of faulting >=1.0 mm in design factors
V
5 1.75 —
E. 1.65-
8
"Eb 1.55 -
§ 1.454
0.)
§ 1.35 .
3
< 1.25 a
DP DNF WF WNF
Site factor levels
Figure 5— 15 Extent of roughness in site factors
1.75 —
1.65 r
1.55 i
l .45 4
1. 35 5 I
1.25 -
m m E E Q a <6 «3 E E
g 3 g E E 2 E E 1‘. M.
Q a. 8 [2‘ 0° N to <1-
N N m' \0'
Design factor levels

Figure 5- l6 Extent of roughness in design factors

58

 

 

CHAPTER 6- ANALYSIS METHODS

In this chapter, a description of the analysis methods employed for investigating the
effects of experimental factors on pavement performance is presented. Broadly, the
analyses were performed at two levels— site-level and network-level (or overall
analysis). In the site-level analysis, as the name suggests, the effects of design factors
were studied at each site and the consistency of the effects across sites was evaluated.
The main advantage of such analysis is that all the sections under comparison, for a given
site, were subjected to similar treatment (mainly, traffic and climate). The network-level
or overall analysis was performed to make use of the wealth of data available from the
SPS-2 experiment, and study the effects (main and interaction) of design and site factors
on performance.

For each test section in the experiment, time-series performance data are
available. Performance index (PI) has been developed to capture the entire time-series
performance of a section in a single number that represents the section’s performance. A

brief description of PI is given first and the details of analysis methods follow.

6.1. PERFORMANCE INDICATORS

As mentioned before, the performance data are available as time-series data. In
order to employ “conventional” statistical procedures that do not involve analysis of
time-series data, it was necessary to develop a method to quantify the performance as a
single quantity. The three options that were considered for this are: area under the
performance curve (AREA), area under the performance curve normalized with age
(AREA/age), and the performance index (PI). However, given the variation in age among

the sites, two sections of different ages can have the same area though the older section

59

has actually shown better performance. For example, the younger section could have
significant distress over only a few years while the older one has more-or-less uniform
distress over a much longer time; while the area under the curve would be the same, the
older section would have exhibited much better performance.

P1 was found to capture better the recent performance in comparison with
AREA/age, though PI and AREA/age have a good correlation. The association between
PI and AREA/age is shown in Figure 6-1 for the case of IRI, which is the performance

measure to which the concept of P1 was applied.

 

 

 

 

3.0
o
2.5 ‘ o
. o
2'0 I 0 o o o 0
o' - ' '~ '
' o
..O
E 1.5 ‘ or :.
. I
g 0
1.0 ‘
0.5 *
0.0 I I I I I
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Area/age

Figure 6-1 Correlation between PI and AREA/age, for IRI
P1 was thus selected as it accounts for both age to initiation and the rate of growth

of a distress. A detailed illustration of the reasons behind selection of PI can be found

elsewhere (Chatti et al, 2005; Buch et al, 2005).

60

PI is defined as:

2019);)
:77:—

1'

PI (1)

Where:

I,- = age of the section at distress measurement in year 1'

3’1 = performance of the section at year 1’

Only the ages at which distress measurements were taken are included in the
calculation of PI. Since PI gives greater weight to the most recent performance of a
section, it may capture the most probable future trend of performance.

Although PI appears to be a viable option it has some inherent limitations. These
limitations are mainly because PI is dependent on the number and timing of the distress
surveys. To address this, the age of test sections was considered as a covariate in all
statistical analyses of PI. Though the actual effects of traffic and age on pavement

performance are quite complex (non-linear) and may not be adequately addressed by the

covariates (linear adjustment), their inclusion in that manner helps in reducing bias.

6.2. SITE-LEVEL ANALYSIS METHODS

In the site-level analysis each section is evaluated in comparison with similar
designs of a site. Within each site, climatic conditions and traffic loading are identical for
all test sections. Thus the main advantage of this analysis is that comparisons are made
among those sections that were subjected to similar loading and environmental
conditions. All site-level analyses were conducted on the Performance Indices (PIS) of
the sections for various performance measures. The difference in performance is assessed
based on average values.

The site-level analysis consisted of series of comparisons, each focusing on the

effect of a particular design factor. Comparisons were done at two levels—A and B. In

61

level-A analyses, all designs (0201 through 0212, or 0213 through 0224) at a given site
were compared such that only one factor is held common within the sections of each
group. For example, in level-A analysis, the effects of PCC slab thickness [203 mm (8-in)
vs. 279 mm (1 1-in)] were studied, within each site, by ignoring base type, drainage, PCC
flexural strength, and slab width.

In level-B analyses, most of the factors are “controlled”. In other words,
individual sections within a given site are paired such that all but one design factor are
the same, this factor being the one of interest. Comparing a given pair of sections will
allow for determining the effect of the particular design factor, with the highest possible
level of constraint (level-B).

The relative effects of levels within each design factor were studied based on the
ratio of mean performance of the sections corresponding to a level over the mean
performance of all levels of the factor.

For factors with two levels (such as PCC thickness and drainage), the relative
performance of each level can range from O to 2, a value of 1 indicating no effect of the
factor [i.e., the amount of distress (performance) corresponding to the two levels of the
factor is the same]. A value less than 1 indicates better performance compared to mean
performance of sections corresponding to both the levels of a factor. Consequently, a
value higher than 1 indicates poor performance. The best possible performance translates
to 0, and the worst possible performance translates to 2. For cases where there is no
distress or same level of distress, each level of a given factor will have relative
performance of 1 indicating no difference in performance. A sample calculation of

relative performance is presented in Table 6-1.

62

For factors with more than two levels, a similar logic can be extended. The
relative performance for various levels of the design factors was calculated for all the
sites in the experiment, and for each performance measure.

Table 6- 1 Example calculation of relative performance for the effect of slab
thickness on longitudinal cracking

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8” PCC slab thickness 11” PCC slab thickness
Section ID Performance Index Section ID Performance Index
201 90.7 203 23.1
202 25.4 204 3.8
205 95.9 207 16.1
206 22.0 208 70.5
209 65.3 211 75.5
210 10.2 212 52.7
Average 5 l .6 Average 40.3
Mean
Performance (51.6+40.3)l2 = 46.0
Relat‘ve 51.6/46.0=l.1 Relative performance 40.3/46.0=O.9
performance

 

 

 

 

6.3. OVERALL ANALYSIS METHODS

The selection of statistical methods for analyses was mainly based on two
criteria—objectives of the study, and the challenges posed by the experiment design and
current performance levels. Two types of analyses were selected—magnitude—based and
frequency-based. Analysis of variance (ANOVA) is a method for comparing means and
is the magnitude-based analysis. Binary logistic regression (BLR) and linear discriminant
analysis (LDA) are frequency-based analyses, which were used to evaluate the effects of
experimental factors on the occurrence or non-occurrence of distress. These methods are
briefly described below. A detailed explanation of analyses for each performance

measure can be found elsewhere (Buch et al, 2005; Chatti et al, 2005).

6.3.1. Frequency-based Analysis

The average age of the SPS-2 test sections is 7.5 years. As a result, the magnitude

and occurrence of distress are “low”. For example, approximately 73% of the sections

63

have not shown any transverse cracking. Though this indicates “good” performance, the
absence of distress in a considerable number of sections presents challenges for statistical
analysis.

Traditionally, magnitude-based methods like multi-factorial ANOVA are used to
establish the impact of design and site factors on pavement performance for an
experiment like SPS-2. For the normal theory least squares ANOVA to be meaningful the
model residuals should (i) be normally distributed and (ii) have a constant variance. For
ANOVA on transverse or longitudinal cracking, these assumptions are violated (see
Figure 6-2), owing to the skewed distribution of cracking occurrence in the SPS-2 test
sections. To address this issue, a logarithmic transformation could be applied to the data;
however, the ANOVA would be restricted only to sections that have shown cracking, as
the transformation function is not defined for a value of zero. The low occurrence of
cracking limits the number of sections that can be used in such an analysis. ANOVA was
used only to study the effects of the experimental factors on roughness.

The frequency-based methods are more suitable when the occurrence of distress is
considerable, even though the magnitude of distress is less, because the categorization is
based on occurrence only. Hence, binary logistic regression (BLR) and linear
discriminant analysis (LDA) were used to study the effects of design and site factors on

the occurrence of transverse and longitudinal cracking.

64

 

 

 

 

 

 

o
20-
10-1 O
U 00 o
2 ° 0 0° 0 o o o
O 00 o O O
‘V %'
'2 may, fig?” 00
o
3'3 o
a:
-10-
-20-
I I I I I I I
-4 —2 0 2 4 6 8
PredictedValueforTC

Figure 6-2 Example of violation of residual assumptions for ANOVA

BLR (Hosmer and Stanley, 2003) is used where the outcome variable is discrete.
This method is based on the maximum likelihood method for estimating the parameters
of interest. The interpretation of effects for various levels of the categorical (independent)
variables is very convenient in terms of the odds ratio when this type of model is used.
Logistic regression models are also very useful for discrimination analysis (of various
groups) when categorical variables are used as independent variables. In the
interpretation of results from BLR, the p-value is of primary importance as it indicates
the statistical significance of the effect.

LDA (Norusis, 2005) is useful in distinguishing between two or more groups of
data. This is done by identifying variables that are significant in classifying the data into

the various groups. For example, in the work here, groups could include pavement

65

sections with cracks and sections with no cracks. The method allows for determining

which variables discriminate between cracked and non-cracked pavements.

6.3.2. Magnitude-based Analysis

Magnitude-based analysis was applied to roughness, which does not pose the
problem of non-occurrence of distress. For the ANOVA, traffic was considered as a
covariate.

The SPS-2 experiment, which is a fractional-factorial design, was rendered
unbalanced because unequal numbers of sites were constructed in each zone-subgrade
combination. The imbalance in the experiment resulted in a fewer or no replications for
some treatments, which affects the statistical power (Kutner et al, 2005). In simple terms,
the lower the power, the lower the reliability of results.

In light of these issues, a simplified analysis based on the concept of the standard
deviate was performed. The performance of designs with respect to each other was
evaluated using the deviation from the mean performance, which is the standard deviate.
The designs were evaluated based on selected distresses, considering one distress at a
time. The standard deviate was calculated for each of the twelve designs within each site
by using the following equation:

(Performance of a given design — Average performance for the given site)

 

S tan dard Deviate = . . _ .
Standard devration of performance for the given Site

Since this measure is calculated individually for each section, considering one site
at a time, it indicates the relative ranking of a section compared to other sections at the
same site. This transformation adjusts for the variation in performance due to site
conditions, as the sections are compared to contiguous sections at their respective sites.

This method is analogous to statistical “blocking” for site effects. The standard deviate

66

for a particular performance was used to compare the individual effects of design factors
by employing one-way ANOVA on the standard deviates of the sections. These analyses
were performed on data from all sections and also on subsets of data stratified by
subgrade type and climate. While a lower value of standard deviate indicates better

performance than the mean, a higher value indicates worse performance.

67

CHAPTER 7- EFFECTS OF EXPERIMENTAL FACTORS ON
PAVEMENT PERFORMANCE

Findings from the statistical analyses conducted to study the effects of design and site
factors on performance of the SPS-2 test sections are presented in this chapter. The
performance measures that were analyzed include transverse and longitudinal cracking,
and roughness (initial roughness and change in roughness). In this chapter, results from
site-level analysis will be followed by results from overall analysis and comparison of
designs, leading to the summary of findings. A list of all analyses conducted on the
performance of SPS—2 test sections is given below:

0 Site-level analyses: Evaluation of the consistency of the effects of design factors

across sites.

0 Overall analyses:
a) Analysis of Variance (ANOVA)

b) Linear Discriminant Analysis (LDA)
c) Binary Logistic Regression (BLR)

0 Comparison of designs (based on standard deviate)

7.1. EFFECT OF CONSTRUCTION ON PAVEMENT PERFORMANCE

Detailed construction guidelines were developed by LTPP for the participating agencies
to minimize variability in construction across sites. The lesser variability across sites, the
lower the “noise”, and easier it is to determine the “pure” effects of the design factors on
pavement performance. However, some deviations have occurred during construction of
some sites and those issues were highlighted in the construction reports prepared by the
participating agencies. Inclusion of test sections with significant construction deviations

in the analysis may distort (bias) the effects of design factors.

68

Depending on the nature of construction issues (construction quality) the
performance of a pavement is affected. Minor issues (such as minor thickness deviation
in base course) may not significantly impact initial performance where as major issues
(such as PCC mix issues, or drainage problems) have greater chances of affecting
performance early in the life of a pavement. Hence, a construction deviation (poor
quality) may be used to identify sections that may potentially show an abnormal
performance.

In this study, any abnormality in early performance was used as an indicator to
identify substandard sections. The performance of all the sections, over time, was
observed for this purpose and those sections that had premature “failure” (in first few
years of service life) were identified. In order to further investigate the construction-
related performance issues, the performance, in terms of key performance measures, of
all pavement sections in the experiment was examined. The analysis is discussed next
with illustrations. Based on the time-series plots for all performance measures it was
found that cracking (transverse and/or longitudinal) was the predominant premature

“failure” for most of the pavement sections.

7.1.1. Transverse and Longitudinal Cracking

Figure 7- 1 and Figure 7- 2 show cracking in all test sections. It can be observed
that some sections exhibit high initial cracking. It was found that most of the sections
with this abnormal performance are from NV (32). As mentioned earlier, a wide range of
construction issues (material-related) that were reported in the construction report for the
site are believed to be the cause. In light of the unusual behavior of test sections at NV

(32) site, data from these sections were excluded from all subsequent statistical analyses.

69

 

Figure 7- 3 and Figure 7— 4 show transverse and longitudinal cracking in all sections

except those from NV (3 2).

 

250 _

N
8
.1.

u—-

M

C
1

Transverse Cracking (No )
8

‘
‘1

 

 

 

Age (years)

Figure 7- 1 Transverse cracking with time - All sections

 

300 _
250
200

150

 

100:

Longitudinal Cracking (m) .

 

50:

 

 

 

0 1 2 3 4 5 6 7 8 9 10 1 l 12
Age (years)

Figure 7- 2 Longitudinal cracking with time - All sections

70

250

 

200

Y T If I I T I 1'

150 i

100*

Transverse Cracking (No.)

50~

 

 

 

 

 

12
Age (years)
Figure 7- 3 Transverse cracking with time - Selected sections (without Nevada)
300

 

250 i

I I T

T r

200

150 -

T I l I I Y

100

Longitudinal Cracking (m)

50

 

l I

A

 

 

MT
1

 

 

12

Age (years)

Figure 7- 4 Longitudinal cracking with time - Selected sections (without Nevada)
7.1.2. Roughness

Figure 7-5 shows the progression of roughness over time in all the SPS-2 sections
except NV (32). It can be observed that only few sections have exhibited an unusual
performance. Exclusion of data from these few sections was not considered necessary, as

their inclusion will not impact the results significantly. Moreover, available data do not

71

suggest any reasons behind this abnormal performance. Therefore, all the pavement

sections [except those from NV (32)] were included for analysis of roughness.

 

 

 

 

 

12

Age (years)

Figure 7- 5 IRI with time - Selected sections (without Nevada)

7 .1.3. Drainage Issues

All the drained sections of the SPS-2 experiment were video taped to assess the
condition of the drainage in late 2001 and early 2002 under the NCHRP Project 1-34C
(Hall and Correa 2002). At that time, some of the sites (e.g. KS, MI, CO, AZ, and NC)
were about 8 years into service, while the California site was just about 1 year old.

A subjective assessment of the quality of the drainage functioning as “good” or
“poor” was reported for each section. A “poor” rating is an indication of; (i) buried
lateral outlet, (ii) outlet fully blocked with silt, gravel or other debris (iii) longitudinal
drains being fully blocked, or (iv) a considerable amount of standing water in the
longitudinal drain. A “good” rating was given to drainage if a reasonably sufficient flow

of water was evident even if some amount material was present in the drains.

72

 

The ratings assigned to each section of the experiment in Table 7-1. As shown in
the table, some of the sections that were designed to be un-drained, according to the
experiment design, were constructed with drainage. Hall and Correa (Hall and Correa
2002) conducted preliminary analysis of the performance of SPS-2 test sections in light
of their assessment of drainage, and a brief summary of their findings are presented
below:

0 With respect to [RI change, larger mean differences were detected for the PATB
sections with “poor” drainage than for PATB sections with “good” drainage,
when un—drained and drained sections were compared. The quality of drainage is
not a significant factor in the differences observed in IRI increase.

0 In the analyses of transverse and longitudinal cracking in drained versus un-
drained SPS-2 sections, larger mean differences were detected for PATB sections
with “good” drainage functioning than for those with “poor” drainage
functioning.

However, the above trends were based only on the average performance and in no
case, were the differences detected statistically significant based on analysis of data from

Release 17 of LTPP database.

7.2. SITE-LEVEL ANALYSES

The results obtained from site-level analyses of the SPS-2 experiment data are
presented here. The concepts of performance index (PI) and relative performance were
used to perform site-level analyses (details in Chapter 6). These analyses were conducted

separately for transverse cracking, longitudinal cracking, and roughness (IRI).

73

 

Table 7- l Subjective ratings of drainage functioning for SPS-2 test sections based on
video inspection results (Hall and Correa 2002)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Test Section ID

0201 0202 0203 0204 0205 0206 0207 0208 0209 02 10 02 l l 02 12

0213 0214 0215 0216 0217 0218 0219 0220 0221 0222 0223 0224
State Base Type

Dense-graded aggregate base Lean concrete base Peréneable asphalt-treated

ase over aggregate
Un—drained Drained

4 G l G I G I G
5 P P I P P P P
6 P G G [ G [ P
8 G ,P P _ P A
10 'L P G P 7 G
19 P P ? ?
20 G G G G
26 " P‘ ‘” if" ‘ P .- ? P’ ?
32 G G G ?
37 '13] . ' P 'P P 111;,
38 G G G G
39 ‘2' '2' ‘2’ ‘2‘ ‘2' :-?,' '2'? '2', G .P.’ . G "P *
53 G G G G
55 ? ? ? ?

 

1

G: Drainage function rated as good

2? = Drainage outlet not found

3P = Drainage function rated as poor
4? = Camera could not be inserted

Site-level analyses deal with each SPS-2 project separately. For each site, the

climatic conditions, subgrade type (for most of the sites) and traffic are same.

Construction conditions, material sources and surveys were also considered to be same

for all sections within each SPS-2 site.

As described in Chapter 6, the site-level analyses consists of two types of

comparisons: (i) Level-A —— In this analysis all designs (201 through 212, or 213 through

224) at a given site are compared (among themselves) such that only one factor (design

feature) is held common within the sections of each group under comparison; (ii) Level—B

— In this analysis, most of the factors (design feature) are “controlled” for comparisons.

The results from level-A and level-B comparisons, in terms of relative performance ratio,

can be found in the Appendix.

74

 

Non-parametric tests (Wilcoxon Signed Ranks test and Friedman test) were
performed on relative performance ratios to determine the statistical significance of the
difference in relative performance ratios of different levels within each design factor. For
example, the relative performance ratio corresponding to transverse cracking, for sections
with 203 mm (8-inch) slab and sections with 279 mm (1 l-inch) slab were compared to
investigate the statistical significance of the consistency of the effect of PCC slab
thickness on transverse cracking across sites. A p-value less than or equal to 0.05 was
considered to be indicative of a statistically significant consistency of an effect. F

In site-level analyses, statistical significance of an effect needs to be interpreted as
the significance of the effect’s consistency across sites but not necessarily as the
significance of its effect on the magnitude of distress.

In this chapter, the discussion of results for level-A and level-B analyses is
presented separately. Some basic descriptive statistics regarding the performance of the
test sections are also presented, to corroborate the results. Though these statistics are not
at site-level they are meant to give the reader an insight about the extent of distresses.

In the SPS-2 experiment, only the test sections built on PATB were provided with
in-pavement drainage. As a consequence of this, the impact of drainage alone or base
type alone cannot be studied. In other words, the effect of PATB and the effect of
drainage cannot be separated. Therefore, an assumption was made that DGAB and PATB
are structurally the “same”, and the analysis was performed by comparing performance of
sections constructed on DGAB and sections constructed on PATB. It is important to note
that the effect of drainage discussed in this thesis would be a result of comparison

between sections on DGAB and sections on PATB. Furthermore, to study the effect of

75

base type, the performance of sections with DGAB, sections with LCB and sections with
PATB were compared. Here too the effect of PATB is combined with the effect of

drainage.

7.2.1. Effect of design features on performance- Comparisons at level-A

The discussion of results from level-A analyses is presented here. These results
are presented for one design feature at a time.
Drainage

To investigate the effects of drainage, sections 201 through 204, and, 213 through
216 were considered as “without drainage” and sections 209 through 212, and, 220
through 224 were considered as “with drainage”. Hence it is important to note that the
effects of drainage that are discussed here are from comparisons only between sections
built on DGAB and sections built on PATB.

From level-A analysis, the effects of drainage on cracking and roughness are
inconclusive, at this point in time. This observation should not be interpreted as drainage
not having a significant impact on pavement performance in general. All the observations
and conclusions need to be interpreted keeping in view the age of the test sections and the
low occurrence of distresses in the SPS-2 test sections. Table 7-2 is the summary of
effects of drainage on cracking (transverse and longitudinal) and roughness.

Base Type

Sections built on each of the three base types, DGAB, LCB and PATB, were
compared at each site to study their relative impact on performance. Base type has a
consistent effect on cracking. However, the effect is not consistent (across sites) for

roughness.

76

Table 7-2 Effects of drainage*, based on Level-A analysis
Performance
Measure

 

Effect Comments

 

o In 5 of the 14 sites, no cracking occurred and the
performance of sections with and without drainage is
thus similar.

0 In 6 of the 9 sites with distressed sections, sections
without drainage exhibited more cracking than ones
with drainage.

0 Overall, 25% of sections without drainage and 12%
of sections with drainage have exhibited cracking.

o 21% of sections in WF zone, 13% of sections in
WNF, 35% of sections in DE zone, and 19% of
sections in DNF zone have exhibited cracking.

0 In 5 of the 9 sites with distressed sections, sections
without drainage exhibited more cracking than ones
with drainage.

0 In 5 of the 14 sites, no cracking occurred and the
performance of sections with and without drainage is
thus similar.

0 Overall, 21% of sections without drainage and 19%
of sections with drainage have exhibited cracking.

o 9% of sections in WF zone, 19% of sections in
WNF, 35% of sections in DF zone, and 25% of
sections in DNF zone have exhibited cracking.

o In 11 of the 14 sites, the performance of sections

R . with and without drainage is comparable.

oughness Inconclusrve . .
(IRI) (p= 0.084) 0 Average latest roughness of sect1ons wrthout
drarnage and sections wrth drarnage are 1.6 and 1.3
m/km.

*Effect of drainage is a result of comparison between sections built on DGAB and sections built on PATB

only.

Transverse Inconclusive
cracking (p=0.299)

 

Longitudinal Inconclusive
cracking (p: 0.411)

 

 

 

 

 

 

Approximately 59% of sections built on LCB have exhibited cracking compared
to 38% of sections built on DGAB and 25% of sections built on PATB. Though the
analysis indicates higher cracking in sections built on LCB, the conclusions need to be
considered in light of the construction issues (Desaraju 2003; Chatti, Buch et a1. 2005)
and, the magnitude and severity of cracking. Table 7-3 is the summary of the effects of
base type. The effect of base type on roughness seems to be inconclusive because at most
of the sites the difference in IRI of sections built on the three base types is not

considerably high. As of latest distress survey, 80% of the sections in the experiment

77

have IRI less than 1.8 m/km. The extent and magnitude of the distresses are to be
considered along with the conclusions.

Table 7- 3 Effects of base type based on Level-A analysis

 

 

 

 

 

 

 

 

 

Performance Effect Comments
Measure
0 In all the 13 sites with distressed sections, higher
T Consistent cracking was observed in sections built on LCB,
ransverse .
cracking effect compared to other sections.
(p: 0.000) 0 25%, 46% and 12% of sections on DGAB, LCB,
and PATB, respectively, exhibited cracking.
o In 11 of the 13 sites with distressed sections,
Longitudinal Consistent higher cracking was observed in sections built
cracking effect on LCB, compared to other sections.
(p: 0.002) 0 21%, 42% and 19% of sections on DGAB, LCB,
and PATB, respectively, exhibited cracking.
0 In 10 of the 14 sites, comparable roughness was
observed in all sections.
0 In 4 of the 14 sites, more roughness was
Roughness Inconclusive observed in sections built on DGAB, compared
(IRI) (p: 0.064) to other sections.
0 Average latest roughness of sections on DGAB,
LCB, and PATB are 1.6, 1.6 and 1.3 m/km,
respectively.
PC C slab thickness

A total of 84 sections with 203 mm (8”) PCC slab and 83 sections with 279 mm
(11”) PCC slab were compared (at site-level) for this analysis. This includes all the
sections in the experiment.

The effect of slab thickness is consistent in the case of transverse and
longitudinal cracking. Though a deviation from target thickness was observed in
considerable number of sections (details in Chapter 4), the analysis indicates a significant
effect of PCC slab thickness on cracking. Table 7-4 is the summary of effects of PCC
slab thickness on cracking. It is to be noted here that 49% of cracking (transverse and/or
longitudinal) has occurred in sections that were built on LCB, of which 67% of the
sections are sections with 203 mm PCC slabs. These statistics suggest a noticeable effect

of both slab thickness and base type. The roughness of both 203 mm PCC slab and 279

78

I...

mm PCC slab sections was found to be comparable at all sites, suggesting an
insignificant effect of slab thickness on roughness. The effect of PCC slab thickness on

IRI is thus inconclusive.

Table 7- 4 Effects of slab thickness, based on Level-A analysis

 

 

 

Performance Effect Comments
Measure
0 In all sites that have distressed sections (13 sites), more
Transverse Consistent cracking was observed in sections with 203 mm (8 inch) PCC
cracking effect slab, compared to sections with 279 mm (11 inch) PCC slab.
(p: 0001) 0 40% of sections with 203 mm (8 inch) PCC slab and 15% of
sections with 279 mm (11 inch) PCC slab exhibited cracking._
0 In 11 of the 13 sites that have distressed sections, more
Longitudinal Consistent cracking was observed in sections with 203 mm (8 inch) PCC
cracking effect slab, compared to sections with 279 mm (11 inch) PCC slab.

(p: 0020) 0 41% of sections with 203 mm (8 inch) PCC slab and 14% of
sections with 279 mm (11 inch) PCC slab exhibited cracking.

o In all the sites, comparable performance was observed in all

 

 

 

 

 

Roughness Inconclusive sectrons. . .

(IRI) (p: 0414) 0 Average latest roughness of both the sectrons With 203 mm (8
inch) PCC slab and sections with 279 mm (11 inch) PCC slab
is 1.5 rn/km.

 

PCCflexural strength

The performance of test sections with target 14-day PCC flexural strength of 3.8
MPa and test sections with target l4-day PCC strength of 6.2 MPa was compared to
study the effect of PCC flexural strength on the performance of SPS-2 sections. A total of
84 sections with 3.8 MPa concrete and 83 sections with 6.2 MPa concrete were
compared. The effect of flexural strength on cracking and roughness appears to be
insignificant. Comparable performance was observed in sections with higher strength
concrete and lower strength concrete.

It is important to consider the deviations from target flexural strength in the
sections. A detailed discussion of the deviations was presented in Chapter 4. The
deviation from target PCC l4-day flexural strength was studied using the data that is

available for 52% of the sections in the experiment. The average l4-day flexural strength

79

of PCC of sections with target strength of 3.8 MPa was 3.6 MPa while in sections with
target strength of 6.2 MPa was 5.6 MPa. Among sections with target flexural strength of
3.8 MPa, 34% of sections had PCC flexural strength (at 14-days) that exceeded the
allowable range of 3.4 MPa to 4.2 MPa, while 16% failed to reach even the lower limit of
the range. In the case of sections with target flexural strength of 6.2 MPa, 34% of
sections had PCC flexural strength (at 14—days) below the allowable range, and none of
the sections exceeded the range. In half of the sections with target strength of 6.2 MPa
that failed to meet the lower limit of the range, the PCC strength reached the required
limit at 28-days. These deviations from target strength could be a reason for comparable
performance of all the pavements.

Tab-1e 7-5 is the summary of effects of PCC flexural strength. At most of the
sites, comparable performance (cracking) was observed for both higher strength and
lower strength concrete sections. The effect is thus inconclusive. The effect of PCC
flexural strength on roughness appears to be insignificant.

Slab width

The widened slab [4.3 m(l4-ft)] sections were compared to those with standard
slab [3.7 m (12-ft)] to study the effect of slab width on performance of the test sections.
For this, 84 sections with standard slab width and 83 sections with widened slab were

compared. The effect of slab width seems to be insignificant, at this point in time.

80

Table 7- 5 Effects of flexural strength, based on Level-A analysis

 

Performance Measure Effect Comments

 

0 In 6 of the 14 sites, lower strength concrete
sections exhibited higher cracking than higher
strength concrete sections.

Inconclusive «1 In 5 of the 14 sites, all sections have performed at

(p=0.400) comparable levels.

0 26% of lower strength concrete sections and 25%
of higher strength concrete sections exhibited
cracking.

0 In 7 of the 14 sites, lower strength concrete
sections exhibited higher cracking than higher
strength concrete sections.

0 In 5 of the 14 sites, higher strength concrete

Transverse cracking

 

 

 

Longitudinal cracking Inggngzltlgve sections exhibited higher cracking than lower
p- ' strength concrete sections.

0 26% of lower strength concrete sections and 25%
of higher strength concrete sections exhibited
cracking.

o In all sites of the experiment, similar performance

Inconclusive was observed in sections with higher strength
Roughness (IRI) (p :0. l 02) concrete and sections with lower strength.

0 Average latest roughness of lower or higher
strength concrete sections is 1.5 m/km.

 

 

 

 

 

In general, the effect of slab width on transverse cracking seems to be
insignificant as widened slab sections and standard slab sections have performed
similarly in most of the sites. But some effect of slab width seems to exist on longitudinal
cracking. In a majority of the sites (9 of 13) slightly higher longitudinal cracking was
observed in widened slab sections compared to standard slab sections. Table 7-6 is the
summary of effects of slab width. The effect of slab width on roughness is inconclusive at
this point in time.

7.2.2. Effect of design features Paired Comparisons at Level-B

Level-B comparisons are those in which all possible factors other than the one of
interest are controlled. The individual sections that are compared under this analysis were
identified in Chapter 6. The effects of drainage, base type, and PCC slab thickness on the

performance measures are presented below.

81

Table 7- 6 Effects of slab width based on Level-A analysis

 

Performance
Measure

Effect

Comments

 

Transverse
cracking

Inconclusive
(p=0.222)

In 6 of the 14 sites, sections with standard slab
width exhibited higher cracking than ones with
wider slab.

In 5 of the 14 sites, both standard slab width and
wider slab sections have shown comparable levels
of performance.

28% and 24% of sections with standard slab width
and wider slab, respectively, have exhibited
cracking.

 

Longitudinal
cracking

Inconclusive
(p=0.362)

In 9 of the 14 sites, sections with wider slab
exhibited higher cracking than ones with standard
slab width.

25% and 26% of sections with standard slab width
and wider slab, respectively, have exhibited
craclfl

 

Roughness (IRI)

 

 

Inconclusive
(p=0.096)

 

In all sites of the experiment, similar performance
was observed in sections with standard slab width
and sections with wider slab.

Average latest roughness of standard slab width
sections and wider slab sections are 1.6 and 1.5
m/km.

 

Drainage

Sections with drainage (i.e. sections with PATB) were compared with sections

without drainage (sections with DGAB) controlling the effects of all other factors,

namely, PCC slab thickness, slab width, and flexural strength. The effect of drainage is

consistent (across sites) on transverse cracking. Slight effect was observed on roughness,

whereas no effect was apparent in the case of longitudinal cracking.

For sections with 203 mm PCC slab, the effect of drainage seems to be consistent
on transverse cracking. Table 7-7 is the summary of effects of drainage. The effect of
drainage on longitudinal cracking is inconclusive. Sections with drainage and without
drainage performed similarly in varying conditions. Among sections with 203 mm slab, a

slight effect (p=0.076) appears to exist on roughness in those with standard slab width.

Sections without drainage have slightly higher roughness than ones with drainage.

82

 

Base Type

Sections built on each of the three base types, DGAB, LCB and PATB, were
compared at each site by controlling the effects of PCC slab thickness and slab width. A
consistent effect of base type on transverse and longitudinal cracking was observed. The

effect of base type on roughness is not clear.

Table 7- 7 Effect of drainage*, based on Level-B analysis

 

Performance

Effect Comments
Measure

 

0 At 5 of the 7 sites with distresses sections, among
sections with 203 mm PCC slab, sections without
drainage cracked more than sections with

Consistent drainage.

effect 0 Among sections with 203 mm PCC slab, 18% and

(p=0.034) 43% of sections with drainage and without
drainage have exhibited transverse cracking.

o Cracking was observed only at two sites in the
thicker slab sections.

0 No discernable trends were observed for
longitudinal cracking.

Longitudinal I . 0 Among sections with 203 mm PCC slab, 25% of

. nconclusrve . . . .

crackrng sectrons wrth drarnage and 25% of sectrons
without drainage have exhibited longitudinal
cracking.

0 Effect of drainage seems to be negligible as the
sections with drainage and without drainage have

Slight effect performed similarly in most of the sites.

(P=0-076) 0 Among sections with 203 mm PCC slab, the

average roughness of sections with drainage and

without drainage are 1.3 and 1.6 m/km.

*Effect of drainage is a result of comparison between sections built on DGAB and sections built on

PATB only.

Transverse
cracking

 

 

Roughness (IRI)

 

 

 

 

 

On average, among sections with 203 mm PCC slab, those built on LCB have
exhibited higher transverse cracking than other sections (see Table 7-8). This trend was
observed in a majority of sites. Among sections with 203 mm slab and standard slab
width, the trend is consistent (p=0.001) across the sites. This effect may be an
“interaction effect”, as the effect of base type was discernable among sections with 203

mm slab and standard slab, and not in sections with 279 mm slab.

83

The effect of base type on longitudinal cracking (see Table 7-8) is consistent
among sections with 203 mm slab, in that sections built on LCB have higher cracking

than those on other bases. The effect of base type on roughness is inconclusive.

Table 7- 8 Effect of base t on cracking, based on Level B analysis

 

 

 

 

 

 

 

 

Performance Effect Comments
Measure
0 In 9 sites, among thinner slab sections, sections
. built on LCB cracked more than sections on
T Cons1stent
ransverse effect other base types.
cracking ( <0 05) o 43%, 64% and 18% of sections on DGAB, LCB
p ' and PATB have exhibited cracking among
thinner slab sections.
0 In 8 sites, of the 12 sites at which cracking was
- observed, sections built on LCB exhibited more
Longitudinal CfgnSlStem cracking sections built on other base types.
cracking 6 ea 0 Among the thinner slab sections, 54% of
(p<0.05) sections built on LCB and 25% of other sections
exhibited cracking.
Roughness (IRI) Inconclusive 0 Effect of base type seems to be neglrgrble as all
the sectrons performed srmrlarly, 1n general.

 

PC C slab thickness

The performance of sections with target PCC slab thickness of 203 mm (8-in) was
compared with that of sections with target PCC slab thickness of 279 mm (1 l-in) by
controlling the effects of base type and PCC flexural strength. The effect of PCC slab
thickness was consistent on cracking (transverse and longitudinal), whereas no noticeable
effect was found on roughness.

The effect of PCC slab thickness is consistent, across sites, on transverse
cracking. Among sections built with DGAB or LCB, sections with 203 mm slab had
higher cracking than sections with 279 mm slab. Table 7-9 is the summary of effects of
PCC slab thickness. In the case of longitudinal cracking, the effect was found consistent
among sections built with LCB with higher cracking in sections with 203 mm slab. The

effect of PCC slab thickness on roughness is inconclusive.

84

 

Table 7- 9 Effect of slab thickness on cracking, based on Level B analysis

 

Performance

 

 

 

 

 

 

 

employed for analyses.

The performance measures that were analyzed to investigate the impact of design

Transverse cracking,

Longitudinal cracking, and

Measure Effect Comments
Transverse Consistent Thinner slab sections exhibit more transverse
crackin effect cracking than thicker slab sections, among sections
g (p<0-05) with DGAB or LCB.
Longitudinal CfcfmSIStem Thinner slab sections exhibit more longitudinal
cracking :3 IICIIS) cracking than thicker slab, among sections with LCB.
R< .
Rou hness (IRI) Inconclusive Effect of base type seems to be negligible as all the
g sections perform similarly
7.3. OVERALL ANALYSIS

The results obtained from statistical analyses performed on SPS-2 data are presented in
this section. Analyses were performed on data from all sites and are thus referred to as
‘overall’ analyses. Linear Discriminant Analysis (LDA), Binary Logistic Regression

(BLR), and Analysis of Variance (ANOVA) are the statistical methods that were

and site factors on rigid pavement performance are as follows:

Roughness (IRI)- initial roughness and change in roughness.

Important factors affecting occurrence of transverse cracking, longitudinal

cracking, and roughness were determined using LDA and BLR, which are frequency-
based methods (details in Chapter 6). For analyses on cracking, the test sections were
grouped into 2 categories— sections with cracking, and sections without cracking. In the
case of IRI, a threshold of 1.5 m/km was used to separate the test sections as groups. The

value corresponds to the threshold between “normal” and “poor” pavements

85

 

(Khazanovich, Darter et al. 1998) for an age of 7.5 years, which is the average age of
SPS-2 test sections. Analyses were then performed to identify factors that significantly
discriminate the groups.

As mentioned in Chapter 6, ANOVA was used to determine the significance of
factors impacting initial roughness (IRI), and change in roughness (first survey to latest
survey). ANOVA method could not be applied to transverse and longitudinal cracking as
the assumption of constant variance of residuals was violated owing to occurrence of
cracking in less than 30% of test sections. Wherever required, a natural logarithmic
transformation of the variable of interest was applied to satisfy the assumptions of
ANOVA. Traffic volume and age of test sections were considered as covariates in all
analyses to adjust for the difference in traffic loading and age among the sites in the
experiment.

The following is a summary of the main findings from each method of analysis.
Basic statistics pertaining to the extent of occurrence of distresses are presented along
with results to corroborate the results. It is suggested that the results be interpreted
keeping in view the extent of distresses that occurred in the test sections. In the
discussion of results, the word ‘significance’ needs to be interpreted as statistical
significance only.

As mentioned before, all analyses were conducted without including data from the
site in Nevada (32), as extensive distresses at the site are related to wide range of
construction issues that occurred at the site but not to pavement performance. Inclusion of

data from this site will affect results from analyses, significantly.

86

7.3.1. Frequency-based Methods

Two frequency-based methods were used- Linear Discriminant Analysis and Binary
Logistic Regression (details in Chapter 6). The results from these analyses are as follows:
Linear Discriminant Analysis

Based on this method all the distresses were analyzed using the following
thresholds for categorization of the sections.

0 Transverse or longitudinal cracking: Cracked versus non-cracked
0 Roughness: IRI (current)< 1.5 m/km versus IRI (current)>1.5 m/km
IRI (initial)<l.25 m/km versus IRI (initial)> 1.25 m/krn.

This analysis was aimed at identifying the experimental factors that best
discriminate between distressed and non-distressed pavement sections. As the pavements
in the SPS-2 experiment have not shown “high” levels of distress, this analysis will help
in finding the significant design and site factors contributing to the occurrence of
distresses (rather than magnitude), at this point in time. Traffic and pavement age, were
considered as covariates in this analysis. Interpretation of results from this analysis is
illustrated here using the results from LDA on transverse cracking. Based on the current

SPS-2 data, LDA yielded the following classification model for transverse cracking:

Score = —4.8 — 0.26 (BaseType) + 0.62 (PCCThickness) + 0.45 (Drainage)

—0.14(Zone) +0.24(LWidth) + 1.05(SG) + 0.15(FS)

Group Centroids: Transverse Crack ‘No‘= 0.51
Transverse Crack ‘Yes’= -1.47

(4)

Where:
BaseType: ‘1’ for DGAB or PATB, ‘2’ for LCB.
PCCThickness: ‘1’ for target slab thickness of 203 mm, ‘2’ for target slab thickness of 279 mm.
Drainage: ‘0’ for No Drainage, ‘1’ for Drainage.
Zone: ‘1’ for WF, ‘2’ for WNF, ‘3’ for DF, ‘4’ for DNF.
LWidth: ‘l’ for 3.7 m, ‘2’ for 4.3 m.
SG: ‘1’ for Fine-grained, ‘2’ for Coarse-grained.
FS: ‘1’ for target 14-day strength of 3.8 MPa, ‘2’ for target l4-day strength of 6.2 MPa.

87

In the above equation, the factors in bold are statistically significant in affecting
the occurrence of transverse cracking. The equation can be interpreted by considering the
group centroids and the particular effect each factor has on the score. For example, PCC
slab thickness has a positive effect on the score, i.e. an increase in PCC thickness from
203 mm (code: 1) to 279 mm (code: 2) will have a positive effect on the overall score.
This implies that increasing PCC slab thickness decreases the chances of cracking.

Similar results were obtained from all LDA analyses. However, for the sake of
brevity only interpretations of the effects are presented in the thesis. Below are the effects
of the design and site factors on pavement performance based on LDA.

Transverse cracking: The effect of drainage, base type, and target PCC thickness were
significant in discriminating between cracked or un-cracked sections. Table 7-10
summarizes the effects of the design and site factors on the occurrence of transverse
cracking.

In the WP zone, the effects of PCC thickness and subgrade soil type were
statistically significant in discriminating between sections with cracking and without
cracking. Table 7-11 summarizes effects of experimental factors based on the results of
LDA on transverse cracking.

It was observed that 33% of sections with thinner slab and 14% of sections with
thicker slab have exhibited cracking. Moreover, 14% of sections with thinner slab have
exhibited high severity cracking where as none of the thicker slab sections exhibited high
severity cracking. While 33% of sections built on fine-grained soils manifested transverse
cracking, 4% of sections built on coarse-grained exhibited cracking. Also, about 22% (13

of the 60 sections) of sections built on fine-grained subgrade soils have exhibited high

88

 

 

 

 

 

severity cracking while 4% (1 of the 24 sections) of the sections built on coarse-grained
soils exhibited cracking.

Longitudinal cracking: The effect of target PCC thickness, base type and the climatic
zone are significant in discriminating between cracked and un—cracked sections. Table 7-
12 summarizes the effect of the design and site factors on the occurrence of longitudinal
cracking. The effects of PCC thickness and base type, in WF zone, were statistically
significant in discriminating between sections with cracking and sections with no
cracking. Table 7-13 summarizes the effects of experimental factors based on the results
from LDA on longitudinal cracking, in WF zone. While 19% of sections with 203 mm
PCC slab have exhibited longitudinal cracking, 10% of sections with 279 mm PCC slab
have exhibited cracking. Also, 7%, 25%, and 11% of sections built on DGAB, LCB and
PATB, respectively, have exhibited longitudinal cracking.

Roughness: The initial and current roughness of the test sections were analyzed by
categorizing the variables using the thresholds mentioned above. Table 7-14 through
Table 7-17 are summaries of results from these analyses. PCC thickness was the only
factor that was found to be discriminating between “smooth” and “rough” pavement
sections based on the initial IRI categories. Based on LDA on current roughness,

drainage, PCC thickness, and base type were found to be the significant factors.

89

 

 

Table 7- 10 Summary of results from LDA on transverse cracking- Overall

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Factor Factor Effects on transverse cracking p-value
Category
Presence of drainage significantly
Drainage reduces the chances of occurrence of Yes (0.001)
cracking
Target PCC Thicker PCC th1ckness reduces the Yes (0.001)
th1ckness chances of occurrence of crackmg
The type of base significantly impacts
Design Base type the chances of the occurrence of Yes (0.044)
cracking
No significant effect. In general, the
Flexural Strength 900-psi mixes tend to mitigate No (0.716)
cracking.
. No 51 nificant effect. l4-foot wide slab
Slab WIdth sectiogns tend to inhibit cracking. No (0'467)
No significant effect. Designs
Climatic Zone constructed in Dry zones tend to crack No (0.147)
more.
Site No significant effect, however the
Subgrade soil model indicates that sections on fine No (0 538)
type subgrade soils tend to crack more than '
sections on coarse subgrade soils

 

 

Table 7- 11 Results from LDA on transverse cracking, WF zone

 

 

 

 

 

 

 

 

Factor Effects on transverse cracking p-value

Drainage No srgnrficant effect. Presence .of drarnage reduces the No (0.151)
chances of occurrence of cracking

Target PCC Sections wrth thicker PCC slabs crack srgnrficantly Yes (0.041)

th1ckness less

Base type astigmficant effect. Sections on LCB tend to crack No (0.214)

Flexural No significant effect. In general, the 900-psi rrrixes No (1 000)

Strength tend to crack more '

Slab Width No Significant effect. l4-foot wrde slab sectrons tend No (0.614)
to rnhrbrt crackmg

Sub rade soil Sections on fine subgrade soils tend to crack

typeg significantly more than sections on coarse subgrade Yes (0.007)
soils

 

 

 

90

 

 

Table 7- 12 Summary of results from LDA on longitudinal cracking

 

Factor

 

 

 

 

 

 

 

 

 

 

 

Factor Effects on longitudinal cracking p-value
category
Drainage No srgnrficant effect. Presence of drarnage 1ncreases No (0.180)
the chances of occurrence of crackrng.
Target PCC Thicker PCC th1ckness reduces the chances of Yes (0.000)
th1ckness occurrence of cracking
- The type of base significantly impacts the chances of
D
esrgn Base type the occurrence of crackig Yes (0'004)
Flexural No significant effect. In general, the 900-psi mixes No (0 834)
Strength tend to mitigate cracking. '
Slab Width No srgnrficant effect. 14-foot wrde slab sectrons tend No (0.834)
to have more crackrng.
Climatic . .
Zone Desrgns constructed in Dry zones tend to crack more. Yes (0.009)
Site Sub rade No significant effect, however the model indicates
. g that sections on fine subgrade soils tend to crack No (0.456)
501] type . .
more than sectrons on coarse subgrade sorls

 

Table 7- 13 Results from LDA on Longitudinal cracking, in WF zone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Factor Effects on transverse cracking p-value
Drainage No srgnrficant effect. Sectrons wrth drarnage have cracked more No (0.193)
than the ones wrthout drarnage
Target PCC Thicker PCC thickness significantly reduces the chances of
. . Yes (0.026)
th1ckness occurrence of crackfl
Base type Sections on LCB crack significantly more than other sections Yes (0.023)
Flexural No srgmficant effect. In general, the 900-psr mixes tend to crack No (1.000)
Strength more.
Slab Width No srgnrficant effect. l4-foot wrde slab sectrons tend to rnhrbrt No (0.463)
crackrng.
Sub a de No significant effect, the model indicates that sections on fine
. gr subgrade soils tend to crack more than sections on coarse No (0.296)
sorl type .
subgrade sorls
Table 7- 14 Summary of results from LDA on initial roughness
Factor Factor Effects on initial roughness p-value
categorL
Drainage Presence of drarnage decreases the chances of higher 1nrtral No (0.090)
roughness.
Target PCC Sections with thicker PCC slab have higher chances of being Yes
thickness built rougher than thinner ones. (0.054)
- No significant effect. Lesser roughness was observed on
D
esrgn Base type sections built with PATB. N0 (0329)
Flexural No significant effect. In general, the 900-psi mixes tend to
No (0.201)
Strength have more roughness.
Slab Width No srgnrficant effect. 14-feet wrde slabs tend to have more No (0.750)
roughness.
Climatic No significant effect. Designs constructed in Wet zones tend
No (0.232)
Site Zone to have more roughness.
Subgrade No significant effect Sections on fine subgrade soils tend to
. . . . No (0.342)
sorl type have more 1n1tral roughness.

 

 

 

 

91

 

Table 7- 15 Summary of results from LDA on initial roughness, in WF zone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Factor Effects on initial roughness p-value
Drainage No Significant effect. Presence of drarnage decreases the chances of No (0.356)
hrgher 1n1t1al roughness.
Target No significant effect. Sections with thicker PCC slab have higher chances
PCC . . . No (0.835)
. of berng built rougher than th1nner ones.
th1ckness
Base type No significant effect. Sections with PATB tend to be built smoother. No (0.594)
Flexural No significant effect. In general, the 900-psi mixes tend to have more
No (0.190)
Strength roughness.
Slab Width No significant effect. l2-feet wide slabs tend to have more roughness. No (0.190)
Subgrade No significant effect. Sections on fine subgrade soils tend to have more
. . . . No (0.071)
sorl type 1n1tral roughness.
Table 7- 16 Summary of results from LDA on roughness
Factor Factor Effects on current (latest) roughness p-value
category
. . . . . . . Yes
Drarnage Presence of drarnage rnhrbrts increase in roughness. (0.000)
Target . . . .
PCC Thrcker PCC th1ckness 1ncreases the chances of higher Yes
thickness roughness. (0.036)
D3518“ Base t Sections on DGAB and LCB have higher increase in Yes
ype roughness. (0. 01 7)
Flexural No significant effect. In general, the 900-psi mixes tend to No
Strength cause more roughness. (0.076)
Slab Width No srgnrficant effect. 12-foot wrde slabs tend to have lesser No
roughness. (0.873)
Climatic No significant effect. Designs constructed in Wet zones tend to No
Site Zone have more roughness. (0.588)
Subgrade No significant effect. Sections on fine subgrade soils have No
soil type more roughness. (0.317)
Table 7- 17 Summary of results from LDA on roughness, in WF zone
Factor Effects on current (latest) roughness p-value
Drainage Presence of drainage inhibits increase in roughness. Yes (0.008)
Target PCC No significant effect. Thicker PCC thickness decreases the
. . No (1.000)
th1ckness chances of higher roughness.
Base type Sections on DGAB have higher increase in roughness. Yes (0.032)
Flexural No significant effect. In general, the 900-psi mixes tend to
No (0.081)
Strengh cause more roughness.
Slab Width No Significant effect. 12-foot wrde slabs tend to have more No (0.193)
roughness.
Subgrade sorl No srgnrficant effect. Sectrons on fine subgrade sorls have Yes (0.004)
type more roughness.

 

 

 

 

92

 

 

 

 

 

 

Binary Logistic Regression (BLR)

The BLR model was used to model the probability of occurrence of distress for various
performance measures. Thresholds similar to the ones used for LDA were used to
categorize the test sections for this analysis. The results (see Table 7-18 and Table 7-19)
are summarized below. The value inside parenthesis is the odds ratio between the levels
of a factor. For example, an odds ratio of 3.6 for the effect of PCC slab thickness on
transverse cracking means that the chances of occurrence of cracking in sections with 203
mm slab are 3.6 times the chances of occurrence of cracking in sections with 279 mm
slab.

In the interpretation of the results, the p-value is of primary importance as it
indicates the statistical significance of the effect. When an effect is statistically
significant it can be said that the effect is not due to random error (i.e. a happenstance).
Hence, irrespective of the odds ratio, a higher p-value (>0.05) will make the odds ratio
less reliable.

Transverse Cracking: The BLR model for transverse cracking was significant with a p-

value of 0.000. In about 89% of the cases, the model correctly differentiates cracked

sections from non-cracked sections. Based on this analysis, the effects of significant

factors are as follows:

0 PCC slab thickness: Sections built with 203 mm PCC slab have significantly higher
probability of cracking than the ones built with 279 mm PCC slab.

0 Base Type: Sections built on PATB have significantly lesser likelihood of cracking

than those built on LCB.

93

 

 

 

0 Subgrade: Sections built on fine subgrade soils have significantly higher probability
of cracking than the ones built on coarse subgrade soils.
Based on the BLR on data from sections in WF zone, the effect subgrade soil type
(0.029 in BLR) was statistically significant in discriminating between sections with
cracking and sections without cracking.
Longitudinal Cracking: The BLR model for longitudinal cracking was significant with a
p-value of 0.000. For about 89% of the times, the model correctly differentiates cracked
sections from un-cracked sections. Based on this analysis, the following factors
significant affect occurrence of longitudinal cracking.
0 Base Type: Sections on LCB have significantly higher chances of cracking compared
to the sections with DGAB.
0 PCC slab thickness: Sections with 203 mm slab have significantly higher chances of
cracking than the sections with 279 mm slab.
0 Climate: Sections in Dry No Freeze have significantly higher chances of cracking

than Wet Freeze.

Roughness: The BLR model for initial roughness was significant with a p—value of 0.023.

Moreover, 61.5% of the times, the model correctly differentiates sections with “poor”

roughness from other sections. Based on this analysis, the following conclusions were

made:

0 Base Type: Sections built on LCB have significantly (p-value=0.040) higher
probability of roughness than the ones built on PATB.

0 PCC slab thickness: Sections built with thinner slab have significantly (p-

value=0.038) lesser probability of being built rougher than those thicker slab.

94

 

The BLR model for latest roughness was significant with a p-value of 0.000. The model
correctly differentiates sections with “poor” roughness from other sections, 78.8% of the
times. Based on this analysis, base type is the significant factor for roughness. Sections
built on DGAB or LCB have significantly (p-value=0.007) higher probability of
roughness than the ones built on PATB. This may also be interpreted that sections with
drainage have significantly lesser chances of becoming rougher compared to sections

without drainage (sections on DGAB).

Table 7- 18 Summary of p-values from BLR for determining the effect of
experimental factors on pavement performance measures- Overall

 

 

 

 

 

Experimental Transverse Longitudinal Rou :hness
Factors cracking cracking Initial Current
Drainage . 0.083 (3.4) 0.616 (0.68) 0.28 (1.6) 0.003 (7.5)
Base type 0.073 (4.9)* 0.099 (2.7) 0.12 (2.4) 0.007

 

PCC thickness 0.019 (3.6) 0.001 (8.7) 0.018 (0.49) 0.381 (0.67)
Flexural Strength 0.550 (1.4) 0.825 (1.13) 0.17 (0.63) 0.077 (0.44)

 

 

 

 

 

 

 

 

 

 

Slab Width 0.389 (1.6) 0.800 (1.15) 0.687 (0.87) 0.876 (0.931)
Subgrade type 0.003 (9.5) 0.361 ( 1.98) 0.186 ( 1.73) 0.283 (1.97)
Climatic Zone 0.571 0.017 (18) 0.400 0.262

 

*LCB vs. PATB.
*Values in parenthesis are odds ratios.

Table 7- 19 Summary of p-values from BLR for determining the effect of experimental
factors on pavement performance measures- WF Zone

 

 

 

 

Experimental Transverse Longitudinal Rou ghness
Factors cracking cracking Initial Current
Drainage 0.736 (0.74) 0.55 (0.35) 0.510 (1.5) 0.004 (47.4)
Base type 0.767 0.237 0.383 0.014

 

PCC thickness 0.650 ( 1.43) 0.084 (11.3) 0.058 (.039) 0.066 (5.05)
Flexural Strength 0.643 (0.70) 0.868 (0.84) 0.141 (0.48) 0.222 (0.37)
Slab Width 0.593 (1.49) 0.234 (4.65) 0.147 (2.1) 0.655 (1.39)

Subgrade type 0.029 ( 19.5) 0.488 (0.23) 0.066 (2.97) 0.368 (3.009)
*Values in parenthesis are odds ratios.

 

 

 

 

 

 

 

 

 

 

95

7.3.2. Analysis of Variance

ANOVA was performed on roughness of sections in the SPS-2 experiment. The
analyses were performed combining data from all sections (overall) in the experiment.
ANOVA was also conducted on data for sections within each zone. It should be noted
that analysis that were conducted only within WF zone is presented. The other zones
have two sites each and this sample size did not yield meaningful results because of less
statistical “power”. The effects of design factors and site factors on pavement
performance are discussed next.

Effects of design factors on pavement performance

The results from this analysis are summarized in Table 7-20. Table 7-21 shows marginal
means corresponding to different levels in each factor. The marginal means were back
transformed using the properties of lognorrnal distribution based on the below formula

(Haider, 2005).

The most important design factor is the base type, which has a significant effect on initial
roughness (IRIO) and change in roughness (AIRI). In addition to base type, IRI0 is
affected by PCC slab thickness, and AIRI is affected by drainage and PCC thickness
(slight effect). The following discussion summarizes significant effects of design factors
on pavement performance:

0 Effect of drainage: Pavement sections with drainage have shown significantly

lower change in roughness than those without drainage.
0 Effect of base type: Pavement sections with DGAB have shown the highest

change in roughness while those with PATB have shown the least change in

96

 

roughness. Sections built with LCB had the highest initial roughness while other
sections had comparable initial roughness.

Effect of PCC thickness: Significantly higher initial roughness was observed on
sections with 279 mm PCC slab compared to sections with 203 mm PCC slab.
However, the change in roughness was slightly higher in sections with 203 mm
PCC slab compared to those with 279 mm PCC slab.

Effect of flexural strength: No significant effect of PCC flexural strength was
found on roughness, at this point.

Effect of slab width: No significant effect of slab width was found on roughness.

Similar ANOVA was performed on data from sections in the WP zone and Table 7-22 is

the summary of results from the analysis. Table 7-23 shows the back transformed

marginal means for all levels of design factors. As mentioned before, the results from

analysis on data from other climatic zones are not presented, as limited amount of data is

available for those zones. The following discussion summarizes significant effects of

design factors on performance of sections in the WP zone:

Effect of drainage: Pavement sections with drainage have shown significantly
lower change in roughness than those without drainage.

Effect of base type: Pavement sections with LCB have shown significantly higher
change in roughness compared to those with PATB and with DGAB. Sections
with PATB have shown the least change in roughness. Sections built with LCB
were found to have the higher initial roughness compared to other sections, which

have comparable initial roughness.

97

0 Effect of PCC thickness: Significantly higher initial roughness was observed on
sections with thicker PCC slab compared to sections with thinner PCC slab.
However, the change in roughness was significantly higher in sections with
thinner PCC slab compared to those with thicker PCC slab.

0 Effect of flexural strength: Significant effect of PCC flexural strength was found
on initial roughness. Sections with higher strength concrete had higher initial
roughness compared to those with lower strength concrete.

0 Effect of slab width: Slight effect of slab width was observed for initial

roughness. Sections with standard slab width were constructed rougher.

Table 7- 20 Summary of p-values from ANOVA for determining the effects of design
factors on pavement performance— Overall

 

 

 

 

 

 

 

 

 

 

 

 

Roughness (IRI)

Factor AIRI IRI,

Drainage 0.002 0.373
PCC thickness 0.094 0.003
Base Type 0.009 0.037
PCC Flexural Strength 0.544 0.246
Slab Width 0.860 0.313
Site (blocked) 0.006 0.000
N=114 N =156

18:03.05 R2=0.344

 

 

Table 7- 21 Summary of marginal means from ANOVA for determining the effect of
main design factors on pavement performance measures—Overall

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Roughness (IRI)
Desrgn Factor AIRI, m/km IRIO, m/km

Drainage No 0.34 1 .33
Yes 0.14 1.29

. 8” 0.32 1.25

PCC thickness 1 1,, 0.22 1.36
DGAB 0.39 1.29

Base Type LCB 0.30 1.37
PATB 0.16 1.25

PCC Flexural 550 0.29 1.28
Strength 900 0.24 1 .32

. 12’ 0.26 1.32

Slab mm“ 14’ 0.27 1.28
MSE 1.343 0.032

 

98

 

Table 7- 22 Summary of p-values from ANOVA for determining the effects of design

factors on pavement performance— WF Zone

 

 

 

 

 

 

 

 

 

 

Factor Roughness (IRI)

AIRI, m/km IRIi, m/km
Drainage 0.012 0.688
PCC thickness 0.033 0.057
Base Type 0.004 0.557
PCC Flexural Strength 0.650 0.009
Slab Width 0.987 0.071
Site (blocked) 0.000 0.000
N=58 N=84

R2=0.505 R2=0.426

 

 

 

 

Table 7- 23 Summary of marginal means from ANOVA for determining the effect of
main design factors on pavement performance measures—WP Zone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Design Factor AIRI, mRn‘omughnCSS ([13:11: rn/km
Draina e N0 030 1.32
g Yes 0.13 1-29
_ 3” 0.29 1.26
PCC thickness 11,, 0.16 1,36
DGAB 0.26 1.30
Base Type LCB 0.34 1-34
PATB 0.1 l 1.27
550 0.20 1.24
PCC Flexural Strength 900 0.23 1, 33
. 12’ 0.22 1.35
MSE 0.947 0.033

 

 

 

 

 

Effect of Site Factors on Pavement Performance

Given the unbalance of the experiment, a one-way AN OVA was performed to study the
main effects of the subgrade soil type (fine-grained versus coarse-grained soils) and
climatic zone (wet versus dry, freeze versus no-freeze), one at a time. The p-values and
mean performances from these analyses are summarized in Table 7-34 and Table 7-25,
respectively. To indicate the direction of effects, “+” and “-“ signs are reported along
with p-values. A “+” indicates that, within a factor, the first level has exhibited more
distress than the second, while a “-“ indicates otherwise. For example, in the case of the

effect of subgrade on roughness, “+” indicates higher roughness in pavements

99

constructed on fine-grained soils compared to those constructed on coarse-grained soils,
as fine-grained soil is the first level and coarse-grained soil is the second.

The p-values indicate that subgrade type appears to be significant in affecting the
initial roughness. Pavements built on fine-grained subgrade soil showed higher initial
roughness than those constructed on coarse-grained subgrade soil. Climate has a
significant effect on initial roughness. Pavements constructed in “wet” climate were
found to have slightly higher initial roughness compared to those in “dry” climate.

Table 7- 24 Summary of p-values from one-way ANOVA for determining the effect
of site factors on pavement performance measures

 

 

 

 

 

 

 

Site Factor Roughness (IRI)
AIRI, m/km IRIi, m/km

Slim

Fine-grained vs. Coarse-grained 0920 (+) 0050 (+)

Climate

Wet vs. Dry 0406 H 0052 ('1‘)

Freeze vs. No Freeze 0632 H 0317 H

Wet Freeze vs. Wet No Freeze ”300 (+) 1000 H

Dry Freeze vs. Dry No Freeze 1000 (~) 1-000 (-)

 

 

Table 7- 25 Summary of marginal means from one-way ANOVA for determining the
effect of site factors on pavement performance measures

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Roughness (IRI)
Site Factor
AIRI, m/km IRIi, m/km

Fine- rained 0.37 1.33

Subgrade Coarsi- grained 0. 38 1 .25
Wet 0.35 1.33
Dry 0.42 1.25
Freeze 0. 36 1.29

Climate No Freeze 0.41 1.33
Wet Freeze 0.36 1.31
Wet No Freeze 0.33 1.38
Dry Freeze 0.38 1.21
Dry No Freeze 0.50 1.29

 

 

 

Effect of Design Factors on Pavement Performance based on standard deviates
As explained before in Chapter 6, the experiment design and the performance of the test
sections have rendered the SPS-2 experiment unbalanced. Most of the sites, 9 of the 14,

in the experiment were constructed in “wet” climate of which 7 are in the WP zone. Also,

100

all 24 unique designs were not built in every soil-climate combination. Furthermore, non-
occurrence of distresses in a considerable number of sections has also contributed to the
unbalance. In light of these concerns, a simplified analysis considering one design factor
at a time (univariate) was performed, as in the case of analysis of the effects of site
factors.

The performance of similar designs was not found to be consistent across sites
indicating a considerable influence of the site conditions at each site. The site conditions
that could have contributed to this variation in performance are traffic, age, construction
quality, measurement variability, and material properties, apart from site factors (i.e.
subgrade and environment). In order to separate the “true” effects of the experimental
factors, this “noise” should be nullified.

The standard deviate indicates the relative performance of a design with respect to
the other designs. As this measure was calculated for each section with respect to other
sections in the same site, it indicates the relative standing of the section compared to
other sections in that site. It thus helps nullify the variation in performance (due to site
conditions) across sites, as the sections are weighed with respect to companion sections in
each site. This approach of using the standard deviate is similar to statistical blocking, for
nullifying the effects due to site conditions.

One-way (univariate) analysis of variance was performed on the standard deviates
of the sections to study the effects of design factors, by considering one design factor at a
time. As the SPS-2 experiment design calls for study of the effects of design factors at
different site conditions, the univariate analysis was performed accordingly. The analyses

were performed on data from all sections and also on subsets of data corresponding to

101

different subgrade types, climates and combination of these. This helped in identification

of the effects of design factors under different site conditions.

To study the “pure” effect of each design factor, comparisons of standard deviates
were also performed by controlling for other factors, as in the case of level-B analyses
(site-level). As said earlier, these comparisons were performed only on roughness. The
method is not appropriate for cracking because of “low” occurrence of the distresses at
this point in time. The effects of design factors, based on the above-mentioned analyses,
on roughness are discussed next.

The effects of the design and site factors, in terms of standard deviate, are shown
in Figure 7-7. The change in roughness was considered as the performance measure for
analyses. A summary of p-values corresponding to these analyses is Table 7-26. The
mean change in roughness corresponding to each comparison presented in Table 7-26 is
shown in Table 7-27.

The effects of the design factors on change in roughness, based on this analysis, are

presented below:

0 Drainage: On the whole, the effect of drainage is statistically significant. Un-drained
sections have exhibited higher change in roughness than drained sections. This effect
is consistent in all sections, irrespective of subgrade soil type. In addition, the effect is
more prominent in “wet” climate. Among the sections located in WF zone, those built
on fine-grained soils have shown greater change in roughness compared to those built
on coarse-grained soils. This effect is marginally significant.

0 Base Type: As the effect of base type is confounded with the effect of drainage

because of the SPS-2 experiment design, the results presented here are from

102

comparisons between the performance of sections built with DGAB and sections built
with LCB, which are both un-drained. On average, sections with LCB have shown
lower change in roughness than sections with DGAB. This effect is not statistically
significant. A significant effect of base type was observed among sections located in
WNF zone, in that sections with DGAB have shown higher change in roughness than
those with LCB. However, this effect should be interpreted with caution, as only two
sites are located in WNF zone.

PCC slab thickness: The effect of PCC slab thickness is significant among sections
built on fine-grained soils, especially when constructed in WF zone. Sections with
203 mm slab have shown higher change in roughness than those with 279 mm slab.
PCC flexural strength: The effect of PCC flexural strength was not found to be
significantly affecting roughness of the SPS-2 sections, at this point in time.

Slab width: The effect of slab width was not found to be significantly affecting
roughness of the SPS-2 sections. On average, among sections built on fine-grained
soils and located in WF zone, those with standard slab width have shown higher

change in roughness than those with wider slab.

7.3.3. Comparison of Designs

The standard deviate of each section in the experiment was used to rank the designs

based on their performance. For each pavement section there is a standard deviate

corresponding to each of the performance measures, with respect to other sections in the

same site. For each design, the standard deviates of corresponding sections at different

sites were averaged to obtain the standard deviate, for each of the performance measures.

The “best” designs were identified for each performance measure. Combining the

103

performance of a design corresponding to different performance measures was
considered inappropriate.

The designs that have an ID among 0201 through 0212 cannot be compared to a
design with an ID among 0213 through 0224, as only one of these two sets of designs are
present at a site. In other words, a site will either have designs from 0201 through 0212,
or, 0213 through 0224. 80, “best” designs were identified from each of the sets of

designs. The design features of all the designs are given in Table 7-39 and Table 7—40.

104

 

 
  

 

 

 

 

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.60 _
r
0.40 E
8 :
Cd _
'S 0.20 :
0 _
Q :
g 0.00 _
5 -0 20 5
o . _
>2 5
-0.40 p
-0.60 I
£6
a D g a a E E g g E 5.
Design Factors
(a) Overall
0.60 :
o 0. 40 : DCoarse
.E
> 0.20 _
5 .
'3' 0.00 _
m 5
g -020 E
2 -040
-0.60
O m m E “1 w E E
g g 3 g E E E E h m
8 9* 8 1‘3 oo o; m V
N N m \0
Design Factors
(b) By subgrade type
0.60 :
o E
a; 0.40 :
'5 020 5 —
Q ' E
'90.“): I I IDIDI I l I I IDIUI IDILJ
m 5
g -0.20 : _ _‘
2 -040
-0.60 I
m m E es «1
E Q 3 o 1— E E E E.
O .—l <
D 9* 8 1‘3 00 N m V
N N m \0’
Design Factors
(c) Wet Freeze (WF) zone

Figure 7- 6 Effect of design factors on change in IRI

105

Hugo q- ma 955:5. 0m 6-55.9 men 8831.33 o». «:5an aninml 05250 5 Sam—50mm

 

w «:6 .. wan w 0:830 Noun m 25 man man scan

 

 

 

 

 

 

Guam: . .

DEE—mo <m.

DRE» 0 .
m 20-635» ..n

 

 

wmmo

So w>aweo>w
100
95303

 

 

 

 

won
38:5:
man: 9

 

 

ob: 9.30

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

208“ magma 85 5988 £339: «mm—#5858 m: oofic 9 3530132 cm 83538 GAPS.

.565 q- no magma. on So»: arm—fin m: 3.553%.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cami: meg? 00:62.50: O<nB= _I ma _ N030 d $9. 32% _ Um
m 0 as“ g Um UZT. m 0 m 0 m
. 75 o.— o.—
CHE—Ema
U o o;
002w ob o.» o;
mama J60 row ob o.» o; . ., , ._ m.
w>Hw o PM o o; o; o
won flaw "En—80mm new 55 o.» o.» o.» oh oh o; c; on o;
.30 BB 0.» oh oh o; Pm o; 0 ob o.—
wb 2:...» oh oh oh PM o; o; oh oh o 9w o;
v00 23:3. asap...
ob Kw» PM oh oh oh oh oh ob o0 c; on o.—
. mg 3 ob ob PM oh. 0; o; 9w o... o; 9— ob o.—
mEU 253
Pu 3 oh 0.» ob oh PM o; 0.» ob _ o; _ 0; 9w 0; o.»

 

 

 

106

Table 7- 28 Design features of designs 0201 through 0212

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Design Features
14-
Design ID , Tar?“ PCC Slab Width neiflffimfl of
Drainage Base type thickness P C C

(mm) (m) (MPa)
0201 No DGAB 203 3.7 3.8
0202 No DGAB 203 4.3 6.2
0203 No DGAB 279 4.3 3.8
0204 No DGAB 279 3.7 6.2
0205 No LCB 203 3.7 3.8
0206 No LCB 203 4.3 6.2
0207 No LCB 279 4.3 3.8
0208 No LCB 279 3.7 6.2
0209 Yes PATB 203 3.7 3.8
0210 Yes PATB 203 4.3 6.2
021 1 Yes PATB 279 4.3 3.8
0212 Yes PATB 279 3.7 6.2

 

 

 

 

 

 

 

 

Table 7- 29 Design features of designs 0213 through 0224

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Design Features
Tar et PCC . Tar et l4-da flexural
SHRP-ID Drainage Base type thigckness Slab Width sgtrength of PCC
(turn) (m) (MPa)
0213 No DGAB 203 4.3 3.8
0214 No DGAB 279 3.7 6.2
0215 No DGAB 279 3.7 3.8
0216 No DGAB 203 4.3 6.2
0217 No LCB 203 4.3 3.8
0218 No LCB 203 3.7 6.2
0219 No LCB 279 3.7 3.8
0220 No LCB 279 4.3 6.2
0221 Yes PATB 203 4.3 3.8
0222 Yes PATB 203 3.7 6.2
0223 Yes PATB 279 3.7 3.8
0224 Yes PATB 279 4.3 6.2

 

 

 

 

 

 

 

 

Transverse Cracking Performance

Based on the transverse cracking performance the designs are ranked as shown in
Figure 7-9 and Figure 7-10. It is evident from the figures that the designs with 203 mm
(8-in) PCC slab and built on a LCB (i.e. 0205, 0206, 0217 and 0218) have shown the
“worst” (relative) performance, while the other sections have shown comparable

performance. This is an “interaction” effect of PCC slab thickness and base type. Also, in

107

general, sections with PATB have performed better than other sections, followed by

 

 

 

 

 

 

 

 

sections with DGAB.
16
14*
12. F—
10~ '
.3 8~
5 6‘ ‘5
Q. 4‘ I
3 2i ..
0T m ...;. ':-‘-: ... I. I 7 7 ' ' _ ' T i I
-4, __
-6
ES 8 8 3 8 8 S 8 8 ‘2 :2 5:!
N N N N N N N N N N N N
O O O O O O O O O O O O

 

Std. Deviate

 

 

 

 

-4

M V V3 V0 l‘ on O‘ O F‘ N M V

p-n —1 —l —- —n —1 cu N N N N N

N N N N N N N N N N

8 O O O O O 8 O O O O 0
Design 11)

Figure 7- 8 Comparison among designs 0213 through 0224
Longitudinal Cracking Performance
The results of comparison of designs with respect to longitudinal cracking
performance are shown in Figure 7-11 and Figure 7-12. As in the case of transverse
cracking performance, the sections with 203 mm PCC slab and built with LCB (i.e. 0205,

0206, 0217, and 0218) have shown the worst performance. Also, in general, sections with

108

PATB have shown better performance than other sections, followed by sections with

 

   

 

 

 

 

DGAB.
12
10—
8-
i 6-
6
a 4-
5 2“
0‘ T
-2- i
—4
888388888822
N N N N N N N N N N N N
O O O O O O O O O O O o
DesignID
Figure 7- 9 Comparison of designs 0201 through 0212
12
a
.s
5
a
9'
(I)

 

 

 

 

M V In 0 |’\ 00 Ch 0 '— N M Vt

— -—1 —! —t —n —1 —i N N N N N

N N N N N N N N N N N

O 8 C o O o O O O O O O
D “@— ID

Figure 7- 10 Comparison of designs 0213 through 0224
Change in Roughness

The change in roughness, from first survey to latest survey, was taken as the performance
measure for this analysis. As shown in Figure 7-13 and Figure 7-14, the change in
roughness was lesser in sections with PATB, in general. On the other hand, sections with

DGAB have slightly higher change in roughness compared to other sections.

109

 

Std. Deviate

 

 

 

 

 

 

 

 

 

.1 N m 3 m o h 00 O\ O '-* N
O O O O O O O O —c —a —a
N N N N N N N N N N N N
O O O O O O O O O O O 0
Design ID
Figure 7- 11 Comparison of designs 0201 through 0212
12
10 e
8 _.
6 1
.3 4 -
E
Q 2 i
3 0—
-2 ,
-4 ,.
-6 -.
-8
M V If) \0 I" 00 0‘ O —' N m V
_. .—¢ '— u—n v—d u-d 0—1 N N N N N
N N N N N N N N N N N N
O O O O O O O O O O O O
DesignID

Figure 7- 12 Comparison of designs 0213 through 0224
In summary, based on the comparison of performance of the designs, PATB sections
(0209 through 0212, and 0221 through 0224) are the “best” designs, especially in terms
of cracking performance and change in roughness. In addition, sections with LCB (0205
through 020208, and 0217 through 0220) have shown worse cracking performance.
Hence, the performance of the designs seems to be determined mainly by the type of

base.

110

7.4. SYNTHESIS OF RESULTS FROM ANALYSES

This section summarizes all the findings from various analyses performed on the
performance data of sections in the SPS-2 experiment. The methods employed in this
study were explained in Chapter 6 and the results obtained from these analyses were
presented above in this chapter.

As mentioned before, broadly two types of analyses (overall) were performed—
magnitude-based and frequency-based. AN OVA, which is a method for comparing
means, is the magnitude-based analysis. Binary Logistic Regression (BLR) and Linear
Discriminant Analysis (LDA) are frequency-based analyses, which give the likelihood of
occurrence and non-occurrence of distresses. The site-level analyses were used to
compare the performance of pavements within each site. The results from site-level
analysis were used to ascertain the consistency of the effects (of experimental factors)
across all sites.

The magnitude-based methods, though powerful, are more appropriate for
analyses of distresses, which have “fairly hi gh” occurrence (roughness) and magnitude.
On the other hand, the frequency—based methods are more suitable- when the occurrence
of a distress is considerable though the magnitude is low (for example, transverse
cracking, and longitudinal cracking).

An attempt has been made to ‘summarize’ the above said effects of design and
site features on the performance. The results were interpreted in light of the type of
analysis, and occurrence and extent of distress. AN OVA being the most “powerful”
among the methods was given higher importance for distresses with “good” occurrence
and/or extent (roughness). However, the results from this analysis suffer seriously in case

of limited (low occurrence of distress) and unbalanced data. Therefore, in these cases, the

111

effects of experimental features, mainly on occurrence of distresses, were investigated
using BLR and LDA. The results from site-level analyses (paired comparisons and
comparison of designs) and data exploration (extent of distresses) were then considered
to confirm the findings. Based on the site-level analyses the consistency of effects was
ascertained.

All results need to be interpreted in light of the experiment design, occurrence and
extent of distresses, and analyses methods used. A “weak” effect at this point in time may
become a “medium” or “strong” effect in the long term. Hence, all the conclusions are

based on “mid-term” performance of the ongoing SPS-2 experiment.

The SPS-2 experiment, entitled Strategic Study of Structural Factors for Rigid
Pavements, is one of nine special pavement studies in the LTPP program. The main
objectives of this experiment are to determine the relative influence and long-term
effectiveness of the structural factors affecting performance of jointed plain concrete
pavements (JPCP). These factors include PCC slab thickness, base type, in-pavement
drainage, PCC flexural strength and slab width. The key conclusions from this study are

summarized below.

7.4.1. Effects of structural factors for rigid pavements — SPS-2 experiment

The effects of design and site factors on the performance of SPS-Z pavements are
presented here. The structural factors include PCC slab thickness, base type, drainage,
PCC flexural strength and slab width. The experiment also includes studying the
secondary effects of site factors, namely subgrade type and climatic zones. The effects of
the experimental factors on each performance measure are discussed below, one

performance measure at a time.

112

Transverse cracking

The occurrence of transverse cracking among pavements with 203 mm (8-inch)
PCC slab thickness is higher than that among those with 279 mm (1 l-inch) PCC slab
thickness. Also, the occurrence of transverse cracking among pavements constructed with
LCB is higher than that among those with PATB/DGAB or with DGAB. Pavements with
PATB/DGAB have shown the “best” performance (least occurrence of cracking). These
effects of PCC thickness and base type are statistically significant, as suggested by the
frequency-based analyses. The analyses indicate a marginal effect of drainage on the
occurrence of transverse cracking. Sections without drainage have slightly higher
likelihood of cracking than sections with drainage.

On average, among sections built with LCB, those with 203 mm PCC slab have
higher occurrence of cracking than those with 279 mm PCC slab. It is important to
interpret these results in light of the construction issues, i.e. shrinkage cracking in LCB.

Pavements built on fine-grained soils have slightly higher chances for the
occurrence of transverse cracking than those built on coarse-grained soils. This effect was
found to be marginally significant.

Longitudinal Cracking

The occurrence of longitudinal cracking among pavements with 203 mm (8—inch)
PCC slab thickness is higher than among those with 279 mm (11 inch) PCC slab
thickness. Also, the occurrence of longitudinal cracking among pavements constructed
with LCB is higher than among those with PATB/DGAB or with DGAB. Pavements with

PATB/DGAB have shown the “best” performance (least occurrence of cracking). These

113

effects of PCC thickness and base type are statistically significant, as suggested by the
frequency-based analyses.

On average, among sections built with LCB, those with 203 mm PCC slab have
higher occurrence of cracking than those with 279 mm PCC slab. It is important to
interpret these results in light of the construction issues i.e. shrinkage cracking in LCB.
Faulting

The extent of faulting among the test sections of the SPS-2 experiment is low,
with joint faulting less than 1 mm in 62% of the sections. Only 33% of the sections have
0 to 20% of the joints that faulted more than 1.0 mm, and just 5% of the sections have
more than 20% of the joints that faulted more than 1.0 mm. A majority of SPS-2 sections
seem to be exhibiting “good” performance with respect to joint faulting, at this point in
time. This performance seems to be reasonable as the test sections are “young” and have
doweled joints at 4.6 m (15’) spacing. Therefore, the results at this point may only
indicate the initial trends/observations that may not be of much practical significance.

In general, pavements with standard slab width i.e. 3.7 m (12’), have shown
higher faulting than those with widened slab i.e. 4.3 m (14’) wide slab. However, the
effect may not be of practical significance because of the low occurrence of faulting.

It is important to note that according to the experiment design of SPS-2, sections
with 203 mm PCC slabs are built with dowels of 25.4 mm diameter, whereas sections
with 279 mm PCC slabs are built with dowels of 32 mm diameter. Hence, the effect of

dowel diameter and the effect of PCC slab thickness on faulting are not separable.

114

Roughness

Pavements without drainage have shown higher change in roughness than those
with drainage. Also, pavements constructed with PATB have shown lower change in IRI
(AIRI) compared to those with DGAB or LCB, while pavements with DGAB have the
highest change in roughness. These effects of drainage and base type are of statistical
significance, as suggested by magnitude-based methods.

Among pavements constructed with standard slab width, sections with DGAB
have shown higher AIRI than those with LCB or PATB. This effect is only marginally
significant. Among pavements built on fine-grained soils, those with 203 mm PCC slab
have higher AIRI than those with 279 mm PCC slab. This effect is more prominent
among sections located in WF zone. Also, the effect of drainage (i.e. sections with
PATB) is more prominent among sections located in WF zone and built with fine-grained
soils. Among sections located in WF zone and built on fine-grained soils, those with
drainage (i.e. sections with PATB) have shown lower AIRI compared to those without
drainage. These effects are of statistical significance.

The above results suggest that the change in roughness can be inhibited by
constructing pavements with PATB and drainage as compared to sections with DGAB or
LCB, especially in the case of pavements built on fine-grained soils. Also, among
pavements built on fine-grained soils, an increase in PCC slab thickness from 203 mm to
279 mm seems to help prevent an increase in pavement roughness.

A simplified summary of results from all analyses is given in Table 7-32. The
summary is only meant to give an overall assessment of the effects. The reader is strongly

recommended to peruse relevant portions of the thesis for a better understanding of all the

115

,9 ‘6

effects. It is important to note that a “strong , medium” or “weak” effect should only be
interpreted in terms of the difference in effects at the various levels of a factor. As an
example, a “strong” effect of PCC slab thickness and a “strong” effect of subgrade soil
type should not be interpreted as PCC slab thickness and subgrade type having the same
strength of effect.

A black circle indicates a “strong” effect; a grey circle indicates a “medium”
effect, and a white circle indicates a “weak” effect. Operational significance was
determined only for “strong” or “medium” effects. It should be noted that an effect can
be statistically significant (meaning that it is not a coincidence) but may not be

operationally/ practically significant, at this point in time.

116

Table 7- 30 Summary of effects of design and site factors for rigid Davements

De - Performance Measures
srgn . _

Factor Transverse Longitudinal
cracking cracking

 

 

Roughness

 

Drainage

 

PCC
thickness

 

Base type

 

Flexural
Strength

 

Lane Width

 

Climatic
Zone

 

Subgrade
type
Note: This table is solely for the purpose of summarizing some of the effects in a ‘simple’ format.

The reader is urged to read relevant text in the thesis for a better understanding.

 

 

 

 

 

 

OOOOOOO
OOOOOOO

 

 

 

 

 

Symbol Description
. Strong Effect
Medium Effect
0 Weak Effect

 

 

 

 

117

CHAPTER 8- ENVIRONMENTAL AND LOAD EFFECTS ON RIGID
PAVEMENTS

8.1. INTRODUCTION

The SPS-8 experiment of the LTPP program is aimed at studying the effects of
environment, in the absence of heavy loads, on pavement performance. Newly
constructed flexible and rigid pavements are included in the experiment. A detailed
description of the experiment design can be found elsewhere (Hanna, Tayabji. et al.
1994). Data from the rigid pavement sections from this experiment were used to
investigate the effects of environment and traffic on rigid pavement performance. While
the effects of environment were studied based on comparisons among the pavement
sections in the SPS-8 experiment, the effects of traffic were deduced from comparisons
between similar sections in SPS-8 and SPS-2 experiments. This chapter contains
observations regarding the effects of environment and traffic on rigid pavement
performance, based on the comparisons mentioned above. An introduction to the SPS—8
experiment is followed by details of the comparisons and observations based on the

comparisons.
8.2. DESCRIPTION OF THE SPS-8 EXPERIMENT

The SPS-8 experiment was designed to evaluate the effects of environment on
pavement performance, in the absence of heavy traffic loads. The study examines the
effects of climatic factors and subgrade type (frost-susceptible, expansive, fine, and
coarse) on pavement sections incorporating different designs of flexible and rigid

pavements, which are subjected to very limited traffic as measured by ESAL

118

accumulation. The study design stipulates that the traffic volume on the test sections be at
least 100 vehicles per day and not more than 10,000 ESALs in a year.

Table 8-1 shows the experiment design matrix for SPS-8. Rigid pavement test
sections consist of 203 mm (8”) and 279 mm (11”) doweled JPCP slabs on 152 mm (6”)
DGAB. These sections are identified as XX-0807 to XX-0812, where ‘XX’ is the site
code and the first two digits of the SHRP ID, i.e. ‘08’, stand for the SPS-8 experiment.
The sections with SHRP ID that ends with an odd number have a target PCC slab
thickness of 203 mm, while the others have a target PCC slab thickness of 279 mm. An
alphabet is introduced before the SHRP ID in case the section belongs to a second site
constructed in the same state.

Table 8- 1 SPS-8 Experiment Design

 

 

 

 

 

 

 

AC/PCC Base
WF WNF DF DNF

Pavement Thickness, Thickness,
yp" mm mm A F c A F C A F C A F C
Flexible 102 203 x x x x x x x x x x x x

178 306 x x x x x x x x x x x x
.. 203 x x x x x x x x x x x x
ngId 279 153 x x x x x x x x x x x x

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A: Active subgrade soil (either frost susceptible or swelling type relative to climatic zone)
F: Fine-grained subgrade soil
C: Coarse-grained subgrade soil

Each section is constructed as uniform as is practical over a length of 183 m (600
feet) to allow 152 m (500 feet) for monitoring purposes and 15 m (50 feet) at each end for
destructive testing. The guidelines also stipulate that an asphalt concrete, untied PCC, or
bituminous surface-treated aggregate shoulder must be constructed as part of the test
section. The shoulder should extend at least 0.9 m (3 feet) outside the edge of the travel
lane and should be constructed according to participating highway agency practice. The
concrete used for the surface layer has to have a target average l4-day flexural strength

of 3.8 MPa (550 psi). Moreover, no subsurface drainage is to be provided to the

119

pavements in the experiment. The other construction guidelines for the experiment are

the similar to the SPS-2 experiment (see Chapter 2).
8.3. CURRENT STATUS OF SPS-8 RIGID PAVEMENTS

Table 8-2 shows the current status of the SPS-8 experiment, as per Release 17 of
DataPave. While the minimum required number of rigid pavement sections to fulfill the
proposed experiment criteria is 24, only 14 rigid pavement sections are currently in the
experiment, distributed over 6 sites. There are two sites [Missouri (29) and Ohio (39)] in
Wet Freeze zone, two sites [Arkansas (5) and Texas (48)] in Wet N 0 Freeze zone, and
two sites [Colorado (8) and Washington (53)] in Dry Freeze zone. There are no sites in
the Dry No Freeze zone. An active subgrade soil can be coarse-grained or fine-grained. A
section that is ‘active’ is not categorized as ‘fine’ or ‘coarse’, but as ‘active’. Subgrade

soil type at a site was determined based on the data provided in the construction reports.

Table 8- 2 Distribution of rigid pavement sections in the SPS-8 experiment

 

 

 

 

 

 

Pavement AC/PCC Base WF WNF DF DNF

Type Thickness Thickness A F C A F C A F C A F C
. . 203 mm x x x x x x

Rigid 279 mm 152 mm x x x x x x

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Each ‘X’ indicates presence of one or more sections fulfilling the criteria of the cell

8.4. DATA AVAILABILITY

Table 8-3 is a summary of data availability for the rigid pavement sections in the SPS_8

experiment.
8.4.1. Construction Reports

As in the case of SPS—2 sites, the construction reports contain information about

the construction process, geometric layout, construction issues, etc. Construction reports

120

are available for all the six sites in the SPS-8 experiment. A summary of the site-specific

information can be found elsewhere (Chatti, Buch et al. 2005).
8.4.2. Climate Data

The data on climate at the SPS-8 sites are available from Automated Weather
Stations (AWS) and not from Virtual Weather Stations (VWS). This information was
used to calculate average annual temperature and precipitation at the sites. The
classification of the sites in different climatic zones was confirmed with the available

data.
8.4.3. Traffic Data

Table 8-4 is a summary of the traffic data available for the rigid pavement sections in
SPS-8 experiment. The traffic on the sections in Ohio (39) is higher than the stipulated
upper limit, which is 10,000 ESAL. Traffic data are also available from the construction
reports of the sites. A summary of traffic data obtained from construction reports is
Table 8-5. The AADT for sections in AR (5), MO (29), and WA (53) is below the lower

limit of 100 vehicles/day.
8.5. DESIGN VERSUS ACTUAL CONSTRUCTION REVIEW

Figure 8-1 and 8-2 show the PCC slab thickness deviations for the two thickness
levels (the horizontal lines indicate target values). Sections at MO and OH have slabs
with considerable deviation from target thickness. Similarly Figure 8-3 shows the
deviations in the base layer thickness (the horizontal line indicates target value). Section
48-A808, and the sections at the sites in Arkansas (5) and Washington (53) were built

with lesser than stipulated base thickness.

121

 

The experiment design stipulates that the target l4-day flexural strength of the
PCC slab concrete be 3.8 MPa (550 psi). Figure 8-4 (the horizontal line indicates target
value) is a summary of concrete test data for each section. PCC in all the sections met the

stipulated flexural strength. All sections except 8-0811 exceeded the target strength.
8.6. DISTRESS OCCURRENCE

The distresses in SPS-8 sections as of Release 17 have been summarized in Table
8-6. Faulting occurred in all the sections except the ones at the WA (53) site. In 12 of the
14 sections in the experiment, less than 40% of the joints have measurable faulting. In
half of the sections, 3 to 20% of the joints faulted more than 1.0 mm. Figure 8-5 shows
the distribution of IRI in SPS-8 sections. A detailed evaluation of distress in SPS-8

sections is presented in the next section.

122

l'fi'.’

Table 8- 3 Summary of data availability for SPS-8 experiment —Rigid pavements

 

 

 

 

 

 

 

 

 

Data Data Availability, % of
Data type .
category sectrons
Construction reports 100
Climatic data
Virtual Weather Station
Annual Temperature 0
Annual Precipitation 0
. . Automated Weather Station
Site location
information Monthly Temperature 7
Monthly precrprtatron 7
Traffic data*
Traffic Open date 0
Monitored 14
Estimated 14
Axle Distribution 0
Subgrade
Sieve analysis 100
Atterberg Limits 100
Classification 100
Lean Concrete Base
Compressive Strength 0
Materials data Portland Cement Concrete
PCC mix data 100
Flexural Strength 7 1
Compressive Strength 86
Split tensile Strength 86
Static modulus of Elasticity 86
CTE 0
Layer details
Type 100
Representative thickness 100
Dowel bar details
Pavement Diameter 0
Length 0
structure .
Spacrng 0
Shoulder information
Type 100
Width 100
Thickness 100
Profile data (IRI) 100
Monitoring" Distress data 100
Faulting data 100

 

*Monitored, Estimated, or Axle Distribution data is considered to be available for a site even if the data is
available only for one year.
“Data is said to be available for a section even if it is available for just one year.

123

Table 8- 4 Summary of available traffic data

 

 

 

 

 

 

 

 

 

Site ID SHRP ID Year Traffic (ESAL)

8 081 1 1997 1000 (Estimated)

8 081 1 1998 1000 (Estimated)

8 081 1 1999 1000 (Estimated)

8 0812 1997 1000 (Estimated)

8 0812 1998 1000 (Estimated)

8 0812 1999 1000 (Estimated)
39 0809 1997 66824 (Monitored)
39 0810 1997 693 17 (Monitored)

 

 

 

 

 

 

Table 8- 5 Summary of traffic data available from the construction reports

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Site ID 2-way AADT used to calculate design Design ESAL,
traffic, vehicles/ day KESAU yr
5 38 Not Available
8 2500 12.95
290800 50 96.5
29A800 118 Not Available
39 500 Not Available
48 Not Available 2.15
53 60 182.5
M
.2
5
.D
2
U
U
a.

 

 

O\ —4 l‘ I"
o —« o o
00 00 00 00
o o o 4;
:2» ob o'x '
N 84‘

Section ID

39-0809

 

48-A807
53-A809

Figure 8- 1 Thickness deviations in sections with target PCC thickness of 203 mm

124

 

 

 

£85.25 ass 08

ofiwoém

Section ID

Figure 8— 2 Thickness deviations in sections with target PCC thickness of 279 mm

 

o— w<-mm
mower?
wow<-wv
Bw<-wv
2 woém
aowcmm
mow<-mm
how<dm
mowodm
howoém
NH wo-w

_ Hmofi
3 M56
oowofi

 

0

mm

1

150

00
50
22

EE $35—25 3am

Section ID

Figure 8- 3 Deviation from target base thickness of 152 mm

 

0.
6
a

     

Hmcobm 35on 36-3

Section ID

day flexural strength of PCC

Figure 8- 4 Average 14

125

Table 8- 6 Distresses in SPS-8 sections

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Long. Trans.
Trans. Long. Corner Spalling Trans. Spall Scaling
Site ID SHRP ID Cracks Cracks Breaks Length Spalling Length No.
5 809 0 0 0 46.2 0 O 0
5 810 0 0 O 37.5 0 0 0
8 811 5 7.7 l 0 2 0.8 l
8 812 0 0 0 0 0 0 0
29 807 3 0.5 0 37.4 1 0.5 0
29 808 0 0 0 4.5 0 0 0
29 A807 0 0 0 0 0 0 O
29 A808 0 0 0 0 0 0 0
39 809 1 0 0 0.7 0 0 O
39 810 0 O O 78.8 0 0 O
48 A807 0 0 0 0 0 0 0
48 A808 0 0 0 0 0 0 0
53 A809 0 0 0 O 0 0 0
53 A810 0 0 0 0 l 0.4 0
10

a

“5

E

E

:3

2

<1.5 1.5-2.0 2.0-2.5 2.5-3.0 3.0-3.5 3.5-4.0
IRI range, m/km
Figure 8- 5 Current IRI in SPS-8 sections
8.7. ANALYSES AND RESULTS

As mentioned before, comparisons were performed among the SPS-8 sections and also
between similar sections in SPS-2 and SPS-8 experiments. The results from these
comparisons are presented in this section.

The SPS-8 sites are 4 to 9 years old, according to Release 17 of DataPave, with an

average age of 7 years. In light of the absence of heavy traffic, SPS-8 sections are not

126

expected to exhibit severe structural distress at this point in time. In line with this
expectation, low levels of distress were observed on the test sections.

No sign of pumping, blow-ups, map cracking or durability cracking were
observed on the test sections. Cracking (transverse and/or longitudinal), mostly of low or
medium severity, was observed in four sections. Spalling was found in eight sections and
most of the spalling occurred at longitudinal joints. Longitudinal spalling ranged from 1
m to 79 111 while transverse spalling was less than 1 m. Scaling and comer break occurred
on section 8-0811 only. Only the sections at the Colorado site, 8-0811 and 8-0812, have
polished aggregate of 302 and 214 m2, respectively. Severe construction issues (poor
climatic conditions during construction), as mentioned in the construction report (Nichols
Consulting Engineers 1998), could be the reason behind the outlying performance of the
test sections.

Keeping in view the number of sections constructed for SPS-8 experiment (14
rigid pavements in 6 sites) and the extent distresses at present, statistical analysis as in the
case of the SPS-2 experiment is not applicable. In future when enough distress occurs at
the SPS-8 sections, non-parametric tests may be applied, given that only 14 sections exist
in the experiment.

For this study, simple mean comparisons (only for sections that exhibited
distresses) were performed to identify the effects of environment and traffic on various
performance measures. The site-level and overall analyses, which are comparisons
among the SPS-8 sections, are followed by a comparison between the performance of
similar sections from SPS-2 and SPS-8 experiments. Some initial trends obtained from

these comparisons are reported below.

127

8.7.1. Site-Level Analysis

The site—level analysis of the SPS-8 rigid pavements was done based on the PI and
the relative performance concepts (see Chapter 6), as in the case of SPS-2 experiment. At
the site—level, roughness and transverse joint sealant damage were analyzed to investigate
the effects of the main site factors (climatic zone and subgrade type) on performance. It
was observed that the roughness of sections constructed in the dry freeze zone is lesser
than those constructed in the wet zones. Furthermore, sections in the wet freeze zone are
smoother than the ones in the wet no-freeze zone. Also, transverse joint sealant damage

appeared to be more prevalent in the wet zones as compared to the dry freeze zone.
8.7.2. Overall Analysis

The initial trends that show the effect of SPS-8 experimental factors on various
performance measures will be discussed in this section. These comparisons were carried
out only for those performance measures that were exhibited to some extent in the rigid
pavement sections.

Transverse Cracking: Figure 8-6 and 8-7 show the average transverse cracking length
by experimental factors. Only three of the fourteen sections have transverse cracking,
ranging from 1 to 5 cracks. Cracking was not observed in any of the pavements
constructed with thicker PCC slab and in any of the pavements constructed on coarse-
grained subgrade soil. Average transverse cracking was found to be more on sections
located in DF zone, which was contributed by section 0811 of site CO (8). The CO(8)
site is the oldest (10 years) in the experiment and “very poor” climatic conditions

prevailed during the construction (Nichols Consulting Engineers 1998).

128

Longitudinal Spalling: Figure 8-8 shows the average longitudinal spalling length by
experimental factors, for rigid pavements. Six of the fourteen sections have longitudinal
spalling ranging from 1 to 79 m with an average of 34 m. Spalling was not observed in
any of the pavements located in the DF zone and in any of the pavements constructed on
coarse—grained subgrade soil. Average spalling was found to be more in sections located
in WNF zone. Also pavements with thicker PCC slab have shown slightly more spalling
than those with thinner PCC slab. Among the pavements built on fine-grained soils,
those built on active soils have exhibited slightly more spalling than those built on non-
active subgrade soil (see Figure 8-9).

Wheel Path Joint Faulting: Average percent of joints that faulted more than 1 mm, by
experimental factors, is shown in Figure 8-10. Seven sections have faulting of more than
1.0 mm at one or more joints. Among these sections, on average, 8% of the joints faulted
more than 1 mm, with a range of 3 to 21% of joints. Average percentage of joints that
faulted more than 1 mm was found to be more on sections located in WNF zone. Also
pavements constructed on active subgrade soil have shown slightly more faulting than
those constructed on others. Among the pavements built on coarse-grained soils, those
built on active soils have exhibited more faulting than those built on non-active subgrade
soil (see Figure 8-11).

Roughness: The average roughness in the SPS-8 sections by experimental factors is
shown in Figure 8-12. The average initial IRI of the SPS-8 rigid pavement sections is 1.8
m/km, with a range of 1.0 to 3.6 m/km. The average change in IRI for rigid pavements is
0.1 m/km with a range of 0.0 to 0.7 m/km. Average roughness was found to be higher for

sections located in WNF zone. Also pavements constructed on active subgrade soil have

129

shown slightly higher IRI than those constructed on others. Among the pavements built
on coarse-grained soils, those built on active soils have exhibited higher IRI than those
built on non-active subgrade soil (see Figure 8-13). It may be noted that similar trends
were observed for faulting and roughness, which may suggest a cause-effect relation

among these performance measure.

 

 

 

 

 

 

 

 

 

 

 

 

 

E 20
25
32 15 r
U
:3
0 10 r
8
is
> 5 r
(0
“ 1 I
CU
:3 O 1 1 l 1 T l
' s: .54
3’ ‘E % “ U "‘ a a
a t— E.
Experimental Factors
Figure 8- 6 Average transverse cracking by experimental factors
10

00

.5

i5 3 ”

‘8

3 6 _

E ’e‘

> v

‘8 4 —

E

2° 2 ~

<2

0 T
F C
Subgrade Type

I Non-Active C] Active
Figure 8- 7 Average transverse cracking by subgrade type

130

NNbJLtJ-h
OUIOLIIO

Avg. Long. Spalling, m

CU!

Avg. Long. Spalling (m)

Avg. % of Joints with fault >1 mm

 

 

 

   

 

WF WNF DF A C F Thin Thick

Experimental Factors
Figure 8- 8 Average long. spalling by experimental factors

 

 

 

 

 

 

 

l

Subgrade Type

- I Non-Active El Active
Figure 8- 9 Average long. spalling by subgrade type

 

 

 

 

 

WF WNF DF A C F Thin Thick

Experimental Factors

Figure 8- 10 Faulting performance by experimental factors

131

 

Avg. % of Joints with fault >1
mm

0.50

Avg. IRI (mlkm) .
.9 .9 S3 9
~— to w A
o o o c

S

0.50
0.40
0.30
0.20

Avg. IRI (mlkm)

0.10
0.00

20

 

 

15-

 

 

 

 

 

,_-——j -

F C
Subgrade Type
I Non-Active Cl Active
Figure 8- 11 Faulting performance by subgrade type

 

 

 

 

JAM! I1

WF WNF DF Thin Thick

Experimental Factors
Figure 8- 12 Average IRI by experimental factors

 

 

 

 

 

 

 

 

Subgrade Type

I Non-Active Cl Active
Figure 8- 13 Average IRI by subgrade type

132

8.7.3. Comparison between Similar Designs of SPS-8 and SPS-2 Experiments

Owing to similarities in the design features of sections in the SPS-8 and SPS-2
experiments, comparisons can be performed among the similar sections to study the
effects of traffic on rigid pavement performance. The sections 0201, 0203, 0213 and 0215
of the SPS-2 experiment are similar in design (i.e. target PCC slab thickness, target PCC
flexural strength, base type and thickness, and drainage conditions) to the sections in the
SPS-8 experiment. Also some sites in the SPS-8 experiment are located in states that also
have SPS-2 sites. These states include Arkansas (5), Colorado (8), Ohio (39) and
Washington (53). In these states, the sections of SPS-2 those are similar in design to the
sections in SPS-8 are most similar because of identical climatic conditions.

In each climatic zone, similar sections from SPS-2 and SPS-8 experiments were
compared to identify the effect of traffic on pavement performance in that climatic zone.
The effect of traffic in each climatic zone can be compared to understand how
environmental conditions affect the influence of traffic on pavements. It is expected that
the SPS-2 pavement sections exhibit much higher distress as compared to those in the
SPS-8 experiment. The sections from Nevada were excluded from this exercise for the
same reason for which they were excluded from all analyses of SPS-2 data.

In general, the distress in SPS-8 sections is lesser than that of SPS-2 sections. The
low distresses exhibited by the SPS-8 sections and the setup of the SPS-8 experiment
preclude application of rigorous statistical analyses. Hence subjective evaluation was
performed based on the magnitude and occurrence of distress. Though faulting data for

all the sections is also available, statistical analysis is not expected to yield meaningful

133

results as considerable faulting (23.0 mm) occurred only in three sections (only at few
joints).

In the Dry Freeze zone, except for section 8-0811 in which many types of
distresses occurred, no major distress was exhibited in the SPS-8 sections. The distresses
in the section 8—0811 could thus be due to some issues other than environment alone. But
the comparable SPS-2 sections have exhibited wide range of distresses implying the
effect of traffic.

In the Wet Freeze zone, based on the comparisons between the sections of SPS-2
and SPS-8, which are from the same state, it can be said that similar distresses occurred
in the sections. But this could be due to traffic of 68000 ESAIJyr (approximately) on the
SPS-8 sections, which is much higher than the stipulated traffic of 10000 ESAIJyear.
From the comparisons between the SPS-2 and SPS-8 sections that are from Wet Freeze
zone, the effect of traffic appeared more clear as the SPS-2 sections exhibited a great
variety of distresses but the SPS-8 sections showed no distresses (except for transverse
cracking which could be due to higher than stipulated traffic). The SPS-2 sections also
showed considerable distress at the same age as that of the SPS-8 sections in Missouri.

In the Wet No Freeze zone, the SPS-2 sections exhibited a wide range of
distresses (including pumping) compared to the SPS—8 sections. Significant faulting (23.0
mm) occurred extensively in the SPS-2 sections but negligible number of joints faulted to
3.0 mm in SPS-8. The SPS-2 sections also showed considerable distress at the same age
as that of the SPS-8 section. From the faulting data it can be said that, the influence of
climate was more in the Wet No Freeze zone as, higher number of joints in SPS-2

sections faulted more than 3.0 mm compared to joints in SPS-8 sections.

134

8.8. SUMMARY OF RESULTS

Currently a total of 14 rigid pavement sections, in 6 sites, are present in the
experiment. There are 6, 4 and 4 pavement sections in WF, WNF, and DF climatic zones,
respectively. Three pavement sections were constructed on coarse-grained soils among
which 2 sections are on “active” and one section is on “non-active” soil. Also, 11
pavement sections were built on fine-grained soil, among which 4 sections are on
“active” soils and 7 sections are on “non-active” soils. These test sections have an
average age of about 7 years. The distresses are too “low” for any meaningful
conclusions to be made, at this point in time. Some observations based on initial trends,
as of Release 17.0, are as follows:

' Longitudinal spalling, on average, was noticed more in sections located in

“wet” climate.

' Spalling was not observed in any of the pavements located in the DF zone and
in any of the pavements constructed on coarse-grained soil.

' Transverse cracking was not observed in any of the pavements constructed with
thicker PCC slabs and in any of the pavements constructed on coarse-grained
subgrade soils.

Based on the comparison between similar sections in SPS-2 and SPS-8
experiment, the effect of traffic seems to be prominent. The effect seemed to be more
important in the case of sections situated in wet climates, especially in the Wet No-Freeze
zone. The results are in line with those observed in a study by Hudson and Flanagan

(Hudson and Flanagan 1987).

135

CHAPTER 9- CONCLUSIONS

This chapter contains a summary of findings from a comprehensive evaluation of the
SPS-2 experiment, based on data from Release 17 of DataPave. Findings from an
evaluation of rigid pavement sections in the SPS-8 experiment are also presented here. A
brief description of the experiments, their current status, and other information detailed in
preceding chapters are also discussed in this chapter. The conclusions are followed by
limitations of the findings and recommendations for future data collection and research.

A detailed description of the experiment design and the current status of the SPS-
2 experiment were presented in Chapter 2. The experiment is a fractional-factorial design
that is aimed at finding the relative influence of strategic design features on performance
of rigid pavements, in different site conditions (subgrade soil type and climate). The
design features include PCC slab thickness, base type, PCC flexural strength, drainage,
and slab width (sometimes referred to as lane width).

Each site within the SPS—2 experiment has 12 pavement test sections, each section
representing a different structural design. There are fourteen sites that are distributed
throughout the United States by climatic zones (wet-freeze, wet-no-freeze, dry-freeze and
dry-no-freeze) and subgrade type (fine and coarse-grained). A review of the current status
(details in Chapter 2) of the experiment indicated some deviations from the intended
experiment design.

A summary of data availability for the SPS-2 experiment was presented in
Chapter 4. The as-constructed details of design features and site factors obtained from the
LTPP database and construction reports were compared to the target values. The

deviations that were observed (in flexural strength, slab thickness, etc.) were reported in

136

 

the chapter. Key performance trends of the test sections were evaluated and the findings
were presented in Chapter 5. Depending on the extent of distress in the pavement
sections, statistical methods were selected for analyses. The analysis methods employed
for this study were detailed in Chapter 6.

All results obtained from analyses of performance of the SPS-2 sections were
presented and discussed in Chapter 7. The performance measures considered for the
analyses are transverse cracking, longitudinal cracking, and roughness [as expressed by

the international roughness index (IRI)].

ISL...

The average age of test sections in the experiment is 7.5 years with a range of 5 to
12 years. It may thus be said that the pavements are “fairly young”, and high occurrence
and levels of distress may not be expected at this point in time. Thus, all conclusions
from the analyses presented in this thesis should only be interpreted as “mid-term”
performance findings. However, the findings are expected to be valid for coming years as
early life performance has an important bearing on long-term performance.

The details of experiment design, analysis and findings of the SPS-8 experiment
were presented in Chapter 8. The experiment was designed to study effects of
environment on pavement performance, in the absence of heavy traffic. The experimental
factors include climate and subgrade soil type. A total of 14 rigid pavement sections at 6
sites were constructed for the experiment. The average age of the rigid pavement sections
in SPS-8 experiment is 7 years with a range of 4 to 10 years.

A majority of results from these analyses should be interpreted with caution in

light of the “low” occurrence of distresses in the test sections. It is suggested that the

137

conclusions be considered while keeping in view the limitations, as explained later in this

chapter.

9.1. EFFECTS OF STRUCTURAL FACTORS FOR RIGID
PAVEMENTS — SPS-2 EXPERIMENT

The SPS-2 experiment, entitled Strategic Study of Structural Factors for Rigid
Pavements, is one of nine Specific Pavement Studies in the LTPP program. The key
conclusions regarding the influence of the experimental factors, based on this study, are

summarized below.

It should be noted that the effects presented herein are statistically significant
unless mentioned otherwise; however, they may not be of practical significance at this
point in time. While statistical significance indicates whether an effect is a happenstance,

the practical significance of the effect matters next.

Transverse cracking: Among the SPS-2 sections, transverse cracking was observed in
31% of the sections. About 65% of the cracked sections have medium severity cracking,
and 37% of the cracked sections have high severity cracking. Most of the sections in the
experiment are thus exhibiting “good” performance. The effects of design and site
features on transverse cracking are as follows:
0 The occurrence of transverse cracking among pavements with 203 mm (8-inch)
PCC slab thickness is higher than among those with 279 mm (1 l-inch) PCC slab.
o The occurrence of transverse cracking among pavements constructed with LCB is
higher than that among those with PATB/DGAB or with DGAB. Pavements with
PATB/DGAB have shown the “best” performance (least occurrence of cracking).
0 Sections without drainage have slightly higher likelihood of cracking than

sections with drainage.

138

 

0 On average, among sections built with LCB, those with 203 mm PCC slab have
higher occurrence of cracking than those with 279 mm PCC slab. It is important
to interpret these results in light of the construction issues, i.e. shrinkage cracking
in LCB.

o Pavements built on fine-grained soils have slightly higher chances for the

occurrence of transverse cracking than those built on coarse-grained soils.

Longitudinal Cracking: Longitudinal cracking was observed in 27% of the test sections
of the SPS-2 experiment and most of the cracking, at this stage, is of low severity. As in
the case of transverse cracking, majority of test sections are exhibiting “good”
performance. The effects of design and site features on longitudinal cracking are as
follows:

0 The occurrence of longitudinal cracking among pavements with 203 mm PCC
slab is higher than among those with 279 mm PCC slab.

0 The occurrence of longitudinal cracking among pavements constructed with LCB
is higher than among those with PATB/DGAB or with DGAB. Pavements with
PATB/DGAB have shown the “best” performance (least occurrence of cracking).

0 On average, among sections built with LCB, those with 203 mm PCC slab have
higher occurrence of cracking than those with 279 mm PCC slab. It is important
to interpret these results in light of the construction issues i.e. shrinkage cracking

in LCB.

Faulting: A majority of SPS-2 sections are exhibiting “good” performance with respect

to joint faulting, at this point in time. Only 33% of the sections have 0 to 20% of the

139

 

joints that have faulting of more than 1.0 mm, and 5% of the sections have more than
20% of the joints that have faulting more than 1.0 mm. In addition, for LTPP data
collection the Georgia Faultmeter, which has a least count of 1.0 mm, is used to measure
faulting at joints and cracks. This least count is “large” (Selezneva et al, 2000), as typical
faulting values in SPS-2 pavements are close to 1.0 mm. In light of these issues, it is too
early to make reliable conclusions regarding the effects of design and/or site factors on
faulting.

Roughness: About 56% of the sections have current roughness less than 1.5 m/km, and
38% of the sections have current roughness between 1.5 and 2.0 m/krn. The effects of
design and site features on roughness are as follows:

0 Pavements constructed with PATB have shown lower change in IRI (AIRI)
compared to those with DGAB or LCB, while pavements with DGAB have the
highest change in roughness, especially in the case of sections with standard slab
width.

0 Among pavements built on fine-grained soils, those with 203 mm PCC slab have
higher AIRI than those with 279 mm PCC slab. This effect is more prominent
among sections located in WF zone.

0 Among sections located in WF zone and built on fine-grained soils, those with
drainage (i.e. sections with PATB) have shown lower AIRI compared to those

without drainage.

The above results suggest that the change in roughness can be inhibited by
constructing pavements with PATB and drainage as compared to sections with DGAB or

LCB, especially in the case of pavements built on fine-grained soils. Also, among

140

pavements built on fine-grained soils, an increase in PCC slab thickness from 203 mm to

279 mm seems to help prevent an increase in pavement roughness.

9.2 ENVIRONMENTAL AND LOAD EFFECTS ON RIGID
PAVEMENT PERFORMANCE

The SPS-8 experiment is entitled Strategic Study of Environmental Eflects in the Absence
of Heavy Loads for Flexible and Rigid Pavements. The study examines the effect of
climate and subgrade type (active, fine, and coarse) on pavement sections incorporating
different flexible and rigid pavements, which are subjected to very limited traffic as
measured by ESAL accumulation.

The SPS-8 pavements have “low” occurrence and extent of distresses, at this point.
Most of the pavements in the experiment are performing at comparable levels. The
observations presented here are just based on average performance of the distressed
pavements. The observations, presented below, need to be considered as initial trends in
light of these limitations.

Longitudinal spalling, on average, was higher in sections located in “wet” climate.
Spalling was not observed in any of the pavements located in the DF zone and in any of
the pavements constructed on coarse-grained subgrade soil. Transverse cracking was not
observed in any of the pavements constructed with thicker PCC slabs and in any of the
pavements constructed on coarse-grained subgrade soils.

Based on the comparison between similar sections in SPS-2 and SPS-8
experiment, the effect of traffic is clear. The effects seem to be more important in the

case of sections situated in wet climates, especially in the Wet No-Freeze zone.

141

El

9.3 LIMITATIONS OF THE EXPERIMENTS AND ANALYSES

All the above findings/observations on the effects of design and construction
features on pavement performance should be considered in light of the limitations
discussed herein. These limitations can be broadly classified under two categories——
experiment-related and data-related.

9.3.1 EXPERIMENT-RELATED ISSUES

0 The SPS-2 and SPS-8 experiments, which are fractional-factorial designs, were
rendered unbalanced because unequal numbers of sites were constructed in each
zone-subgrade combination. This unbalanced design limits the “power” of the
experiments. In the SPS-2 experiment, 7 sites are located in the WP zone
compared to 2, 3 and 2 in WNF, DF, and DNF zones, respectively. Furthermore,
in some of the sites not all sections were constructed on the same subgrade soil
type [for example, NV (32)].

0 The SPS-2 experiment has 12 designs constructed per site. Hence, the
experiments do not have any “true” (statistical) replication of designs. In other
words, though two sites (with 12 designs at each site) are located in a climate-
subgrade soil combination, the traffic, age, and material-related properties vary
between the sites.

0 The variation in age of the sites is considerably high for the experiments. If the
sites were reasonably similar in age, the findings would be more reliable. It
should be noted that age of the test sections was included as a covariate in all

statistical analyses to address the above issue to some extent.

142

0 In the SPS-2 experiment, in-pavement drainage was provided only for sections
built with PATB. Moreover, all sections with PATB were provided with drainage.
As a result of this, the effect of PATB and the effect of drainage are inseparable
(confounded).

0 The sections with 203 mm (8-inch) thick PCC slabs have dowels of 32 mm (1.25-
inch) diameter and sections with 279 mm (1 l-inch) thick PCC slabs have dowels
of 38 mm (1.5-inch) diameter. The effect of PCC slab thickness, especially on
faulting, is thus not “pure”. E

0 Only the lower limit for traffic volume was specified for the SPS-2 experiment. L
This resulted in considerable variability in traffic across sites.

9.3.2 DATA-RELATED ISSUES

0 Reasonably accurate monitored traffic data is not available for all the sites in the
experiments. This has further complicated the issue of controlling for traffic.

0 Large measurement variability was observed, over time, for some of the distresses
(faulting in SPS-2). This variability has made the time-series trends unclear for
some of the performance measures.

0 Though thorough construction guidelines were developed by the LTPP to
minimize construction variability across sites, some deviations occurred. These
deviations along with the material variability have added to the variability in
performance across sites. If material-related information were available for all the
sections, the issues caused by performance variability could be better addressed.

0 Backcalculated layer moduli are unavailable for most of the sections. Some of the

material-related issues could be dealt with if the data were present.

143

The data regarding the coefficient of thermal expansion (CTE) of concrete is not
available in the DataPave IMS database (Release 17.0). Therefore, CTE of

concrete could not be considered in the analyses.

9.4 RECOMMENDATIONS FOR FUTURE DATA COLLECTION

recommendations for future data collection and research are given below.

1)

2)

3)

4)

5)

AND RESEARCH

Based on the above issues and the experience of the authors with the LTPP data,

Reasonably accurate monitored traffic data should be made available for all the

II.
‘0‘

sites to allow for better adjustment of traffic loading variation across sites.

All the core sections of the experiment should be closely monitored until failure
or to a stage when the long-term performance (at least 15-20 years) has been
captured.

The core sections should be strictly supervised to prevent any maintenance
activity, as per the experiment designs. This will ensure that the actual long-term
performance of the pavements is observed.

Most of the test sections will soon enter a critical stage in their service life; in
light of this, to reap maximum benefits from the experiments, the sections should
be monitored at regular intervals and with greater accuracy.

Hall and Correa (Hall and Correa 2002) have identified issues regarding in-
pavement drainage for. The findings from this study should be considered for
inclusion in the DataPave IMS database. This may help study the effect of in-

pavement drainage more accurately.

144

 

6) Some of the sections have shown premature “failure”. These sections should be
considered for exclusion from DataPave, as they do not contribute to the study of
long—term pavement performance.

7) Some of the sites in the SPS-2 experiment are very close to the thresholds
(regarding average annual precipitation and freeze index) defined for delineation
of climatic zones. The definitions of the climatic zones may need reconsideration
in light of this.

8) The definition of pumping, in the case of the SPS-2 experiment, should be
revisited to allow for its inclusion in future studies.

9) Accurate material data should be made available for all the sections of the
experiments to allow for addressing the variability in material quality across sites.
Also, backcalculated layer moduli data should be made available for all sections
in DataPave to help perform mechanistic analyses.

10) Most of the construction-related issues are available only from construction
reports. These issues and/or deviations should be better highlighted well within
the DataPave database.

11) The spatial location of some distresses (such as cracking) is sometimes important
for research. It is practically cumbersome for the users to obtain distress maps and
interpret the spatial location of the distresses. It is therefore recommended that
each section be “discretized” (segmented) for data collection and the related data
be made available in DataPave. This would greatly decrease the level of

subjectivity in the data.

145

12) In general, the extent of distresses on the SPS-8 test sections is “low”, at this point
in time. The performance data should thus be collected for sufficiently long time
(15 to 20 years for all the sections in the experiment) to capture the effects of
environment. A meaningful statistical analysis may then be performed to study

the effects of environment.

146

APPENDIX

147

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nd w; Nd m; dd d.N Nd d.N dd nd m; .m ml?» mm m:
Nd w._ w; Nd dd d.N dd d.N Nd d.N dd 0 ml? m Mddfi
d; d._ d._ d._ d4 d4 d._ d._ d._ d._ d._ 0 .55 mm nmu
dd 2 3 dd Nd w; md 04 #4 d4 vd h .35 dd wde
d4 dd Nd mg md N: do d.N dd dd d.N an E5 mm de
Nd w; n; md dd d4 dd fiN dd d.N dd ...— E N mdmfi
md m; md m4 dd d.N dd dd d.m d.N dd .m E5 dN hmh
d; do do d4 dd d.N dd d.m dd d4 d. H ..— E dm Rh
dd d.N dd d.N dd d.N dd d.m dd d. d d; 0 E5 3 dwm
dd : dd 3 Nd m4 dd mg mg d.N dd U mzn e deN
dd 3 : dd Nd M: dd d.m dd d._ d4 0 .mzn v Nddm
m4 md wd N._ dd m; dd d.m dd d; d; m mm mm New
d._ d._ dd : wd N; Nd o; N; 54 md U+m m: Nm ddw
dd d.N d.N dd dd d.N md mN dd dd d.N n+0 he w 89
o:

3 N— ddd dmm 3 w 9:...— moa— m<GG z > ovauwaam quake—MU 33m 958,—.

. 3a.. in .Auwfibm 5 685—35 2.5 3am own—:95 mad—220 «E96539

52.: 2:5 .95on on: aim Dom

 

 

 

 

 

 

 

mac—080 8835b .8 £335 <-_o>o.._ d 2de

148

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nd . mg Nd w; dd .. d.N Nd . m.N md NA wd b m2?» hm m:
v4 dd dd v; dd d.N dd w; NA d.N dd 0 56$ m Mddu
d4 d4 d._ d4 d4 d4 d4 d._ d._ d4 d4 0 .55 mm EN
d.N dd d.N dd dd d.N dd d.m dd d._ d; m .55 dd add
u; md ed 0; Nd w; md 5N dd Nd w; ...— m>> an de
Nd w; w; Nd dd d.N dd mN nd d.N dd ..— .ES N 33
vd d; md m4 . dd d4 dd dd _.N d.N dd m .55 dN rm»
.2 5d m4 md wd N._ d.N dd dd md m4 r.— .ES d— Fm
M: Nd dd d.N mg Nd ~d d.N dd dd d.N U ES dm dam
dd d.N d.N dd d.N dd dd d.N dd d4 d; U .mzn e deN
d." vd vd 04 dd d.N md m4 dd m4 md U .mza v Nddm
d.N dd d.N dd dd d.N dd d.N dd d. _ d. m m an mm Ndv
N." dd : dd md m4 dd 04 N; w; Nd U+h mn— Nm ddw
d. _ dd yd d4 dd d.N Nd 5N dd dd v4 h+0 n:— w ddv
o:
3 Na ddd dmm 3 w 9::— mUA £499 Z > odauuasm gem—U 85m 953,—.
«08 man £59.25 5 £85.35
.523 23‘— ?uES—h 00m an? on: odb 8am 0933.5 wan—920 ESE—398‘—

 

 

 

 

 

 

 

wcfiofio .mcfiszwc2 so max—«5w <Ao>3 N 2de

149

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

d4 d4 d4 d4 d._ d._ wd N._ d._ 2 dd r.— mZB hm m:
3 dd 2 dd d. _ d. _ wd d. fl NA NA wd U “—23 m mddfi
d; d._ 3 dd d4 d4 2 dd 2 d._ d4 0 p.55 mm Em
d. d d. d d. ~ d. d dd 2 dd 2 d._ d. _ d. _ ..— m>> dm wdd
d._ d4 d; d; : dd d4 d; d._ d._ d._ h .55 mm de
dd 2 : dd dd 2 Ed 3 N._ m; nd a m3 dN 3m—
dd 3 d. _ d. m : dd dd 3 dd d; d. d ..— E$ dN hmh
dd 2 d. _ d. _ d. _ d. _ : dd d. d dd 2 ..— m>> du 2h
wd N. H 2 dd : dd dd 2 d4 d4 d._ U .55 dm ddm
d. d d. _ d. d d. _ : dd d; d. _ 2 2 dd U mzn d mva
d. d d._ dd 3 d4 d. d dd dd N. a 2 dd 0 mzn v Ndd~
d. d d. d d; d. d d. _ d. d dd 3 d. d d4 d._ ..— .mn mm New
dd 2 3 dd d. _ d. ~ wd d. d N4 N4 wd Of.— mn Nm ddw
d._ d._ d4 d._ d4 d._ 3 : md dd : n+0 “a w ddv
o:
3 N a ddd dmm : a 955 m0.— m<wn Z > ocean—aw era—“MU 38m 953,—.
good Ed .5325 E 68535
.52: 2.3 .333...— oom ea 8.. 25 83 $8.29 55 aaéuacm

 

 

a co mowing <-_o>oq m 2an

150

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 3 od 3 3 3 3 O.” o.o 3 3 3 ..— R
od 2 od ca od 3 o._ 3 3 3 3 S o m ...zB
3 3 3 3 3 3 S 3 3 3 3 3 0 mm
Qo 3 9o 3 o.o 2 mo 2 Qo 9m 2 no ..— an
3 2 ca o.o 3 3 3 3 od o.o 3 3 m an
3 3 ed on od 3 no 3 3 3 3 3 m em .55
3 S 3 3 od 3 od 3 3 3 S 3 ..— 2
od 3 o.o 3.. S o._ S 3 3 3 3 S ..— a
3 3 od 3 S 3 S 3 3 3 3 3 o 3
3 M: No 3 no 5 o.o 3 3 3 3 3 o e
No 3 do 3 3 S 3 3 3 3 3 3 u .. ...zn
3 3 od 3 A: 3 3 3 3 3 3 S ..— mm
No 3 x 2 3 x No 3 x ..—
x x o a... ..E

3 3 x x x x x .—

x x x x x u a
: w 2 w = a = a z a 2 a

e_ u_ e_ u_ v_ N“
:3 3G: 9:3 muflwwam aam awhwu
:2
@535 8.— ? 8am

 

 

 

 

 

$2225. 82m 00m do Subm— -wcflo€o 8855.: :0 £923.» m-_o>ox_ v 2an

151

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 3 3 3 o.o 3 3 ed a R
S 3 3 3 3 od 3 od 0 m .565
3 3 3 3 3 A: o.._ 3 0 mm
3 3 3 9o 2 no 3 3 m an
S 3 3 od 3 od 3 3 .— an
S 3 od od 3 ed od o.o m 3 ES
3 3 3 3 3 od 3 o.o m 8
3 3 3 3 3 od o.~ ed a 2
3 3 3 3 3 3 ca od 0 2
3 3 ca o.o oN o.o 3 9o 0 o
3 ed oN ed oN od oN o.o u e .mzn
ca o.o 3 3 3. od ca od ..— mm
2 3 x 2 m o x m
x o a ..E
x x 3 3 S no .._
3 3 3 o _ x u a
:z a 92 a :2 a dz a
2 2 3 2 2: o:
S ocaauflam 83m use—“W5
own-:95 he «comm

 

 

owmcwfim do “ovum -mcflofio 0963.85 :0 mead—mam m-_o>3 m 2an

152

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 3 3 3 3 3 od o.o 3 3 3 3 r.— R
3 3 3 3 3 3 3 3 od 3 3 3 o m 33
3 3 3 3 3 3 3 3 3 3 3 3 0 mm
3 3 3 3 3 no 3 3 to 3 no 3 m an
3 3 3 3 3 od 3 ed od 3 3 3 r.— an
3 3 3 3 3 3 3 3 od 3 no 3 r.— 3 E5
3 3 3 3 3 3 od 3 3 o.o 3 o.o m 8
3 3 3 3 3 3 3 o.o o.o 3 3. 3 m 3
3 3 3 3 3 3 3 3 3 3 3 3 u S
3 3 3 3 od 3 3 3 0.0 3 2 3 u e
3 3 3 3 3 3 3 3 3 3 3 3 o 9 m2:
3 o.o o.o 3 3 3 3 3 3. 3 3 3 m mm
No 3 S 3 we 3 we to x m
54 md 0 NM ha—

x x 3 3 3 3 no u

3 3 3 3 x o a
m3 369 93.. m3 36a FEE m3 36: m3.— mua 36: m3.”

:2 n 92 a :z a nz :

3 2 3 2 mew—mag 35m 23 .85

2
09:. 03¢ .3 «Sam

 

 

 

 

 

09¢. 3mm mo 80th -wcflofiu 083385 so mama—«cm 9354 o 2an

153

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 mo 3 3 mo 3 3 ad ad 3 3 3 m R
3 mo 3 3 mo 2 3 3 3 no 3 3 o m m2?»
3 3 3 mo 3 3 3 3 we 3 3 3 0 mm
3 3 3 3 we .3 mo 3 3 3 3 mo m an
3 3 3 3 3 no 3 we 3 we 3 3 m an
we 3 mo 2 no , 3 2 mo 3 3 3 3 m 3 E5
2 3 3 mo 3 we no 3 3 mo 2 .3 m 3
mo 3 mo 2 3 3 3 we we 3 3 3 m 3
3 9o 3 3 3 3 mo 3 3 mo 3 3 u 3
3 mo 3 3. 2 mo 3 3. 3 ad 3 we 0 e
3 3 3 ad 3 mo 3 3 3 3 3 3 o 9 m2:
mo 3 3 3 3 ad 3 3 3 ad ad 3 m mm
3 3 x 3 3 x x 3 3 ..—
x x O Nm ha—
3 3 x 3 3 x x a
x 3 3 x x u w
2 w = a 2 a z a = a S a
8a
m3 559 :55 amp 83m BEN
92 2:: saw 835:0
$05.35 8m S .83

 

 

 

 

 

3.2205 23 o3 3 “Sam -3 8 2353 323 5 29¢

154

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 M3 3 mo 3 ad 3 mo m R
3 mo 3 3 .3 no 3 S U m E
3 no 3 3 3 3 3 3 0 mm
mo 3 3 3 3 mo 3 mo m an
mo 3 3 mo 3 3 ad 3 m mm
3 we 3 B 3 1 mo 3 9o m 3 E5
3 mo Qo 3 mo 3 3 3 r,— 3
3 mo 3 3 mo 3 mo 3 m S
3 mo 3 3 mo 3 3 3 U 3
3 3 3 ad 3 mo 3. 3 o e
3 mo 3 mo 3 we 3 mo 0 v mzn
3 mo 3 3 3 3 3 3 m mm
3 3 x 3 3 x m
L x u «m ,3
x x we 3 x m
we 3 x x U m
CZ n— DZ 9 DZ 9 :2 D
3 N“ 3 NH 2: o:
: «Baham 33m gamma
augmen— uo Scam

 

 

0333.5 mo Subm— AMH co max—«3 m-_u>3 w 3an

155

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 : wd m4 ad wd d._ 3 ad N4 ad wd h an
d. 3 d._ ad 2 3 wd d3 2 wd ad v.3 ad 0 m “~23
d3 m; wd 03 d3 d3 ad 2 : wd d._ «A 0 mm
2 ad 3 NA ad ad 2 3 wd : : wd «— an
ad d3 2 ad NA ad 2 ad 03 2 ad d3 «— mm
3 2 wd ad m3 wd 3 2 md 2 m3 dd ...— wN r55
2 d3 ad N; wd d3 «4 ad 3 d. _ m3 wd b ca
ad 3 d3 wd d3 NA NA , wd d4 ad ad. m; r..— aw
m3 d3 md : ad d3 ad _ d3 3 2 3 ad U A:
0.3 d3 d3 3 3 ad 3 d3 ad ad d3 d3 0 w
ad 2 ad ad m4 ad ad m4 ad d3 2 d. 3 U v RZQ
2 2 ad 3 d3 ad m4 wd ad ad d._ d._ h mm
2 ad d._ ad 2 ad «4 ad X h
a , wd 3 0 mm m:

x x 3 M3 3 3 3 m

wd N; 0.3 d._ x 0 w

mm: m<wn 9:5 GOA mafia 9::— mon— m<wn 9::— mod m<wn mks—
DZ Q QZ G :2 9 DZ Q 2: .28
2 2 2 S «afiwmsm 83 EM
3 w
09$ 35: ac Beaum—

 

 

 

 

 

oaah 8mm do 809cm -5: so Ema—«3 9354 a 23%

156

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O4 O4 0.0 ON 0.0 ON 0.0 ON 0.0 ON O4 O4 r..— FM
0.0 ON 0.0 ON O4 O4 0.0 ON O4 O4 O4 O4 U m E
O4 O4 O4 O4 O4 O4 O4 O4 O4 O4 O4 O4 U mm
0.0 ON O4 O4 O4 O4 O4 O4 O4 O4 O4 O4 r..— QM
0.0 ON 0.0 ON 9O #4 ON 0.0 ON 0.0 O4 O4 rm an
0.0 ON 0.0 ON 0.0 ON O4 O4 O4 O4 O4 O4 an @N KB
MO 54 0.0 ON O4 O4 0.0 ON O4 O4 O4 O4 rfl 6N
O4 O4 O4 O4 O4 O4 0.0 ON 44 0.0 O4 O4 rm aw
O4 O4 a4 4.0 O4 O4 O4 O4 O4 O4 0.0 ON U Ow
ON 0.0 0.0 ON O4 O4 O4 O4 O4 O4 O4 O4 0 O
0.0 ON 0.0 ON O4 O4 0.0 ON N.O w4 0.0 ON U V rmzn
0.0 ON O4 O4 O4 O4 O4 O4 O4 O4 O4 O4 ry— mm
N.O O4 X N.O w4 X X a..O m4 r..—
x X D NM a:—
O4 O4 N O N X X an
X 4 4 X X U w
: a = w = w = m S a : w
cmm a: 8m .3
m3 :5: m3.— oumumwam 23m 3th6
92 a
@8535 03 a. .83

 

 

 

 

 

305425 37, 00m do “outm— -w=U_ofiU 72:45:33 :0 £333 @354 O4 23%

157

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 3 3 3 3 3 3 3 m an
3 3 3 3 3 3 3 3 U m m2?
3 3 3 3 3 3 3 3 0 mm
3 3 3 3 3 3 3 3 m an
3 3 3 3 3 3 3 3 m an
3 3 3 3 3 3 3 3 m 3 ES
3 3 3 3 3 3 3 3 r.— 3
3 3 3 3 3 3 3 3 r,— S
3 3 3 3 3 3 3 3 U S
3 3 3 3 3 3 3 3 O c
3 3 3 3 3 3 3 3 0 v mzn
3 3 3 3 3 3 3 3 m mm
3 3 x 3 3 x m
x D an hm—
x x 3 3 x m
3 3 x x U m
DZ G DZ 9 OZ 9 OZ Q
3 N4 3 N4 2: a:
3 32333 33 3.2wa
3:55.... .4: Beam

 

 

 

 

 

owmcufin do 80km -mFEofiU 3:65:33 no 39333 m-_o>3 3 23%

158

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 3 3 3 3 3 3 3 3 3 3 3 m an
3 3 3 3 3 3 3 3 3 3 3 3 U m m2?
3 3 3 3 3 3 3 3 3 3 3 3 0 mm
3 3 3 3 3 3 3 3 3 3 3 3 r.— am
3 3 3 3 3 3 3 3 3 3 3 3 m an
3 3 3 3 3 3 3 3 3 3 3 3 m 3 E5
3 3 3 3 3 3 3 3 3 3 3 3 m 3
3 3 3 3 3 3 3 3 3 3 3 3 m S
3 3 3 3 3 3 3 3 3 3 3 3 U :4
3 3 3 3 3 3 3 3 3 3 3 3 U o
3 3 3 3 3 3 3 3 3 3 3 3 U c mzn
3 3 3 3 3 3 3 3 3 , 3 3 3 m mm
3 3 3 3 3 3 3 3 x ,..4
I 3 u «m 3

x x 3 3 3 3 3 m

3 3 3 3 x O a

M404 43:00 44.4.34 M404 m4<mvn4 mh<a4 4404 44<UA4 44,4.<A4 4404 443044 44,4334
DZ n4 DZ A4 OZ A4 A42 A4 24% and
2 2 E S coauham 33m assume
44 w
245 334 .4: “8.4.444

 

 

 

 

 

09C. 03m 40 Hootm -chQEU 3:63.323 no 33433 m-_o>o‘4 N4 2341

159

SITE WISE SUMMARIES
Arizona (4)

Site Description
The site in Arizona is located in the eastbound direction of I-10, approximately 35 miles
west of Phoenix between Tonopah and State Spur on route 85, in Maricopa County. The

other inventory data for the sections are summarized in Table 1.

Table 1 Inventory data for AZ (4)

 

 

 

 

 

 

 

 

 

Site code 4

Climatic zone Dry No Freeze
Average annual precipitation 232 mm
Average annual freezing index 0.0 °C-days
Traffic open date 1St of October 1993
‘Proposed’ traffic 1092 KESALs/year
Subgrade soil type Coarse grained
Inside shoulder type PCC
Outside shoulder type PCC

 

 

 

 

Construction Issues
According to the construction report, no major problems were encountered during
construction. The following are the main construction issues encountered at the site:
0 In the construction of DGAB, AASHTO No. 57 coarse aggregate was utilized in
as the backfill material in the pavement base drain.
0 The geotextile wrapped around the PATB edge was short and could have caused
intrusion of soil from adjacent DGAB.
0 Transverse drains were installed perpendicular rather than in a herringbone

fashion.

160

0 Transverse cracking in the LCB layer was observed in sections 4-0217, 4-0218, 4-
0219 and 4-0220 prior to the placement of the slab.
0 Longitudinal tie bars were placed uncoated and also the bars were smaller (20”
long) than specified (30”).
o PCC slump variations and segregation occurred in several test sections.
Site Status
The deviations in the design and/or site features at this site are explained here. No
deviations in the site factors have been observed at the site. As planned, the site has
coarse-grained roadbed soil and it is situated in the Dry No Freeze climatic zone. A
summary of the status of the design features of the site is Table 2. Only one section, 4-
0222, has been constructed thicker/thinner by more than 0.5”. All the other design

features have been constructed as intended in the SPS-2 design.

Table 2 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP 11) 5'?" Th'd‘ms’ "" Base Type Drainage ”me man" As designed?
Desrgn Actual ft
0213 8 7.9 DGAB No 14 Yes
0214 8 8.3 DGAB No 12 Yes
0215 11 11.3 DGAB No 12 Yes
0216 11 11.2 DGAB No 14 Yes
0217 8 8.1 LCB No 14 Yes
0218 8 8.3 LCB No 12 Yes
0219 11 10.8 LCB No 12 Yes
0220 1 l l 1.3 LCB No 14 Yes
0221 8 8 .2 PATB Yes 14 Yes
0222 8 8.6 PATB Yes 12 No
0223 ll 1 1.1 PATB Yes 12 Yes
0224 1 1 10.7 PATB Yes 14 Yes
Target Std. Dev. Actual (average) Average Within 10%?
Flexural 3‘ ' —
. 550 10 572 Yes
Strength, psr
900 58 837 Yes

 

 

 

 

 

 

161

Data Availability
A summary of the data available for the sections in this site is Table 3. The section 4-
0213 is a seasonal monitoring section and that could be the reason for the higher amount

of data available for that section.

Table 3 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultig
0213 6 6 8 5
0214 6 6 8 4
0215 18 18 10 40
0216 6 6 8 5
0217 6 6 8 5
0218 6 6 8 5
0219 6 6 8 5
0220 6 6 8 5
0221 6 6 8 5
0222 6 6 8 5
0223 6 6 8 5
0224 6 6 8 4

 

 

 

 

 

 

162

Arkansas (5)

Site Description

The site, a four lane divided highway, is located on the westbound lanes of I-30 in Hot
Spring County, Arkansas. The other inventory data of the site in Arkansas have been
summarized in Table 4. It is to be noted that there is a discrepancy in the notation of the
sections between DATAPAVE and the construction reports. The sections were numbered
from 5-0213 to 5-0224 in DATAPAVE while the sections were numbered from 5-0201 to
5-0212 in the construction reports. However the notation in DATAPAVE was used

during the analysis.

Table 4 Inventory data for AR (5)

 

 

 

 

 

 

 

 

 

 

 

Site code 5

Climatic zone Wet No Freeze
Average annual precipitation 1381 mm
Average annual freezing index 38 oC-days
Traffic open date lSt of November 1995
‘Proposed’ traffic 1903 KESALs/year
Subgrade soil type Coarse grained
Inside shoulder type AC

Outside shoulder type AC

 

 

Construction Issues
Only a few construction issues occurred during the construction of the section at this site.
They are as follows:
0 The dowel basket assembly that got “entangled” with the paver had to be removed
while the paving operations were halted.
0 The longitudinal joints at the site were not sealed until early 1997 and pumping

was evident at the joints by that time.

163

Site Status
Though as per the design the sections at this site are to be founded on fine—grained soils,

most of the sections (i.e. except 0222 and 0223) have been built on coarse-grained
subgrade soils. The subgrade soil type for each test section has been decided based on
four different sources of information in the LTPP database- TST_L05B, SP82_LAYER,
TST_SS04_UG08, and TST_SSOl_UGOl_UG02. The data from SPS2_LAYER has been
found to be contradictory with the data from other sources for most of the sections at this
site. The construction report has indicated that the site has all the sections on fine-grained
subgrade soil. A decision has been made considering all the sources. But, as planned, the
site has been constructed in 3 Wet No Freeze climatic zone. Table 5 is a summary of the
status of all the design factors. Sections 0213 and 0215 have been constructed with a PCC
thickness deviation of at least 0.5 inch. Also the average l4-day flexural strength of the
PCC at the sections with target strength of 900-psi has been found to be lesser by more

than 10% the target.

Table 5 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP ID 8"" Thic'mm’ “" Base Type Drainage Lane Wid‘h’ As designed?
Design Actual ft
02 13 8 7.4 DGAB No 14 No
0214 8 8.4 DGAB No 12 Yes
0215 11 11.5 DGAB No 12 No
0216 l l 1 l DGAB No 14 Yes
0217 8 8.3 LCB No 14 Yes
0218 8 8.2 LCB N0 12 Yes
0219 11 11.1 LCB No 12 Yes
0220 l 1 10.7 LCB No 14 Yes
022 l 8 8.3 PATB Yes 14 Yes
0222 8 8.3 PATB Yes 12 Yes
0223 l 1 10.9 PATB Yes 12 Yes
0224 l 1 10.9 PATB Yes 14 Yes
Target Std. Dev. Actual (average) Average Within 10%?
Flexural 550 30 545 Yes
Strength, psi 900 226 666 N0

 

 

 

 

 

 

 

164

Maintenance construction was done on all the sections at this site. Longitudinal lane—
shoulder joints, transverse joints, and cracks were sealed as maintenance. The
maintenance work was done in 1997 and 2002, and resulted in the change of
‘construction number’ in the database since those years.

Data Availability
Table 6 is the summary of the monitoring data available from the LTPP database. Though

the site is more than 8 years old, the data available is only for four or five tests.

Table 6 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faulting
0213 5 5 4 4
0214 5 5 4 4
0215 5 5 4 4
0216 5 5 4 4
0217 5 5 4 4
0218 5 5 4 4
0219 5 5 4 4
0220 5 5 4 4
0221 5 5 4 4
0222 5 5 4 4
0223 5 5 4 4
0224 5 5 4 4

 

 

 

 

 

 

165

California (6)
Site Description
The project at this site is the youngest of all the sites in the SPS—2 experiment. It is
located on the northbound truck lane of SR 99, Delhi (Merced County), California. The
test sections were built as part of a realignment of SR 99 and a conversion to a four-lane

freeway. The other inventory data are summarized in Table 7.

Table 7 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 6

Climatic zone Dry N 0 Freeze
Average annual precipitation 299 mm
Average annual freezing index 0.2 °C-days
Traffic open date 1St October 2000
‘Proposed’ traffic 2405 KESALs/ year
Subgrade soil type Coarse-grained
Inside shoulder type AC
Outside shoulder type PCC

 

 

Construction Issues
No major problems were encountered during construction of the pavement sections at

the site. The main construction issues are as follows:
0 Cracks developed at several places in the LCB layer right after placement
because the curing compound was not placed properly.
0 Considerable segregation occurred in the LCB layer, due to large aggregate used
in mix.
0 The PATB layer was bladed, following an inspection, to make the surface

uniform.

166

0 The sides of the PATB material were completely covered by the overlaying PCC
material and the cement paste rendered the PATB almost ineffective. Later it was
cleaned up ‘sufficiently’.

0 Unlike in other sites of the SPS-2 experiment, two levels in dowel diameter, 32
mm and 38 mm, exist in sections with target PCC slab thickness of 203 mm and
in sections with target PCC slab thickness of 279 mm.

Site Status

No deviations in the site factors have been observed at the site. As planned, the site has
coarse-grained roadbed soil and it is situated in the Dry No Freeze climatic zone. A
summary of the status of the design features of the site is Table 8. Only one section, 4-
0211, has been constructed thicker by at least 0.5”. The lane width of test section with
target of 14 ft was found to be 13 ft (Table: SPS_GENERAL). The data of testing on the
PCC of the slab are not available from the database. All the other design features have

been constructed as intended in the SPS-2 design.

Table 8 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP ID Slab Thickness, in. Base Type Drainage Lane Width, As designed?
Design Actual ft
0201 8 8.2 DGAB No 12 Yes
0202 8 7.7 DGAB No 13 N0
0203 1 l l 1.4 DGAB No 13 N0
0204 1 1 1 1.4 DGAB No 12 Yes
0205 8 8.3 LCB No 12 Yes
0206 8 8.1 LCB No 13 N0
0207 l l 1 1.4 LCB No 13 N0
0208 1 1 10.8 LCB No 12 Yes
0209 8 8.3 PATB Yes 12 Yes
0210 8 8.2 PATB Yes 13 No
021 1 1 1 11.5 PATB Yes 13 No
0212 1 l l 1.2 PATB Yes 12 Yes
TargL Std. Dev. Actual (average) Average Within 10%?
Flexural 550 N.A. N.A. -
Strength, psi 900 NA. NA. -

 

167

 

Data Availability

A summary of the data available for the sections in this site is Table 9.

Table 9 Summary of status of available data

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultig
0201 3 3 3 2
0202 3 3 3 2
0203 3 3 3 2
0204 2 2 3 l
0205 3 3 3 2
0206 3 3 3 2
0207 3 3 3 2
0208 3 3 3 2
0209 3 3 3 2
0210 3 3 3 2
0211 3 3 3 2
0212 3 3 3 2

 

 

 

 

 

 

 

168

Colorado (8)

Site Description

The site in Colorado was constructed on I-76 (east bound) near Adams County in Denver,
Colorado. Six sections (0213 through 0216, 0218 and 0219) were a part of reconstruction
project and the other six sections were a part of new alignment project. The other

inventory data of the site have been summarized in Table 10.

Table 10 Summary of inventory data

 

 

 

 

 

 

 

 

 

Site code 8

Climatic zone Dry Freeze
Average annual precipitation 370 mm
Average annual freezing index 327 °C-days
Traffic open date 1" of November 1993
‘Proposed’ traffic 400 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type PCC

Outside shoulder type PCC

 

 

 

 

Construction Issues

According to the construction report of the site, six sections each were constructed in
‘cut’ and on ‘fill’. Sections 0213, 0214, 0215, 0216 and 0221 were constructed on fills.
The Phase 1 of construction included the construction of the new alignment and the Phase
2 of the construction was reconstruction of I-76. The major construction issues at the site
are as follows:

0 Subgrade pumping occurred on several sections during Phase 1 construction due

to wet weather and high water table at some locations.
0 Many PATB sections had too many fines in the mix. The mat in section 0221 was

replaced due to this problem.

169

o In section 0218, construction was stopped sometimes due to delay in delivery of
material and equipment. The dowel basket assembly was torn up at station

141+50 but not replaced in this section.

Site Status

The deviations in the design and/or site features at this site are explained here. The site is
one of the three sites in the SPS-2 experiment that have sections on both fine-grained and
coarse-grained subgrade soils. The site has five test sections (0214, 0216, 0219, 0223,
and 0224) on coarse-grained soils and the other seven sections on fine-grained soils. As a
majority of the sections have fine-grained soils, the site has been categorized under fine-
grained subgrade soil type (see Table 10). But according to the SPS-2 experiment design,
the site was supposed to be having coarse-grained subgrade soils only. This is the major
deviation from design at this site. A summary of the status of the design features of the

site is Table 10.

Table 11 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP ID Slab Thickness, m' Base Type Drainage Lane Width’ As designed?
Desrgn Actual ft
0213 8 8.7 DGAB No 14 No
0214 8 8.4 DGAB No 12 Yes
0215 1 1 1 1.4 DGAB No 12 Yes
0216 11 11.8 DGAB No 14 No
0217 8 8.6 LCB No 14 No
0218 8 7.7 LCB No 12 Yes
0219 11 11.1 LCB No 12 Yes
0220 1 1 1 1.1 LCB No 14 Yes
0221 8 8 . 3 PATB Yes 14 Yes
0222 8 8.7 PATB Yes 12 No
0223 11 11.8 PATB Yes 12 No
0224 11 11.7 PATB Yes 14 No
Target Std. Dev. Actual (average) Average Within 10%?
Flexural 1— ‘ '
Strength, psi 550 45 526 Yes
900 58 906 Yes

 

 

 

 

 

 

170

Six of the twelve sections have PCC thickness exceeding the respective target PCC

thickness by more than 0.5 inch. The target l4-day flexural strength has been met, based

on average 14-day flexural strength values.

Data Availability

A summary of the data available for the sections in this site is Table 12. The initial

faulting and distress survey was conducted only in 1996 though the section was opened to

traffic in 1993.

Table 12 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faulting
0213 5 5 7 5
0214 4 4 7 5
0215 5 5 7 5
0216 4 4 7 5
0217 6 6 8 7
0218 7 7 8 7
0219 7 7 8 7
0220 7 7 8 7
0221 7 7 8 7
0222 7 7 8 7
0223 7 7 8 7
0224 7 7 8 7

 

 

 

 

 

 

171

Delaware (10)

Site Description

The site is located on US 113 between Milford and Georgetown, Delaware. The site was
included in the additional two southbound lanes to an initial to-lane roadway. Within the
area of the site, there is an intersecting highway that is assumed to cause only
insignificant impact on truck traffic through the test sections. Other inventory data has

been summarized in Table 13.

Table 13 Summary of inventory data

 

 

 

 

 

 

 

 

 

Site code 10
Climatic zone Wet Freeze
Average annual precipitation 1144 mm
Average annual freezing index 103 °C-days
Traffic open date 1St of May 1996
‘Proposed’ traffic 430 KESALs/year
Subgrade soil type Coarse grained
Inside shoulder type AC
Outside shoulder type AC

 

 

 

 

Construction Issues

A variety of construction issues have been reported in the construction report. Problems
were encountered with weather and poor performance of concrete in some test sections.
The main issues have been summarized below:

0 All the SSO-psi PCC has been replaced with the Delaware DOT Type ‘B’ mix that
gives a flexural strength of approximately 650 psi. Extensive cracking (poor
performance) prompted the Delaware DOT officials to take this decision.

0 Some 900-psi concrete was also replaced (sections 0202 and 0206), with 900-psi

mix with 7.5-bag mix, after breaking and removing cracked PCC.

172

0 Also 900-psi mix was found hard finish during paving operations.
0 Concrete patching was done at many locations where cracks appeared on the
PCC.
0 Transverse shrinkage cracks appeared in LCB before PCC was laid.
0 During construction of the LCB, depressions occurred due to stoppage of the
paver. Transverse cracks were observed at some of these depressions.
0 “High spots” were milled before paving was done for some of the sections.
0 Edge drains did not extend to the full length of the PATB in section 0211.
o A transverse construction joint was placed within section 0212.
o The road was opened to traffic before all the joints were sealed.
Site Status
As said above, the 550-psi SHRP mix has been replaced with Delaware DOT Type ‘B’
mix after the 550-psi SHRP mix was found to be performing poorly (shrinkage cracking).
This is the main deviation at this site. A summary of the status of the design factors at
this site is in Table 14. Seven of the twelve test sections have been built with PCC layers
thicker than their respective target thickness, by at least 0.5 in. Also, the flexural strength

of the PCC of the sections are different from the target strength by more than 10%.

173

Table 14 Summary of status of design factors

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP ID Slab Thickness, in. Base Type Drainage Lane Width, As designed?
DeSIgn Actual ft
0201 8 8.3 DGAB No 12 Yes
0202 8 8.8 DGAB No 14 No
0203 11 11.7 DGAB No 14 No
0204 11 ll DGAB No 12 Yes
0205 8 9.2 LCB No 12 No
0206 8 8.9 LCB No 14 No
0207 l 1 l 1.3 LCB No 14 Yes
0208 l 1 12.1 LCB No 12 No
0209 8 8.2 PATB Yes 12 Yes
0210 8 8.3 PATB Yes 14 Yes
0211 11 11.8 PATB Yes 14 No
0212 1 1 12.4 PATB Yes 12 No
Target Std. Dev. Actual (average) Average Within 10%?
Flexural - - '
. 550 101 657 No
Strength, psr
900 152 757 No
Data Availability

A summary of the data available for the sections in this site is Table 15.

Table 15 Summary of data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress IRI FWD
Manual Faultingfi
0201 7 7 7 5
0202 6 6 7 4
0203 6 6 7 4
0204 6 6 7 4
0205 5 5 7 4
0206 5 5 7 4
0207 6 6 7 4
0208 6 6 7 4
0209 7 7 7 5
0210 5 5 7 4
0211 5 S 7 4
0212 6 6 7 4

 

 

 

 

 

 

174

Iowa (19)

Site Description

The site is located in the northbound lanes of US. 65 in central Iowa, northeast of Des
Moines. The project was included in the relocation of the US. 65 in both the northbound

and southbound lanes. A summary of other inventory data is in Table 16.

Table 16 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 19

Climatic zone Wet Freeze
Average annual precipitation 901 mm
Average annual freezing index 580 °C-days
Traffic open date lSt of December 1994
‘Proposed’ traffic 377 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type AC

Outside shoulder type AC

 

 

Construction Issues
No major construction problems have occurred during construction of the report, as per
the construction report. Some of the main issues are as follows:
0 In six of the twelve sections at the site (0213, 0214, 0215, 0217, 0219, and 0221),
underground structures were located. The depth of the structures has a range of
2.4 m to 12.2 m, with reference to the profile grade.
0 At least 0.3 m of geotextile was removed from the longitudinal edge of the
sections because of the low permeability of the geotextile.
0 During placement of the PCC slab for the test section 0222 incorrect dowel

baskets were placed. The section was thus relocated to avoid this area.

175

Site Status

Five sections (0215, 0216, 0221, 0223, and 0224) have been built with PCC thickness

greater than corresponding target PCC thickness by a margin of 0.5 in. Also the average

14-day flexural strength of the sections is lesser than the corresponding target flexural

strength by a margin greater than 10% of the target flexural strength. Table 17 is a

summary of the status of the design features at the site.

Table 17 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP iD Slab Th'c'mess’ “" Base Type Drainage Lane Wid‘h’ As designed?
Design Actual ft
0213 8 8.5 DGAB No 14 Yes
02 14 8 8.4 DGAB No 12 Yes
0215 11 11.8 DGAB No 12 No
0216 l 1 11.6 DGAB No 14 No
0217 8 8.1 LCB No 14 Yes
0218 8 8.2 LCB No 12 Yes
0219 11 11.2 LCB No 12 Yes
0220 1 1 1 1.4 LCB No 14 Yes
022 1 8 9.4 PATB Yes 14 No
0222 8 8.3 PATB Yes 12 Yes
0223 1 1 11.7 PATB Yes 12 No
0224 1 1 11.6 PATB Yes 14 No
Target Std. Dev. Actual (average) Average Within 10%?
Flexural 550 31 467 N0
Strength, psi 900 47 753 No
Data Availability

A summary of the available monitoring data for the sections at this site is in Table 18.

Though the site is almost 10 years old, the deflection data is available only for about 4

16818, 011 an average.

176

Table 18 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faulting
0213 6 6 8 4
0214 5 5 8 3
0215 6 6 8 3
0216 6 6 8 3
0217 6 6 8 4
0218 6 6 8 4
0219 6 6 8 4
0220 6 6 8 4
0221 6 5 8 4
0222 5 5 7 3
0223 5 5 8 3
0224 5 5 8 3

 

 

 

 

 

177

Kansas (20)

Site Description
The site is located in the westbound driving lane of Interstate 70 near Abilene in
Dickinson County. The project was included in the reconstruction of I-70 and was built

on fill. The other inventory data are summarized in Table 19.

Table 19 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 20
Climatic zone Wet Freeze
Average annual precipitation 820 mm
Average annual freezing index 259 0C-days
Traffic open date lSt of August 1992
‘Proposed’ traffic 757 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type PCC
Outside shoulder type PCC

 

 

Construction Issues

The main construction issues encountered at this site are as follows:

0 Underground structures were present at the site in sections 0208 through 0212 and

0204. The drains were at least 1.5 m below the pavement surface.
0 Two partial-depth repairs were done to the test section 0204 in the year 1995.
0 Vertical curves exist within the limits of the site.
0 PATB was difficult to place. Excess PATB was removed with a trimmer.
0 Existing subbase and shoulder material was retained.

0 Subgrade was dried up prior to construction using Type ‘C’ Fly Ash.

178

Site Status

Table 20 is the summary of the status of design factors at the site. Two sections, 0202 and
0209, have at least 0.5 inches as deviation from target PCC slab thickness. Test section
0202 has been built 0.6” thinner and 0209 has been built 0.5” thicker than corresponding
target thicknesses. Also the average 14-day flexural strength of the sections with target
14-day modulus of rupture as 550-psi is more than 10% higher than the target (see Table
20).

Partial depth repairing was performed on section 0201 in the year 2995 resulting in a

 

change in ‘construction number’ for the section since that year.

Table 20 Summary of status of the design factors

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SHRP iD 8"” Th'c'mess’ “" Base Type Drainage Lane w'd‘b’ As designed?
Designi Actual ft
0201 8 7.7 DGAB No 12 Yes
0202 8 7.4 DGAB No 14 No
0203 l 1 l 1.1 DGAB No 14 Yes
0204 1 1 l 1.3 DGAB No 12 Yes
0205 8 7.8 LCB No 12 Yes
0206 8 7 .9 LCB No 14 Yes
0207 1 1 1 1.3 LCB No 14 Yes
0208 l 1 l 1 LCB No 12 Yes
0209 8 8.5 PATB Yes 12 No
02 10 8 8.3 PATB Yes 14 Yes
0211 11 11.1 PATB Yes 14 Yes
0212 1 1 10.9 PATB Yes 12 Yes
Flexural Target Std. Dev. Actual (average) Average Within 10%?
Strength, psi 550 47 613 No
900 50 843 Yes
Data Availability

Table 21 is the summary of the monitoring data availability. Though the site is about 12

years old, the monitoring data available is for less than or equal to six tests for distress

 

179

and deflection data. Also, unlike in the case most of the sections in the experiment, the

faulting data and distress data are not available to the same extent in this site.

Table 21 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultingfi
0201 6 4 11 4
0202 6 4 12 3
0203 6 4 l2 3
0204 6 4 12 4
0205 6 4 12 4
0206 6 4 12 4
0207 6 4 12 4
0208 6 4 12 4
0209 5 3 12 3
0210 6 4 12 4
0211 6 4 12 4
0212 6 4 12 4

 

 

 

 

 

 

180

Michigan (26)

Site Description

The Michigan SPS-2 site is located on the US 23 (Ottawa Lake, Monroe County), which
is a rural principal arterial. The project was included in the reconstruction of US 23.
Consear Road bisects the site. Most of the sections were constructed on fills. Some of the
sections (0218 and 0219) were constructed on a superelevation. The other inventory data

for the site has been summarized in Table 22.

Table 22 Summary of inventory data

 

 

 

 

 

 

 

 

 

Site code 26

Climatic zone Wet Freeze
Average annual precipitation 866 mm
Average annual freezing index 382 °C-days
Traffic open date 1St of November 1993
‘Proposed’ traffic 1505 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type AC

Outside shoulder type AC

 

 

 

 

Construction Issues
The major construction issues, from the construction report, at the Michigan site are as
follows:
0 The traffic flow over all of the test sections is not uniform as Consear road bisects
the site.
0 Moisture content of the compacted subgrade was not maintained in the range of
85 to 120% of the optimum moisture content on sections 0213 through 0220.

0 The DGAB for section 0221 segregated.

181

 

0 The Geotextile fabric did not extent to the stipulated minimum depth of 1’ under
the pavement.

0 Rutting (1/2” to 1-3/4”) occurred in PATB due to traffic that was allowed to pass
over the outside shoulder area of PATB during construction.

0 A transverse construction joint was located in the LCB of section 0218.

0 Longitudinal cracking of LCB was observed in 0217 and 0220, which could be
due to the paving machines that were allowed to operate from the outside
shoulder area.

0 LCB in sections 0218, 0219 and 0220 and PCC in sections 0215 and 0219 had
lesser than 1” of slump, which is the stipulated value.

0 Embankment clay dried out and desiccation cracks appeared

0 Ruttin g developed from 0-15 to 0+15 near the inner wheel path and 0-02 to 0+15

in the outer wheel path of 0221

Transverse shrinkage cracks appeared in LCB soon after construction

Extra amount of water entered the pavement structure since this section is located

on superelevation, which drains toward the outside shoulder.

PCC Concreting delayed by a month in 0216

Site Status

Table 23 is the summary for the status of the design features of the sections at this site.
Four of the twelve sections (0213, 0214, 0217, and 0218) were built with PCC thickness
deviation of at least 0.5 in. The average l4-day modulus of rupture of PCC of the sections
with target 14—day strength of 550-psi is greater than the target by a margin of 10% of the

target strength (see Table 23).

182

Table 23 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP ID Design Actual Base Type Drainage Width, ft designed?
0213 Si 8.6 DGAB No 14 No
0214 8 8.9 DGAB No 12 No
0215 11 11.2 DGAB No 12 Yes
0216 1 1 11.4 DGAB No 14 Yes
0217 8 8.5 LCB No 14 No
0218 8 7.1 LCB No 12 No
0219 1 1 10.9 LCB No 12 Yes
0220 1 l l 1.1 LCB No 14 Yes
0221 8 8 .2 PATB Yes 14 Yes
0222 8 8.4 PATB Yes 12 Yes
0223 l 1 1 l PATB Yes 12 Yes
0224 1 1 1 1.2 PATB Yes 14 Yes
Flexural Target Std. Dev. Actual (average) Average Within 10%?
Strength, 550 30 617 No
psi 900 4 973 Yes

 

 

 

 

 

 

 

Sections 0213, 0215, 0217 and 0218 were de-assigned from the experiment in the years

1999, 2000, 1999, and 1998.

Data Availability

A summary of the extent of monitoring data available in the LTPP database for the site is
Table 24. Though the site is about 11 years old, the deflection data is available for only 3
to 6 tests. Unlike in the case of other sites, this site has high variation in the extent of

available data among the test sections. The section 0218 has the least amount of data.

183

Table 24 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitorinfig data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultingfi
0213 5 4 6 5
0214 6 6 10 5
0215 5 5 7 4
0216 8 7 10 5
0217 4 3 5 4
0218 3 3 4 3
0219 7 7 10 5
0220 8 7 9 5
0221 9 8 9 6
0222 7 7 10 5
0223 8 8 10 5
0224 8 7 10 4

 

 

 

 

 

 

184

Nevada (32)

Site Description

The site is located in north central Nevada, in the outer eastbound lane of Interstate 80, in

Humboldt and Lander Counties. The other inventory data are summarized in Table 25.

Table 25 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 32

Climatic zone Dry Freeze
Average annual precipitation 222 mm
Average annual freezing index 276 °C-days
Traffic open date 1St of September 1995
‘Proposed’ traffic 800 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type PCC
Outside shoulder type PCC

 

 

Construction Issues
The Nevada site has many serious construction issues according to the construction
report. The PCC layer had a wide range of construction issues. It is highly unlikely,
according to the construction report, that any of the sections will last the intended life for
the experiment. A majority of the problems with the PCC paving came as a result of the
mixes being significantly different from those typically used. Major construction issues,
from the construction report, are as follows:
0 The site was constructed after removal of an existing AC pavement. After
removal of then existing AC layer, cement treated base, and DGAB, it was found
that the subgrade was ‘unsuitable’ as per NDOT specifications. For this, lime

stabilization was done to the top one foot of the in-situ subgrade soil.

185

0 Higher deflections were observed in stabilized subgrade soil of sections 0201,
0205, 0207, and 0209, compared to that of other sections

0 The PCC consisted of mixes different from the ones stipulated by the SHRP. The
sections that were supposed to have PCC of 550-psi 14-day flexural strength have
a 475-psi mix and the sections with target l4-day flexural strength of 900—psi
have a 750-psi mix. This change was made to the design as it was found difficult
to attain the 900-psi strength stipulated by SHRP, with locally available materials.

0 Sections 0205, 0207 and 0208 had shrinkage cracking in the LCB before paving
operations were canied out.

0 Section 0212 had severe cracking following paving and was removed in 1995.
The section was replaced with nonconforming materials and thus was removed
from the experiment. Also sections 0203, 0205, and 0208 had shrinkage cracks
following paving.

Site Status

Table 26 is the summary of status of the design features at this site. Six of the eleven
sections (see Table 25) have been built with PCC slab thicker than target thickness at
least by a margin of 0.5 in. Also the average 14-day flexural strength of the sections is
more than 10% (of target strength) below the target strength.

Full-depth repairing was conducted on section 0201 in 1999 and 2000. This has been
reported in the database as a rehabilitation construction event. In addition sections 0202

and 0206 have been de-assigned from the experiment in 1997.

186

Table 26 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP m Design Actual Base Type Drama” Width, ft designed?
0201 8 9.2 DGAB No 12 No
0202 8 8.2 DGAB No 14 Yes
0203 11 11.9 DGAB No 14 No
0204 11 11.8 DGAB No 12 No
0205 8 8.5 LCB No 12 No
0206 8 7.8 LCB No 14 Yes
0207 l 1 10.9 LCB No 14 Yes
0208 1 l 1 1 LCB No 12 Yes
0209 8 8.9 PATB Yes 12 No
0210 8 10.1 PATB Yes 14 No
021 l 1 1 1 1.3 PATB Yes 14 Yes
0212 ll - - - - -
Flexural Target Std. Dev. Actual (average) Averggegz’ithm
Strenig‘h’ 550 33 490 No
”S 900 87 730 No
Data Availability

Table 27 is a summary of extent of monitoring data available for the sections at this site.

Table 27 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress IRI FWD
Manual Faulting
0201 7 7 7 6
0202 2 2 2 2
0203 7 7 7 4
0204 8 8 7 17
0205 7 7 7 6
0206 2 2 2 2
0207 7 7 7 4
0208 7 7 7 4
0209 8 8 7 5
0210 8 8 7 6
0211 7 7 7 4
0212 - - - -

 

 

 

 

 

 

187

North Carolina (37)

Site Description

The site is located in the southbound lanes of U. S. 52 near Lexington, N. C. It is a four
lane divided highway. There is an interchange to US 64 on the site and 0204 is on the

south of the interchange. The other inventory data are summarized in Table 28.

Table 28 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 37
Climatic zone Wet No Freeze
Average annual precipitation 1151 mm
Average annual freezing index 47 °C-days
Traffic open date 1st of July 1994
‘Proposed’ traffic 715 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type PCC
Outside shoulder type PCC

 

 

Construction Issues
No major construction issues have occurred at the site, according to the construction
report. The main construction issues are as follows:
0 Shrinkage cracks occurred at several locations in LCB before paving was done.
Cracks were covered with tar paper before paving. Repairing was done to the
PCC slabs that had cracks that reflected from LCB.
o The PATB layer was placed 5” thick instead of 4” thick in 0209 and 0210.
o All sections in the site were constructed with dowels with diameter of 38 mm.

The stipulation requires dowels with diameter of 25 mm for sections with 203

188

mm—thick PCC slab and dowels with diameter of 38 mm for sections with 279
mm-thick PCC slab.

- Cement or lime stabilization was done to top 7 or 8 inches of the subgrade for
sections 0204 and 0207.

0 The DGAB and LCB extended only to two feet from the pavement edge and not
to the shoulder edge as stipulated for SPS-2.

0 Edge drains were located at a two-feet offset from the pavement edge instead of at
8-feet, which is an SPS-2 specification. Stone was used as trench backfill instead
of PATB.

0 Shoulders were made of Econocrete instead of asphalt concrete.

0 In section, 0203, a contraction joint was located in LCB.

0 No compaction was done around the TDR probes in section 0201. This may cause

post construction settlement of the pavement.

Site Status

A summary of the status of the design features at the site is Table 29. Sections 0201,
0202, 0207, 0209 and 0210 were constructed at least 0.5” thicker than their respective
target thickness. The flexural strength data is available only for three sections. Sections
with target 14-day flexural strength of 550-psi have average 14-day flexural strength

higher than target by a margin of more than 10% (of target strength).

189

Table 29 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP ID Design Actual Base Type Drainage Width, ft designed?
0201 8 f 9 DGAB No 12 No
0202 8 8.9 DGAB No 14 No
0203 l l l 1.2 DGAB No 14 Yes
0204 1 1 11.2 DGAB No 12 Yes
0205 8 8 LCB No 12 Yes
0206 8 8.4 LCB No 14 Yes
0207 11 11.6 LCB No 14 No
0208 l l l 1.2 LCB No 12 Yes
0209 8 8.6 PATB Yes 12 No
02 10 8 9.1 PATB Yes 14 No
021 1 1 l l 1.4 PATB Yes 14 Yes
0212 l 1 10.9 PATB Yes 12 Yes
Flexural Target Std. Dev. Actual (avegge) Average Within 10%?
Strength, 550 61 693 No
psi 900 N/A 850 Yes
Data Availability

A summary of monitoring data availability is Table 30. The large extent of data available

for the section 0201 could be because the section is a DLR and SMP section.

Table 30 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress IRI FWD
Manual Faulthg
0201 8 9 9 47
0202 5 6 9 4
0203 5 6 9 4
0204 5 5 9 4
0205 5 6 9 4
0206 5 6 9 4
0207 5 6 9 4
0208 5 5 8 4
0209 7 8 9 7
0210 5 6 9 4
0211 5 6 9 4
0212 5 5 9 4

 

 

 

 

190

North Dakota (38)

Site Description
The site is located in the eastbound lanes of I-94 in eastern North Dakota, west of Fargo.
I-94 is a rural interstate. Table 31 is a summary of other inventory data. The project is

reconstruction of an existing PCC pavement.

Table 31 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

Site code 38

Climatic zone Wet Freeze
Average annual precipitation 545 mm
Average annual freezing index 1313 °C-days
Traffic open date 1st of November 1994
‘Proposed’ traffic 420 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type AC

Outside shoulder type AC

 

 

 

Construction Issues
The site is located on a flat terrain, which is part of the old Lake Agassiz. The extremely
wet clay soils delayed the construction during rains. The project opening was thus
delayed by more than a month. The key observations from the construction report are as
follows:
0 The LCB was hard to place. For this the mix was changed to increase the strength
of this layer.
0 Shrinkage cracks in LCB reflected on the PCC layer in section 0217 and the
cracks were sealed.

0 The PATB was difficult to place, as it was very “fluid”.

191

Site Status

The status of the design features at the site is summarized in Table 32. No data for l4-day
flexural strength of PCC is available in the LTPP database (Release 17). All the sections

have been constructed with actual PCC thickness deviation (from target thickness) less

than 0.5”.

Rehabilitation repairing was done to all the sections of the site. AC shoulder restoration

was done to all the sections in the year 1997. Partial-depth repairing was done to section

0217 in 1998 and in 1999. Partial-depth repairing was also done in 1999 in 0216.

Table 32 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP ID Design Actual Base Type Drainage Width, “ design ed?
0213 8 8.2 DGAB No 14 Yes
0214 8 7.9 DGAB No 12 Yes
0215 ll 11 DGAB No 12 Yes
0216 l l l 1.2 DGAB No 14 Yes
0217 8 7.9 LCB No 14 Yes
0218 7.9 LCB No 12 Yes
0219 l 1 10.9 LCB No 12 Yes
0220 l 1 10.9 LCB No 14 Yes
022 1 8.1 PATB Yes 14 Yes
0222 8 8.2 PATB Yes 12 Yes
0223 l l l 1.1 PATB Yes 12 Yes
0224 l 1 10.8 PATB Yes 14 Yes
Flexural Target Std. Dev. Actual (average) Average Within 10%?
Strength, 550 N/ A N/ A -
psi 900 N/A N/A -
Data Availability

Table 33 is the summary of the extent of monitoring data available for the sections in this

site. Though the site is about 10 years old, the IRI and FWD data is available only for 5

tests each.

192

 

Table 33 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultmgfi
0213 7 7 5 5
0214 6 6 5 5
0215 6 5 5 5
0216 6 5 5 5
0217 7 7 5 5
0218 6 6 5 5
0219 6 6 5 5
0220 6 6 5 5
0221 7 7 5 5
0222 6 6 5 5
0223 6 6 5 5
0224 6 6 5 5

 

 

 

 

 

 

193

I".

Ohio (39)

Site Description

The Ohio site is located in the northbound lanes of U. S. 23 in Delaware County, central

Ohio. The four-lane highway is a rural arterial.

Table 34 Summary of inventory data

 

 

 

 

 

 

 

 

 

Site code 39
Climatic zone Wet Freeze
Average annual precipitation 972 mm
Average annual freezing index 375 °C-days
Traffic open date 1" of October 1996
‘Proposed’ traffic 608 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type AC*
Outside shoulder type AC

 

 

 

 

*Section 0211 has a PCC shoulder

Construction Issues

According to the construction report, no major construction problems and/ or deviations

have occurred at this site.

Site Status

Table 35 is summary of the status of design features at the site. A substantial deviation
from the target flexural strengths is to be noted. The average l4-day flexural strength of
sections with 550-psi as target strength is much higher than the target and that of the

sections with target strength as 900-psi is much lesser than 900-psi.

194

Table 35 Summary of status of design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP 1D Design Actual Base Type Drainage Width, ft designed?
0201 8 7 .9 DGAB No 12 Yes
0202 8 8.3 DGAB No 14 Yes
0203 l 1 10.9 DGAB No 14 Yes
0204 1 1 l 1.1 DGAB No 12 Yes
0205 8 8 LCB No 12 Yes
0206 8 7.9 LCB No 14 Yes
0207 l l 1 1.1 LCB No 14 Yes
0208 l l l l LCB No 12 Yes
0209 8 8. l PATB Yes 12 Yes
02 10 8 8 PATB Yes 14 Yes
021 1 l 1 1 1.4 PATB Yes 14 Yes
0212 l 1 10.6 PATB Yes 12 Yes
Flexural Target Std. Dev. Actual (average) Average Within 10%?
Strength, 550 56 684 No
psi 900 153 614 No

 

 

 

 

 

 

Moreover, the average l4-day flexural strength PCC in sections with target strength of

550-psi is greater than that of PCC of sections with target strength of 900-psi.
Data Availability

Table 36 is a summary of the extent of monitoring data available for the test sections at
this site. Section 0204 is a SMP section. This could be the reason for higher extent of data

available for the section.

195

Table 36 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faultng
0201 5 5 8 8
0202 5 5 8 7
0203 5 5 8 8
0204 6 6 8 21
0205 5 5 8 9
0206 4 4 8 7
0207 4 4 8 8
0208 4 4 8 8
0209 5 5 8 8
0210 5 5 8 7
0211 5 5 8 8
0212 5 5 8 6

 

 

 

 

 

 

196

Washington (53)

Site Description
The site is located in the northbound lanes of SR 395 in eastern Washington. The route is

an urban principal arterial.

Table 37 Summary of other inventory data

 

 

 

 

 

 

 

 

 

 

 

 

Site code 53

Climatic zone Dry Freeze
Average annual precipitation 308 mm
Average annual freezing index 265 °C-days
Traffic open date 1" of November 1995
‘Proposed’ traffic 462 KESALs/year
Subgrade soil type Fine grained
Inside shoulder type AC

Outside shoulder type AC

 

Construction Issues
The major observations from the construction report are as follows:
0 Average moisture content of the subgrade soil was 5.8% below optimum.
- Construction traffic provided compaction effort.
0 The section 0203 that was built on cut has most variation in deflections, as
observed from the FWD testing.
0 Section 0207 had a high compressive strength of LCB compared to other sections,
which could be due to the low water-cement ratio.
0 In sections 0209 and 0212, the embankment soil was accidentally placed on
shoulder and because of this the PATB voids could have been clogged. Also
during paving, the PCC slurry that spilled over the PATB in shoulder, though

scrapped off later, could have clogged the PATB voids.

197

o Patching was done to the fabric of edge drains in 0209 and 0212.

0 Surface voids appeared immediately due to the mix being unconsolidated.

0 Following paving, shrinkage cracks appeared throughout the section 0206.

Site Status

The status of the design features has been summarized in Table 38. Five of the twelve

sections have deviation of at least 0.5” from the respective target PCC thickness. All the

five sections have been built at least 0.5” thicker than the respective target thickness.

Also, the average 14—day flexural strength of the sections with target 14-day flexural

strength of 550-psi is below the 10% (of target strength) error range.

Table 38 Summary of status of design factors

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP ID Design Actual Base Type Dmage Width, ft designed?
0213 8— 8.7 DGAB No 14 No
0214 8 8.3 DGAB No 12 Yes
0215 11 11.1 DGAB No 12 Yes
0216 11 11.2 DGAB No 14 Yes
0217 8 8.5 LCB No 14 No
0218 8 8.6 LCB No 12 No
0219 11 11.1 LCB No 12 Yes
0220 1 1 l 1.2 LCB No 14 Yes
0221 8 9 PATB Yes 14 No
0222 8 8.3 PATB Yes 12 Yes
0223 11 11.8 PATB Yes 12 No
0224 l 1 1 1.3 PATB Yes 14 Yes
Flexural Target Std. Dev. Actual (average) Average Within 10%?
Strength, 550 55 485 No
psi 900 35 831 Yes
Data Availability

Table 39 is the summary of monitoring data availability.

198

 

Table 39 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress . IRI FWD
Manual Faulting
0213 8 8 8 6
0214 8 8 8 6
0215 8 8 8 5
0216 8 8 8 6
0217 8 8 8 6
0218 8 8 8 6
0219 8 8 8 6
0220 8 8 8 6
0221 8 8 8 6
0222 8 8 8 6
0223 8 8 8 6
0224 8 8 8 6

 

 

 

 

 

 

199

Wisconsin (55)
Site Description

The site is located on the westbound and eastbound STH-29, a rural arterial road, in

Marathon County, Wisconsin. The other inventory data are summarized in Table 40.

Table 40 Summary of inventory data

 

 

 

 

 

 

 

 

 

 

 

Site code 55

Climatic zone Wet Freeze
Average annual precipitation 815 mm
Average annual freezing index 998 °C-days
Traffic open date lSt of November 1997
‘Proposed’ traffic 151 KESALs/year
Subgrade soil type Coarse grained
Inside shoulder type No data
Outside shoulder type No data

 

 

Construction Issues
No major construction issues exist for the site. The main issues are as follows:
- Undisturbed soil samples could not be obtained for testing, as the project was a
replacement of a PCC pavement. The PCC slab was removed and fill was placed.

0 A stiff or rigid layer exists below 20 feet.

Site Status

Table 44 is a summary of the status of the design features of the sections at the site.
Section 0222 was constructed 0.5” thicker than the target thickness of 8”. The average
14-day flexural strength of the sections with target l4-day strength of 550-psi is higher by

10% of the target strength.

200

Table 41 Summary of status of the design features

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Slab Thickness, in. . Lane As
SHRP ID Design Actual Base Type Dramage Width, ft design ed?
0213 8 8.5 DGAB No 14 Yes
0214 8 8.4 DGAB No 12 Yes
0215 11 11.3 DGAB No 12 Yes
0216 11 11.1 DGAB No 14 Yes
0217 8 8.2 LCB No 14 Yes
0218 8 8.4 LCB No 12 Yes
0219 11 11.3 LCB No 12 Yes
0220 1 1 l 1.2 LCB No 14 Yes
022 1 8 8 .3 PATB Yes 14 Yes
0222 8 8.5 PATB Yes 12 No
0223 l 1 1 1.3 PATB Yes 12 Yes
0224 1 1 1 1.4 PATB Yes 14 Yes
Flexural Target Std. Dev. Actual (avegige) Average Within 10%?
Strength, 550 28 633 No
pSi 900 53 884 Yes
Data Availability

Table 42 is a summary of the monitoring data available for the sections at the site.
Though the site is about seven years old, the distress data and FWD data are available for

only three tests.

Table 42 Summary of monitoring data availability

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Monitoring data availability, No. of surveys
SHRP ID Distress IRI FWD
Manual Faulting
0213 3 3 7 3
0214 3 3 6 3
0215 3 3 7 3
0216 3 3 7 3
0217 3 3 7 3
0218 3 3 7 3
0219 3 3 7 3
0220 3 3 7 3
0221 3 3 7 3
0222 3 3 7 3
0223 3 3 7 3
0224 3 3 7 3

 

 

 

 

 

 

 

201

REFERENCES

202

Armaghani, J. M. "Factors Affecting Performance of Concrete Pavements." 5th
International Conference on Concrete Pavement Design and Rehabilitation, 33-

45.

Brown, S. (1979). "Design of Pavements with Lean Concrete Bases." Transportation
Research Record, 725, 51-58.

Buch, N., Chatti, K., Haider, S. W., Pulipaka, A., Lyles, R. W., and Gilliland, D.
"Statistical Challenges presented by the SPS-2 Experiment and Data." 8th
International Conference on Concrete Pavements, Colorado, USA.

Cedergren, H. R. (1988). "Why All Important Pavements Should Be Well Drained."
Transportation Research Record, 1188, 56-62.

Chatti, K., N. Buch, et a1. (2005). Final Report, LTPP Data Analysis: Influence of Design
and Construction Features on the Response and Performance of New Flexible and
Rigid Pavements. Washington DC., NCHRP.

Desaraju, P. (2003). "Influence of Design and Construction Features on the Response and
Performance of SPS-2 Test Sections," Thesis, Master of Science, Michigan State
University, East Lansing.

Elkins, G. E., Schmalzer, R, Thompson, T., and Simpson, A. (2003). "Long Term
Pavement Performance Information Management System Pavement Performance
Database User Guide." FHWA.

Fleckenstein, L. J ., and Allen, D. L. (1996). "Evaluation of Pavement Edge Drains and
Their Effect on Pavement Performance." Transportation Research Record, 1519,
28-35.

Forster, S. F. "HIPERPAV- Guidance to Avoid Early-Age Cracking in Concrete
Pavements." Fifth CANMET/ACI International Conference on Durability of
Concrete, Barcelona, Spain, 1109-1122.

Forsyth, R. A., Wells, G. K., and Woodstrom, J. H. (1987). "Economic Impact of
Pavement Subsurface Drainage." Transportation Research Record, 1121 , 77—85.

Gharaibeh, N. G., and Darter, M. I. (2001). "Benefits and Costs of Jointed Plain Concrete
Pavement Design Features." Transportation Research Record, 1778, 1-8.

Grater, S., McCullough, F., and Zollinger, D. (1999). "Developing let Century
Specifications for Portland Cement Concrete Pavements." Recent Developments
in Design and Specification of Concrete Pavement Systems, American Concrete
Institute, Farmington Hills, MI.

203

Hagen, M. G., and Cochran, G. R. (1996). "Comparison of Pavement Drainage Systems.’
Transportation Research Record, 1519, 1-10.

Haider, S. W., “The Use of Long-Term Pavement Performance Data for Quantifying the
Relative Effects of Structural and Environmental factors on the Response and
Performance of New Flexible Pavements,” in Department of Civil and
Environmental Engineering, Ph. D. Dissertation. Michigan State University, 2005.

Hall, K. T. and C. Correa (1999). Estimation of Present Serviceability Index from
International Roughness Index. 78th Annual Meeting of the Transportation
Research Board, Washington, D. C.

Hall, K. T., and C. Correa (2003). "Effects of Subsurface Drainage on Performance of
Asphalt and Concrete Pavements." NCHRP's Report No. 499.

Hanna, A. N ., Tayabji., S. D., and Miller, J. S. (1994). "SHRP-LTPP Specific Pavement
Studies: Five-Year Report." SHRP-P-395, Strategic Highway Research Program,
National Research Council.

Hansen, W., Jensen, E. A., and Mohr, P. (2001). "The Effects of Higher Strength and
Associated Concrete Properties on Pavement Performance." FHWA-RD-00-16I ,
FHWA, U. S, Department of Transportation.

Hansen, W., Mohamed, A. R., Byrum, C. R., and Jensen, E. "Effect of High Strength
Concrete on Pavement Performance." Material Science of Concrete- The Sidney
Diamond Symposium, 191-204.

Harvey, J., Roesler, J ., Farver, J ., and Liang, L. (2000). "Preliminary Evaluation of
Proposed LLPRS Rigid Pavement Structures and Design Inputs." FHWA/CA/OR-
2000/02, Pavement Research Center, Institute of Transportation Studies,
University of California at Berkeley, Berkeley, CA.

Huang, Y. H. (1993). Pavement Analysis and Design. New Jersey, Prentice Hall.

Hudson, W. R. and P. R. Flanagan (1987). "An Examination of Environmental Versus
Load Effects on Pavements." Transportation Research Record 1121, 34-39.

Jiang, Y. J ., and Darter, M. I. (2001). "Structural Factors of Jointed Plain Concrete
Pavements: SPS—2 - Initial Evaluation and Analysis." FHWA-RD-01-167, FHWA.

Khazanovich, L., Darter, M., Bartlett, R., and McPeak, T. (1998). "Common
Characteristics of Good and Poorly Performing PCC Pavements." F H WA-RD-97-
131 , FHWA, U. S. Department of Transportation.

Kutner, M. H., Christopher, J. N., John, N., and William, L. (2005). Applied Linear
Models, McGraw-Hill, Irwin.

204

Mallela, J ., Titus-Glover, L., and Darter, M. I. (2000). "Considerations for Providing
Subsurface Drainage in Jointed Concrete Pavements." Transportation Research
Record, 1709, 1-10.

NCHRP. (2002). "Research Results Digest No. 268."

NCHRP[1-37A] (2004). Guide for Mechanistic-Empirical Design of New and
Rehabilitated Pavement Structures. Champaign, IL, ARA Inc., ERES Consultants
Division.

Nichols Consulting Engineers, C. (1998). Colorado SPS-8 Construction Report, Colorado
Department of Transportation.

Nichols Consulting Engineers, C. (1998). Materials Sampling, Field Testing, and
Laboratory Testing Plan- SHRP SPS-2 Projects, Nevada Department of
Transportation.

Norusis, M. J ., SPSS 12.0: Statistical Procedures Companion. NJ .: Prentice Hall, 2003.

Perera, R. W., and Kohn, S. D. (2001). "LTPP Data Analysis: Factors Affecting
Pavement Smoothness." Project No. 20-50 (8/13), NCHRP.

Peshkin, D. G., Smith, K. D., Darter, M. 1., and Arnold, C. J. (1989). "Performance
Evaluation of Experimental Pavement Designs at Clare." Transportation
Research Record, 1227, 24-33.

Rao, C., Barenberg, E. J ., Snyder, M. B., and Schmidt, S. "Effects of Temperature and
Moisture on the Response of Jointed Concrete Pavements." 7th International

Conference of Concrete Pavements, 23-38.

Ray, G. K. "35 years of Pavement Design and Performance." 2nd International
Conference on Concrete Pavement Design, 3-8.

Ray, M. A. "European Synthesis on drainage, sub-base erodibility, and load transfer."
2nd International Conference on Concrete Pavement Design, 27-39.

Ring HI, G. W. "Drainage of Concrete Pavement Structures." 1 st International
Conference on Concrete Pavement Design and Rehabilitation, 365-378.

Selezneva, O., Jiang, J ., and Tayabji, S. D. (2000). "Preliminary Evaluation and Analysis
of LTPP Faulting Data, Final Report." FHWA-RD-00-076, FHW A.

SHRP. (1986). "Final Report for Strategic Highway Research Program Research Plans."
NCHRP Project 20-20, Transportation Research Board, Washington, DC.

205

SHRP. (1990). "Specific Pavement Studies Experimental Design and Research Plan for
Experiment SPS-2, Strategic Study of Structural Factors for Rigid Pavements."
National Research Council, Washington DC.

Smith, K. D., Mueller, A. L., Darter, M. I., and Peshkin, D. G. (1990). "Performance of
Jointed Concrete Pavements." F H WA-RD-89-I 37, FHWA, U. S. Department of
Transportation.

Smith, K. D., Peshkin, D. G., Darter, M. 1., and Mueller, A. L. "Effect of design features
on concrete pavement performance." 4th International Conference on Concrete

Pavement Design and Rehabilitation, 2344.

Smith, K. D., Wade, M. J ., Yu, H. T., and Darter, M. I. "Design Considerations For
Improved JPCP Performance." 6th International Conference on Concrete
Pavement Design and Materials for High Performance, 89-109.

Springenschrrrid, R., and Fleischer, W. "Developments in the Design and Construction of
Concrete Pavements for German Expressways (Autobahns)." 7th International
Conference on Concrete Pavements, 437-450.

Suthahar, N., Ardani, A., and Morain, D. (2000). "Early Evaluation of LTPP Specific
Pavement Studies-2 in Colorado." CDOT-DTD-R-00-2000-2, Colorado
Department of Transportation, Denver, CO.

Tart Jr., R. G. (2000). "Pavement Distress and Roadway Damage caused by Subsurface
Moisture and Freezing Temperatures- Case Histories from Alaska."
Transportation Research Record, 1709, 91- 97.

Voi gt, G. F. (2000). "Specification Synthesis and Recommendations for Repairing
Uncontrolled Cracks that Occur During Concrete Pavement Construction."
American Concrete Pavement Association, Skokie, IL.

Vongchusiri, K., Buch, N., Desaraju, P., and Salama, H. "An Evaluation of LTPP SPS-2
Sections in Michigan." Maintenance and Rehabilitation of Pavements and
Technological Control, Portugal.

Wade, M. J ., Smith, K. D., Yu, T., and Darter, M. I. (1995). "Monitoring of European
Concrete Pavements." Transportation Research Record, 1478, 90-99.

Wu, S. (1992). "Concrete Pavement Performance: A 23-year Report (Abridgement)."
Transportation Research Record, 1370, 3942.

Wu, S.-S., and Qian, Y. "Effect of Base Type on Concrete Slabs Response to Loading."

6th International Conference on Concrete Pavement Design and Materials for
High Performance, 339-358.

206

111111111111311111111111111611