Beruflich Dokumente
Kultur Dokumente
RECALffiRATION OF THE
ASPHALT LAYER
COEFFICIENT
By
Kendra Peters-Davis
-1 Dr. David H. Timm, P.E.
August2009
By
Kendra P.Davis, Graduate Research Assistant
Dr. David H. Timm, Associate Professor ofCivil Engineering
Auburn University
August 2009
"The contents ofthis report reflecl Ihe views ofllle authors who are solely responsible for
Ihe facls and the aceuracy ofthe dala presenled herein. The conlents do nol necessarily
reflecl Ihe official view and polieies oflhe Nalional Cenler for Asphall Technology of
Auburn University. This repOlt does nol conslitule a standard, specification, or
regulalion."
Pelers-Davis and Timm
ACKNOWLEDGMENTS
The authors wish to thank the Alabama Department ofTransportation for their SUppOlt of
this research through the Southeastern Superpave Center. Also, their longstanding
commitment to the NCAT Test Track made this research possible.
ABSTRACT
The A1abama Department ofTransportation (ALDO}) currently uses the 1993 DARWin
version ofthe American Association ofState Highway and Transportation Officials
(AASHTO) Guide for the Design ofPavement Structures when designing flexible
pavements. The AASHTO design methodology is based on information obtained at the
AASHO Road Test, which was performed from 1958 to 1960 near Ottowa, lIIinois. This
road test provided an empirical correlatiotl between pavement thickness and traffic
loadings. However, the results stemming fi'om the road test are Iimited to the pavement
materials utilized, applied traffic and the climate ofIllinois. Using the results ofthe
AASHO Road Test, a flexible pavement design equation was developed atHl introduced
in the 1986 AASHTO Guide for Design ofPavement Structures that includes inputs of
soil resilient modulus, traffic, structural capacity (structuralnmuber), reliability,
variability, and ride quality (change in serviceability). The strllctural nllmber is
calculated usillg the layer thickllesses, material drainage propelties, and layer
coefficients, which are used to express the relative strength contribution of each
pavement layer to the overall pavement structure. In this study, these inputs were
analyzed to determine the relative influence of each on the resulting hot mix asphalt
(HMA) thickness. It \Vas found that the HMA layer coefficient, resilient modulus, and
traffic inputs are by far the most influential parameters. Since the soil modlllus and
traffic are generally given parameters for a particular design, it was decided that the layer
coefficient be recalibrated to provide the greatest potential savings in HMA thickness.
Fmthermore, the layer coefficient has not been updated to account for advancements
made in construction methods, gradation requirements, and paving materials since the
AASHTO Road Test, and therefore should be reanalyzed.
The recalibration was performed USillg traffic alld performance data collected for
the structural sections of the 2003 and 2006 National Center for Asphalt TeclUlology
(NCAT) Test Track cycles. These data were used in conjunction with traffic equations
developed from the AASHO Road Test as well as the AASHTO flexible pavement
design equatioll to find both the calculated alld predicted equivalent single axle loads
(ESALs). Once these values were found for each section, a least squares regression was
perfonued to determine new HMA layer coefficients. The resulting average layer
coefficient was 0.54 for all sections, with a standard deviation ofO.08. Using this
parameter instead ofthe AASHTO recommended coefficient of0.44 results in an
approximate HMA thickness savings of 18%. From these results, it is recommended that
ALDOT adopt this value as their new layer coefficient for flexible pavement desigl.1s.
ii
Peters-Davis and Timm
TABLE OF CONTENTS
iii
Pele/'s-Davis and Till/II/
IV
Pelers-Davis and Tillllll
LIST OF TABLES
Table 2. 1 [-!MA Layer Coefficients fi'om AASHO Road Test (afler HRB, 1962) .......... 13
Table 2. 2 Minimum Thicknesses (afler AASHTO, 1993) ............................ ....... ............. 2O
Table 2.3 RecOlrunended Design Values (afler Holman, 1990) . .....................................22
Table 3. 1 Vallle Ranges Used for Flexible Design Eqllation Inpllts ............................... 29
Table 3. 2 Correlation between HMA Thickness and Other lnpllts ................................ .30
Table 4. l Thickness Data for the Test Track Sections ................................................... .46
Table 4. 2 Resilient Modllllls Data (afler Taylor, 2008) ...................................................47
Table 4. 3 Selviceability Data for Section NI (2003 Test Track) .................................. ..47
Table 4. 4 Predicted ESALs Applied for Section NI (2003 Test Track) .........................47
Table 4. 5 Axle Weights for 2003 Test Track (afler Priest and Timm, 2006) ................. .49
Table 4. 6 Axle Weights for 2006 Test Track (Taylor, 2008) .............................. ............50
Table 4. 7 Average Axle Weights for 2003 Test Track ....................... ....................... ...... 50
Table 4. 8 Average Axle Weights for 2006 Test Track ...................... ............................. .50
Table 4. 9 Section NI (2003 Test Track) Traffic Calclllation Reslllts ............................. 50
Table 4. 10 Calclllated ESALs for Section NI (2003 Test Track) ................................... 51
Table 4. 11 ESAL Differences Assllming al = 0.44 for Section NI (2003 Test Track) ... 52
Table 4.12 Regression Statistics for Section NI (2003 TestTrack) ................................ 53
Table 4. 13 Regressed HMA Layer Coefficients .............................................................. 55
v
Pele/'s-Davis and Tillllll
LIST OF FIGURES
Figure 2. 1 AASHO Road Test Vehicles (Smith et al., 2004) ................................ .......... 11
Figure 2. 2 41h Power Relalionship between Axle Weight and Pavelllent Dalllage (afier
HRB, 1962) ........................................................... ............................................................. 12
Figure 2. 3 Determining al Based on HMA Modulus (afier AASHTO, 1993) ............... . 14
Figure 2. 4 AASHTO Flexible Design Nomograph (AASHTO, 1993) .......................... .17
Figure 2. 5 Relationship between IRI and PSR for AH Pavement Types (AI-Omari and
Darter, 1994) .......... ........................... .............. ........................ ......... .................................. 19
Figure 3. I General Trend ofHMA Thiekness with Layer Coeffieient (al) ..................... 31
Figure 3. 2 General Trend ofHMA Thiekness with Traffie (W I8) . ... ... ........... .... .. .. .. ...... 31
Figure 3.3 General Trend ofHMA Thiekness with Resilient Modulus (MR) . . ... 32
Figure 3. 4 Gelleral Trend ofHMA Thiekness \Vith Reliability (R) ...... .......................... .32
Figure 3. 5 General Trend ofHMA Thiekness \Vith Selvieeability (.1PSI) ......................33
Figure 3. 6 General Trend ofHMA Thiekness with Variability (So) ............................... 33
Figure 3. 7 Resulting HMA TIliekness fmm Changing Layer Coeffieient (al) ...... ......... .34
Figure 3. 8 Resulting HMA Thiekness from Changing Resilient Modulus (MR) ........... .35
Figure 3. 9 Resulting HMA Thiekness frOIll Changing Servieeability (.1PSI) ................ .35
Figure 3. 10 Resulting HMA Thiekness from Changing Variability (So) ........................ 36
Figure 3. 11 Resulting HMA Thiekness fi'om Changing Reliability (R) .......................... 36
Figure 4. 1 Reealibration Proeedure llluslration ............................................................... 39
Figure 4. 2 Layout oflhe NCAT Test Traek ................................................................. .. .40
Figure 4.3 2003 Test Traek Struetural Seetions (Tilllm el. al, 2004) . ............................ .4 1
Figure 4. 4 2006 Test Track Structural Seetions (Timm, 2009) ...................................... .42
Figure 4. 5 ARAN Ineltial Profiler at NCAT Test Traek ................................................ .43
Figure 4. 6 PSI Data from Seetion NI (2003 Test Traek) . .............................................. .44
Figure 4.7 psr Data from Seetion N3 (2003 and 2006 Test Traek Cycles) .................... .45
Figure 4. 8 Triple Traiter Truek at NCAT Test Traek ..................................................... .49
Figure 4. 9 Box Traiter Truek al NCAT Test Traek. ............. ................................ .......... .49
Figure 4. 10 Sample of Detailed Axle Data ................................ ...................................... 51
Figure 4. II Caleulated vs. Predieted ESALs ................................................ ................... 54
Figure 4. 12 Regressed Layer Coefficients ....................................................................... 56
Figure 4. 13 Treneh Showing Sigus ofDebonding in Seetion NIO .................................. 57
Figure 4.14 Treneh Showing Delalllination ofHMA Lifis in Seetion NI0 ..................... 57
Figure 4. 15 Caleulated vs. Predieted ESALs Using a l = 0.54 .................................... ..... 58
Figure 4. 16 Caleulated vs. Predieted ESALs Using al = 0.44 ....................... .................. 59
Figure 4. 17 Change in Resulting HMA Thiekness frolll al = 0.44 to al = 0.54 . ............. 60
VI
Pefe/'s-Davis and Tilll/ll
CHAPTER 1 - INTRODUCTlON
BACKGROUND
The Alabama Department ofTranspOltation (ALDOT) currently uses the 1993 DARWin
version ofthe American Association ofState Highway and TranspOltation Officials
(AASHTO) Guide for Ihe Design ofPavement Structlll'es when designillg flexible
pavemenls. The AASHTO designmethod is based on informal ion obtained at the
AASHO Road Tesl, which was perfonned from 1958 to 1960 near Ottowa, IIIinois.
This road test provided an empirical correlalion between pavemenl Ihickness and Iraffic
loadings. Equations were developed from Ihe road test lo determine Ihe pavement
thickness required for a patticular design, and although they have been modified
somewhat, are still in use today. Through the road test research and the developed
equalions, Ihe sh'lIctural nllmber (initially lermed the thickness index) was introduced to
define Ihe overall slruclural capacity of a pavemenl cross section. The stmchlral number
for flexible pavement design is malhematically defined by Ihe following eqllation
(AASHTO, 1993):
(1.1)
Where a is Ihe emp'ically based layer coefficient Ihat represenls the sh'lIclmal
capacity oflhe malerialused in Ihe i h layer, Ihe /11 coefficienls are used to describe the
drainage properties of each layer, and Ihe D tenns are the thicknesses of each respective
layer. The slruchlral nUl11ber can also represent the capacity of each individuallayer in
Ihe pavement cross seclion. For exal11ple, Ihe slruchlral nUl11ber for the HMA can be
expressed as:
(1.2)
The same concepl can be used lo calculale Ihe SN for olher pavement layers.
The slruchlral numbers for each layer in Ihe pavement cross section can be summed to
equal Ihe total struchlral number for the pavement as expressed in Equalion 1.1. While
Ihe respective layer thicl<l1esses and drainage conditions are relatively easy to quantify,
Ihe layer coefficienls are not so straightforward. No direct melhod exists for
eslablishing new layer coefficients as new HMA mix types are created, and they are
dependent upon many different parameters including material stiffness, tensile slrength,
compressive slrength, moishlre conditions, and even Ihe layer's position within the
pavement cross section.
The recommended layer coefficient for dense graded HMA mixes (al) is 0.44,
which comes from the results ofthe AASHO Road Test (HRB, 1962; AASHTO, 1993).
ALDOT uses this value for designing flexible pavements, as do many other
h'ansportation agencies. This coefficient is based on the Iimited parameters used at the
road test. The trucks had only bias-ply tires with pressures of around 70 psi. No triple
or quad axles were utilized, and no super singles tires were used . Additionally, a
maximum of2 million equivalent single axle loads (ESALs) were applied over the
course ofthe road test. Olllya limited number of cross seclions were tested, one soil
7
Pelel's-Davis and 1'il/ll/l
Iype was used as Ihe subgrade, one type of gravel was used as Ihe base malerial, and one
type ofHMA \Vas used. The Ihickest HMA pavemenl was 6 inches. The entile road lesl
only lasled (wo years. AIl Ihe results from (he road (est are a product oflhe climate in
norlhern IIIinois. FUl1hermore, no Superpave mixes, open-graded friction courses,
stone-mah'ix asphalls, or olher advanced paving materials were available at the time.
These ne\V mixes provide better performance against rulting, fatigue cracking, and other
distresses, and Iherefore il seems logical Ihat Ihese improved asphalt mixes would have a
higher slructural capacily (SN). Since Ihere are only two inpuls needed lo calculate Ihe
sh'uchlral number oflhe lIMA as shown in Equation 1.2, this implies Ihat Ihe layer
coefficient shollld be higher for Ihese new mixes. Ifa new, and presumably higher,
layer coefficient were established for these improved mixes, then Ibe required HMA
Ihickllesses wOllld decrease. Consequently, Ihis would result in lower malerial and
construclion costs and an overall more efficient pavement sh"llclure.
Due lo Ihis reasoning, it was concluded Ihat there is a need to analyze the current
recommended value for Ihe HMA layer coefficient of0.44. It should be updaled to
reflect Ibe changes and improvements Ihat have occurred in HMA malerials and
conslruclion over Ihe last 50 years. In addition, while a savings in lIMA thickness is
expected, the magnitude ofthat savings is uncertain; therefore, Ihe sensilivity oflhe
layer coefficient 011 the resulting HMA Ihickness should be analyzed as well.
OBJECTIVES
There were Iwo primaly objeelives ofthis investigation . One \Vas lo determine Ihe
sel1sitivity oflhe layer eoeffieient (al) on the reslllting lIMA thiekness using Ihe 1993
AASHTO melhod for flexible pavemel1t designo The seeol1d objeetive was to recalibrate
Ihe layer coefficient for newer mixes, and compare that value to Ihe currently used layer
coefficienlof0.44.
SCOPE
The sensitivity analysis was performed using Ihe 1993 AASHTO Design Guide flexible
pavemellt desigll equalion lo creale a large dalabase of resulting lIMA thicknesses from
changing individual inpuls lo Ihe equalion. This dalabase allowed easy viewing oflhe
trends and relalive intluences ofeach ofthe inpuls, including Ihe lIMA layer coefficient.
The primary objeclive ofthis investigalion was lo recalibrale Ihe layer
coefficienls for current asphalt mixes. Data from the slructural sections ofthe 2003 and
2006 National Center for Asphalt Technology (NCA T) Test Track cycles were utilized
to achieve this objective. The Tesl Track provides unique opporhmities for such
researeh: aeeelerated traffie is applied lo ereate aceelerated levels of pavement damage
in a relatively shOl1 period oftime, which was vely useful for Ihis invesligation.
Detailed traffic and performance data collected over the course ofthe test cycles were
used to perfonn a least squares regression to arrive at new layer coefficients for each lesl
section and recommendations were made for the use of a new lIMA layer coeffieient.
ORGANIZATION OF REPORT
A literature review is provided in Chapter 2 Ihat flll1her describes Ihe AASHTO design
method, ALDOT's pavement design procedure, and the origins ofthe layer coefficient.
8
Pele/'s-Davis and Tillllll
9
Pelers-Davis and Timm
INTRODUCTION
The current flexible pavement design methodology used by the Alabama Depal1ment of
Trausportation (ALD01) is derived fi"om the results ofthe American Association of
State Highway Officials (AASHO) Road Test conducted in the late 1950's. A basic
nomograph and equation were developed fiom the road test results that includes inputs
of soilmodulus, traffic, ride quality (serviceability), reliability, variability, and the
capacity ofthe pavement structure (structural number). Using the aforementioned
flexible pavelllent desigtt equation or nOlllograph in conjunction with these inputs, the
designer arrives at a design thickness for the hot lIlix asphalt (HMA).
To fmd the structuralnulllber ofthe entire pavement cross section or its
respective layers, the layer thicknesses, drainage propelties, and layer coefficients are
needed. Layer coefficients, also called structural coefficients, are used to quantify a
particular layer's relative ability to function and perform within the pavement cross
section. For dense graded HMA mixes, the AASHTO recommended layer coefficient is
0.44, and this value is used by ALDOT for all plant lIlix designs (AASHTO, 1993;
I ALDOT, 2004). Recollllllended layer coefficients for other materials are listed in
Chapter 2. This value is based on results from the very Iimited test conditions ofthe
AASHO Road Test, and has not been recalibrated for newer pavelllent types and other
I variables.
10
I
Pelel's-Dal'is and Timm
I
LANE 1 LOOP 2 LANE 2 LOOP 2
I G KIF'
t,lIO P 4 LOOP 4
18 Klps 32 Klp.
LOOP 5
22 .4 KIlt
LOOP 6
3 0 Kfp.
Results
The results oftbe AASHO Road Test were used to develop the first pavement design
guide, known as the AASHO Inlel'im Guide fol' Ihe Design ofRig id and Flexible
Pavemenls. This design guide was issued in 1961 , and had major updates in 1972, 1986,
and 1993. The 1993 AASHTO Design Guide is essentially the same as the 1986 Design
11
Pele/'s-Davis and Timm
Guide for Ihe design of new fleXible pavements, and is stillused today by many
transp0l1ation agencies, including ALDOT.
TIle primary objective for Ihe AASHO Road Test was lo determine the
relalionship between pavement loading and deterioration. Using replicate cross sections
in different test loops (that were loaded wilh different axle weights), researchers at the
road test were able to view Ihe differences in pavement distresses such as rutting,
cracking, and slope variance, that were caused by increasillg axle loads. The
relatiollship found was an approximate fomih power relationship : a unit increase in axle
weight causes increased damage to Ihe fomlh power. To put Ihis relationship into
context, iflhe axle weight is doubled, it causes approximately sixteen times more
damage to the pavement. Figure 2.2 illustrales Ihe general relationship between loading
and damage found at the road test.
.'"
el)
E
.3
el)
:g>
~
AxleWeight
Figure 2. 24th Power Relationship hetween Axle Weight and Pavement Damage
(aftcr HRB, 1962).
12
Pelers-Davis and 1'illlll/
Where Ihe a lenns are Ihe layer coefficienls, Ihe D terms are Ihe Ihicknesses of each
layer, and Ihe subscripls 1,2 and 3 represent Ihe HMA, base, and subbase pavement
layers, respectively. Ifthe layer coefficients are all equal to one, tllen Ihe thickness
index is simply the total thickness ofthe pavement cross section. However, for the
AASHO Road Tesl investigalion, Ihese paramelers were allowed to vary so Ihat each
pavement layer could have a cerlain capacity per unit Ihickness (HRB, 1962). These
parameters were varied because, for example, a four inch HMA layer contribules
considerably more lo pavement capacity Ihan a fOlll inch subbase layer. From Ihis
concept slems Ihe general definition oflhe layer coefficient: an empirical relationship
belween the layer Ihickness and structural number that expresses a layer's relative
contribulion lo Ihe performance oflhe pavemenl struclure (AASHTO, 1993). The layer
coefficienl depends upon many variables, including Ihe resilienl modullls, layer
thickness, underlying supporl, pos ilion in Ihe pavemenl slruclure, and shess slale
(AASHTO, 1993; Pologruto, 200 1).
Using Ihe Ihicklless index eqllation, several analyses ofvaliance were condueled
on the dala from Ihe AASHO Road Test to delermine the layer coefficients for eaeh
pavement sublayer. Table 2.1 shows Ihe HMA layer coefficienls (al) found fiom Ihose
analyses, Ihe lIumber oftest sections in eaeh loop used to find Ihose coeffieienls, and the
2
1Il0del R values. Loop 1 is not ineluded in Ihe lable beeause it was never hafficked; il
was used to evaluate envirollluenlal impaets on pavements.
Table 2. 1 HMA Layer Coefficienfs from AASHO Road Test (after HRB, 1962)
Loop Layer Coefficlent (a,) Test Sections R'
2 0.83 44 0.80
3 0.44 60 0.83
4 0.44 60 0.90
5 0.47 60 0.92
6 0.33 60 0.81
Based upon this table and other delails tiom the results, Ihe value of 0.44 was
recolllmended for use as the HMA layer coeffieient (HRB, 1962). According to Ihe
authors, a weighled average was laken to arrive at Ihis recomlllended value. However, it
is uncertaiu how these values were weighled.
In 1972, a relationsltip was crealed Ihat linked the layer coefficient lo Ihe elastie
modulus (E) ofthe HMA at 70F, and is shown in Figure 2.3. This graph can only be
used ifthe modulus is between 110,000 and 450,000 psi. The AASHO Road Test
recommended layer coefficient of0.44 corresponds lo a modulus of 450,000 psi
(AASHTO, 1993). In 2006, Priest and Timlll found a relalionship relating temperature
and sliftless for all Ihe slruclural seclions in Ihe 2003 Tesl Track cycle. Using Iheir
relalionship, Ihe average HMA modulus was calculaled as 811,115 psi. Iflhe curve in
Figure 2.3 were exhapolaled out lo Ihis modulus value, lile resulting layer coefficienl
would equal 0.54.
13
Pelers-Davis and l'imm
0.5
- ....----- -
-...,
1:
Q)
0.4
0.3 ~ ---
/
lE
Q)
o
U
~ 0.2
-
::J
U
::J
:>
(J)
0.1
o
O 2 3 4 5
HMA Elaslic Modulus al 70F (10'psi)
(2.2)
Where W" is the number of x axJe load applications at time /, and W"8 is the number of
18 kip axle load applications at time l.
The EALFs for each axle load group are used to find the total damage done
during the design period, which is defined in terms of passes of the standard axle load,
as shown in the following equation (Huang, 2004):
14
Pelers-Davis and Tillllll
Where 111 is the llumber ofaxle load groups, EALFi is the EALF for the ith axle load
group, and ni is the number ofpasses ofthe ith axle load group dUl'ing the design periodo
These basic traffic equations can be used in cOl\junclion with the following
equations thal were also developed fi'om the AASHO Road Test to characterize traffic
for a given flexible pavemenl design (Huang, 2004):
G =10 [4.2- P ,]
1 , g 4.2-1.5
(2.4a)
w: ]
log _" G - -G'
=6.l252-4 .7910g(L x + L,)+4.3310gL, +-' (2.4c)
[ Wll8 Px PIS
Whel'e G, is the log ofthe ratio of loss of serviceability (ride quality) at sorne lime 1 lo
the potentialloss of serviceability al terminal selviceability (P,) = 1.5. The initial
sel'viceability is assumed to be 4.2; this value was typical fol' flexible pavements at the
AASHO Road Test, and is used as the initial value for ALDOT pavement designs as
wel!. P., is a function of design and load variables, Lx is the axle group load in kips, L, is
the axle code (1 for single, 2 for tandem and 3 for tridem), SN is the structuralnumbel',
W" is the number of x axle load applications at time 1, WtlS is the numbel' of J8 kip axle
load applications at time 1, and PIS is the value of Px when L., is equal to 18 and L2 is
equal to one.
1 [ M'SI ]
logW;8 = ZR S +9.3610g(SN + 1) - 0.20 + og 4.21~~: (2.5)
004 + --~o;;-
, (SN + 1),,19
+ 2.3210g M R - 8.07
15
Peters-Davis and Timm
111e reliability level selected is typically based upon the predicted traffic leve!. A
low volmne road (defined as less than 500 ESALs per day travelillg in both directions)
requires 85% reliability, medium volume (between 500 and 1750 ESALs per day)
requires and high volume (greater than 1750 ESALs per day) requires 95% (Holman,
1990). The reliability level seIected corresponds to a standard normal deviate, ZR, which
is calculated using the following equation (Huang, 2004):
16
Pefers-Davis and Ti1ll11/
\
"\
- N .,2
;
n.
<l
.,
.;
"~ -~ i
z
.3
!2 \ ~~ - eZg
:ae ~
1 "u
1'.
f~ ~
~
u
, ~
"
<lO
<
~
1" ~ ~
<n
~
<> <
.; .~
Cl I'~ '\ ~21- '
(
.
",
~
8'
N
M
O O
+ ~ "N
1'" (iS)!) \11'1 ' snlopo~ lUa!l!Sal:l
"l .. 1:;; JOS paqpeOl:j e'\~ IQaIJ3
,
:"'1
: i,-
~
le
~
u i "
~
2 g ~
t
'" I :i!
.
If)
~ ti) o 11 le"
+
'" '" d o
."
lO
O
~
O
-: "
o " (1)
'O
...f -,:,- '"
tE t.l.
O (/) :E <J CI)
+
~
,,;
;: ~ ,
~ (t:IJO!IJ!W) t1M'IUO!IOJlJcldl/ POO'1 JI' \!' I IOU!S
...'" IU IO,,!nb3 dPt "SI 0101 paIOW!I'i3
...
8'
~
O
"
.
")
'"+0
ii <Il
~ 'o:
N
&l
.,
I it;::...
8'
-,
17
Pelers-Davis and Tillllll
The M R ofthe subgrade soil seen in Ihe equalion has been adjusled (o account for
seasonal changes, and is tenned the effective M R. The AASHTO Design Guide (1993)
recommends taking an annual average for backcalculated resilient modulus data and
dividing it by three lo obtain the effective modulus. Tls is done to account for
differences in testing procedures fi'om the road test and Ihe current lesting method using
Ihe falling weight deflectometer (FWD). At the road lest, screw driven laboratory
devices were used to determine Ihe soil stiffness. Due lo Ihe slow response lime of such
devices, the apparent stiffness oflhe soil was vely 10\V (around 3,000 psi). With the
much more rapid loading ofFWD lesting, the moduli are typically around three times
higher, and Iherefore are divided by three to arrive at similar numbers to Ihose used at Ihe
road test.
AI-Omari and Dalter (1994) studied the relationships between serviccability, IRI,
and pavement distresses. Plots werc created that relate IRI to serviceability (PSR) as
shown in Figure 2.5. The data in this graph comes fi'om the slates ofLouisiana,
Michigan, New Jersey, New Mexico, and Indiana, and include flexible and rigid
pavements, as well as combinations ofthe two. The eqllation shown in the figure is for
18
Pete/'s-Davis and Tillllll
all pavement types and for units of mm/m, and is recommended by the National Highway
Institllte (NHI) for use if no state-specifie data is available (Holman, 2003). Based upon
only the flexible pavement data ftom their study, AI-Omari and Darter developed an
equation that converts lIU (in/mile) measurements to PSI:
6~-----------------------------------.
el
1.5+---------=--~~~~------------~
1t-----------~.-----~~----------ER
o
3 7
Note: 1 nun/m = 1163.36 i!lfmile
IR/(mm/m)
Figure 2. 5 Relationsbip betwcen IRI and PSR for AlI Pavemcnt Types (Al-Omari
and Darter, 1994).
A stlldy cOlldlleted in 1994 by Gulen et al. stated that lile eqllation developed by
Al-Omari and Dalter was biased and not statistically correet beeallse it was foreed to pass
through anlRI va/lIe ofzero when PSR was eqllal to 5. The authors developed their OWIl
re/ationships relating IRI and PSI for the state of Indiana and recommended that the
selection ofa model be based upon the needs ofthe end user (Gulen et al., 1994).
In 1999, Hall and Munoz developed relationships for relating lRI and PSI for both
asphalt and concrete pavements. They analyzed data from the AASHO Road Test that
ineluded parameters of slope variance (SV) aIld PSI, and then deve/oped a corre/ation
between SV and IRI for a broad range of road roughness levels. Their findings for
flexible pavements can be expressed mathematieally as:
19
Pefers-Davis and Tillllll
SV = 2.2704(JRl)' (2.9c)
Based upon the similarity ofthe AI-Omari/Darter and Holman equations, it was
decided to focus on those relationships for this study. Since the equation developed by
AI-Omari and Darter is the result of a much larger performance database and is
recommended by the NHI, it was determined thal their equation would be optimal for
converting the nu data to present serviceability values.
Structural Number
As seen in the nOlllograph in Figure 2.4, the designer arrives at a structuralnumber for a
given set of inputs. This structural number is used to find the design lIMA thickness (DI)
for a given base and subbase thickness (D2 and D3, respectively) fi'om the following
relationship (AASHTO, 1993):
Minimum Tbickuess
When designing flexible pavements, minimum thickness values should be used to prevent
impractical or uneconomical designs. TIle 1993 AASHTO Design Guide recommends
the minimum thicknesses shown in Table 2.2 for asphalt concrete and aggregate base
layen; based upon the traffic level (ESALs) ofthe designo
20
Pefers-Davis and Tilll/ll
Scnsitivity to Inpllts
A search was condllcled lo find lhe relalive inflllence of eacJl ofthe inputs in the flexible
pavement dcsign eqllation on the reslllting HMA lhickness. Very little information was
available for this topic. Therefore, it was conc1l1ded that rcsearch should be condllclcd to
determine tlle sensilivity ofthese inputs.
Some information can be in ferred from the design nomograph shown in Figure
2.4 . For example, looking at the far left ofthe nomograph, the reliability scale changes as
the vallle increases. This means that at lower reliability levels (50 to 70%), changing the
design reliability by a single percentage has little effect on the reslllting HMA thickness.
However, at higher levels (90 to 99%), a unit change will have a more apparent affect on
the HMA thickness. This trend of changing scales is also seen on the standard deviation,
resilient modulus, and strllcturalnumber, though the scales change at varying degtees.
The traffic level does not demonsh'ate the same trend because it is on a lag scale between
the two tmuing tines. In other words, for traffic, it is the arder ofmagnitude tha! is
critical rather than relatively small differences within an arder ofmaguitude.
ALDOT PRACTICE
To design flexible pavements in the state of Alabama, ALDOT currently follows lhe 390
Procedure (ALDOT, 2004). This procedllre provides guidance for condllcting pavement
designs according to the 1993 AASHTO Desigu Guide, and inc1udes additional
information for materials tesling and obtaining h'affic infonnation. The end reslllt ofthe
390 procedure is a "materials repOlt" that contains traffic data, materials test reslllts, and
printollts detailing the pavement stnictural design from the DARWin software, which
is the embodiment ofthe 1993 AASHTO Design GlIide. The h'affic data and materials
test reslllts are inpllts to the DARWin software, and lhe output obtained from lhe software
is lhe pavemenl Shl1ctllral desigu.
ALDOT also has anolher pavement design guidance document, Guidelines for
Flexible Pavelllenl Design in Alaballla that describes how to determine each ofthe inputs
for the flexible pavement desigtl equation to be entered into the DARWin software.
Fmther details can be obtained from the original document (Holman, 1990).
The current ALDOT flexible pavement design procedlll'e, employed tlu'ollgh
DARWin, utilizes Equation 2.5 and the corresponding nomograph shown in Figure 2.4.
ALDOT currently uses default recommended values for the layer coefficients, change in
serviceability, reliability, and variability inputs (Hollllan, 1990). These values are
sUllllllarized in Table 2.2. The relllaining inputs of drainage coefficients, traffic, and
subgt'ade resilient modulus, are calclllated on a project-by-project basis. The h'affic
ranges used for reliability and sel'viceability are the same as those discussed previollsly.
21
Pete/'s-Davis and Tilll/ll
Where Si is the level of saturation and Dq is used to describe the drainage quality.
The design traffic (ESALs) is ealculated for each project from the average annual
daily traffie and pereent trueks fol' the designo To quantify the traffic, h'uck volumes are
ealculated at various locations withill the project lenglh atld an average is computed. The
truck volume fi'om the location closest to, but just over, the average value is used as the
design value. This design h'affic value is then multiplied by several factors (365
days/year, Lane Distribution, Directional Dish'ibution, Growth Factor, EALF) to alTive at
the design ESALs (Holman, 1990).
The ALDOT procedure (Holman, 1990) has a provision for estimating subgrade
soil modulus (MR ) from California Bearing Ratio (CBR) testing according to the
following equatioll:
M R = 10(O.S>J.I,gCRR+2.911) (2.12)
22
Pelel's-Davis and Till1/11
ALDOT assumes Ihis value lo remain Ihe same Ihroughout the year unless Ihere are data
lo Ihe conlrary (Holman, 1990). The CBR value is found fiom Ihe soil in Ihe saluraled
condilion, and Iherefore is expected lo be Ihe worst case scenario for Ihe soil subgrade.
In addilion lo using CBR dala lo eslimale Ihe resilient modulus, ALDOT has been
conducting exlensive triaxial resilient modulus lesls (AASHTO TI07) oftheir subgrade
materials for the pasl seven years. Procedure 390 slates Ihal Ihe design M R for soils
classified as A-I, A-3, A-2A, and A-2-5 shall be the averageMR values generaled by
AASHTO TI07 at a confining pressure of 4 psi alld oplimum moislllre conlent. For soil
classes A-6 01' A-7 (A-7-5 01' A-7-6), 2 psi confining pressure is lIsed, and samples are
compacled on Ihe wel side of optimum moisture to generate more conselvalive design
soil moduli. For all other soil classes, the design M R used is Ihe average M R value
generaled al 2 psi confining pressure and optimum moisture content (ALDOT, 2004).
23
Pelers-Davis and Timm
In 1989, Coree and White determined layer coefficients for ten bihuninous mixes
as single values ami as distributions. These mixes were typically IIsed by the Indiana
Department of Highways (IDOH) . They used Odemark's principie of equivalent
stiffness, which compares different material types using a ratio of strength parameters.
Applying this principIe to find a new layer coefficient reslllted in the following equation
(Coree and White, 1989):
Y,
{I/OOH = Q MSHO [E
I
DOH
] (2 .13)
E "'SHO
Where alOOH is the desired layer coefficient for an IDOH mixhlre, aAASIIO is the AASHO
Road Test layer coefficient, and EJ/JOH and EAASHO are the moduli of an IDOH asphalt
mixhlre and the AASHO Road Test asphalt mixhlre, respectively.
Equation 2.13 can be solved for a single value ofthe layer coefficient. To
determine the coefficient distributions, equations termed the PoellBonnaure et al.
relationships were IItilized. A fuI! description ofthese equations anel concepts is beyond
the scope ofthis report, but can be fOllnd in the reference (Coree and White, 1989).
Distriblltions ofparameters were created for the AASHO Road Test and the IDOH
mixtures. 111ree approaches were IIsed to determine the layer coefficients, including a
deterministic method, a probabilistic method, and a method combining both deterministic
and probabilistic methods. The detenninistic method used single values fol' each
parameter in Equation 2.13, while the probabilistic method used distributions for all
parameters in the same equation. 111e combination method used dish'ibutions for the
mixhlre stiffuess variables only. Using the deterministic approach, the average layer
coefficient was 0.44. From the strictly probabilistic approach, the average coefficient
was 0.53 with an average standard deviation ofO.287. The combination approach
resulted in an average layer coefficient ofO.47 with an average standard deviation of
0.101. The authors cautioned agaiost using the results ofthe probabilistic approach since
the idea of a layer coefficient being represented by a distribution is not widely accepted.
They stated that the larger values for the layer coefficients of ID OH mixtures compared
to those of the AASHO Road Test could be attributed to the mixes being stiffer; the
penetration grade ofthe AASHO Road Test mixes was 85 - 100, alld the IDOH mixes
were classified as 60 - 70. Fillally, the authors think that the shift in resulting layer
coefficients from the three different approaches was due to the dish'ibutiolls being
asymmetrical.
To determine the layer coefficients for crumb-rubber modified (CRM) mixes for
the Kansas Dep31tment ofTranspOltation (KDOT), Hossain et al. (1997) used falling
weight deflectometer (FWD) data collected from testing the in-place pavement structure.
Three backcalculation methods were used to convelt the deflection data fi'om 41 different
test locations illto moduli. Good agreel1lent was found between the methods, and the
moduli were determined to be sound. To detel1nine Ihe layer coefficients from the
moduli, the sh'uctural nUl1lber was found using the following equation recoml1lended by
the AASHTO Design Guide (1993) :
24
Pelers-Davis and Tillll/l
Where D is the total thicklless of the pavement cross sectioll aboye Ihe subgrade (in
inches), and El' is the effective modulus oflhe pavement cross section aboye the subgrade
(in psi). Using the backcalculated moduli and Ihe layer thicknesses, the layer coefficients
were calculated for Ihe CRM mixes.
Allother method was used to calculate Ihe layer coefficienls for comparison to Ihe
. results ofthe AASHTO method. This method was termed the eqllal mechanistic response
(EMA) and used layer equivalency concepts (similar to the procedure ofVan Wyk et al.,
1983) to find another set of layer coefficients. The average layer coefficient for CRM
mixes found using the AASHTO method was 0.28. The average layer coefficient fiOIn
the EMA method was 0.33. From this study, a pal1icular layer coefficient was not
recommended for use in desigu. However, an equation was developed to estimate the
layer coefficient for CRM mixes, and is defined as follows (afier Hossain et al., 1997):
I FWD data were gathered fiom the Slrategic Highway Research Program (SHRP)
database . Seasonal data, collected over more than 5 years, were used for eight different
locations Ihroughoul the state ofVermont to determine the procedure for finding the layer
coefficients. The effective struclural number (Equation 2.14) was calculated for the
backcalcnlated moduli and the results were plotted against the days oflhe year from
25
Pelers-Davis and Tillllll
which they were calculated. The resulting graph illustrated that for Vermont, the SN'ff
remained fairly constant fi'om May to October. High variations were found during the
spring-thaw periodo Therefore, only the data fi'om the stable months were used to find
Ihe layer coefficients. A pilot project was created to determ ine the resulting layer
coefficients for one test section to verifY the vaJidity of the recommended procedure. To
find the layer coefficients for each respective layer in Ihe pavement structure, the SN'ff
was calculated immediately aboye and below the layer in question. The difference
between these two values was found and then divided by the layer thickness. The SNeffat
the top ofthe subgrade was defined as zero. The results ofthe pilot project found layer
coefficients within the AASHTO range for all materials except the HMA, which was
0.639. This value, although approximately 50% higher than the recoillmended 0.44, was
not discounted from the results because the Marshall stability and resilient modulus for
the material were well beyond the AASHTO recommended upper Jimits for each
parameter. Since the reslllts were encouraging, FWD lesting was performed at three
other test sites to find layer coefficients. A total of25 individual testing locations were
used in the author's analysis. Elastic layer simulation (ELS) was used to verifY the layer
coefficients found fi'om the AASHTO procedure, and the two methods were compared.
From the AASHTO method, the average HMA layer coefficient was 0.60. The average
found lIsing the ELS was 0.59. From these results, the author reconunended using the
AASHTO method for determining new layer coefficients, and that further research
should be performed to see if similarly high HMA layer coefficients are found for other
HMA types and in other locations (pologruto, 2001).
In 2005, Jess and Timm used the AASHTO procedure to find layer coefficients
fi'om backcalculated moduli at the NCAT Test Track in Opelika, Alabama. The
AASHTO method was used (Equation 2.14) to find the effective struchlral numbers for
26 different test sections. A control test section with a HMA layer coefficient of0.44
was used to determine the comparison layer coefficients for the other test sections. Prom
this shldy, the average layer coefficient found was 0.59 with a standard deviation ofO.13.
Additionally, the sections inclllded in the shldy were unusually thick; on the order of24
inches ofhot mix asphalt. Because ofthis, the sections did not experience significant
pavement distress alld conseqllently the struchlral coefficients were merely caJibrated to
sllrface deflection and not changes in serviceability. Therefore, it was decided to
recommend a conservative value based on the average minus one standard deviation
which resulted in 0.46 (Jess and Timm, 2005). Obviously, switchillg from 0.44 to 0.46
do es not significantly alter pavement cross sections in designo
Harold Von Quintus conducted a study on layer coefficients for the Kansas
Department ofTranspOltation in 2007. He concluded that the HMA layer coefficient for
the wearing surface and base mixtures shollld be increased. He stated that the magnitude
ofthe increase should be dependent upon a detailed analysis of material properties,
constructions record s, and pavement performance, and not solely on the HMA modulus
(Von QlIintus, 2007).
SUMMARY
This Jiterature review briefly discussed the AASHO Road Test and its Jimitations, along
with the findings fi'om the road test. The origins ofthe layer coefficient were explained,
and the recommended value for the HMA layer coefficient was discussed. The AASHTO
26
Pelers-Davis and 11111111
traffie and flexible pavement design equations amI their respective inputs were presented
alld described. An explanatioll was provided for ALDOT's flexible pavement design
procedure, and default values were given that are eommonly used by ALDOT for the
AASHTO design equation. Finally, the proeedures and results fi'om past layer eoefficient
ealibration effOlts were provided. As noted iu this ehapter, there was a lack of
information available on sensitivity ofthe AASHTO equations to their inputs. Therefore,
the following chapter presents a sensitivity analysis as part ofthis investigation.
27
Pele/'s-Davis and Ti/ll/ll
INTRODUCTlON
The 1993 AASHTO Design Guide flexible pavement design equation reqllires several
inpllts to find a reslllting HMA thickness. These inputs obviously all affect the HMA
thickness since they are inclllded in the calculation to find it; however, obtaining the
relative influellce of each would be a valuable tool for optimizing pavement designs. A
sensitivity analysis of all the inputs was perfonned to achieve thal goal. The reslllts can
provide a pavement designer with the knowledge ofwhich inputs have the greatesl
influence on the HMA Ihickness, and which have the leasl. Such infonnation could be
used to determine further research needs for the more influential input s, 01' the need lo
better characterize those inputs for the greatest benefit to cost ratio.
METHODOLOGY
Fundamental Equatiolls
A sensitivity analysis was conducted to determine which parameters in the AASHTO
Design Gllide (1993) flexible pavernent design equation are the most influential on the
thickness ofthe hot mix asphalt (HMA). Eqnation 3. 1 shows the parameters needed to
arrive at the HMA thickness (DI), which comes from the strllctural nllmber (SN) shown
in Equation 3.2. The other parameters include the number ofESALs over the design
period (Ws), the design reliability (R), Ihe allloullt ofvariability (So) in the design, the
soil 1II0dlllllS (MR), and the expected loss ofserviceability over the lifetime ofthe
pavement (/JPSI) .
I [ I1PSI ]
IOgW8 =ZRSo + 9.3610g(SN + 1) - 0.20+ og 4.21~~ (3.1)
0.4+ --~-=
(SN +1)'.19
+2.3210gM" - 8.07
Inputs
Before conducting the sensitivity analysis, a baseline pavement condition was
created to determine the relative influence of altering the inputs. A three-layer pavement
cross-section ofHMA, granular base, and sllbgrade soil was used. For Ihe base layer of
granular material, a thickness (D2) of 6 inches was llsed with a layer coefficient (a2) of
0.14. The layer coefficient is typical for crushed stone aggregate bases . A value of 1.0
was llsed for the drainage coefficient (/112)' This value was chosen for s implicity, but is
28
Pele/'s-Davis and Tillllll
also fi'equently used for ALDOT pavement designs. It was assumed that no subbase
layerwas used, so the final tenn in Equation 3.2 was zero. AII other parameters in (he
design equation were varied to determine theil' relative influence on HMA thickness.
Table 3.1 shows the values used fol' eacb input. 111ese values wel'e chosen based upon
typical values used fol' flexible pavement design, and extremes wel'e added to get a wide
I'ange ofthicknesses and to characterize trends.
To obtain the HMA layer thickness using the flexible design equation, fust the
stl'llctural number was found using the bisecnon method. This method was needed since
equation 3.1 is diflicult to solve explicitly. Once a structural number was computed,
Equation 3.2 was rearl'anged to solve fol' DI, which is the HMA layer tbickness. Using
this concept and the varied parameters as shown in Table 3.1, the HMA thickness was
I'ecalculated each time an input was changed. Thel'e were five inputs used for the
variability, and four inputs used fol' the other inputs (traflic, I'esilient modulus, reliability,
change in serviceability, and layer coefficient). 111erefore, a total of 5 x 45 = 5,120 HMA
thicknesses were calculated for tbis investigation.
29
Peters-Davis and Timm
Input Trcnds
To determine the general trend of each input on lhe resultillg HMA thick:ness, a few
poinls were selected ofthe 5,120 available for each input and were plotted to view the
relatiollship. These plots were not meant to exactly quantity the relationship of each
input with the HMA thickness, but rather to view the general h'end and finther illustrate
the relative influence of each input on the HMA thiclmess.
The thickllesses found from alteril1g the layer coefficient for the HMA ({I)
generally follow a llegative trend as shown in Figure 3.1. The graph ShOWll is for a set
traffic level (le8 ESALs), resilient modulus (20,000 psi), variability (0.40), reliability
(80%), and change in serviceability (2.0) . To generate graphs for the other inputs, the
layer coefficient was set to 0.44. Figures 3.1 through 3.6 are ordered fi'om the most
influential input to the least (as in Table 3.2), and observation ofthe graphs provides a
better understalldil1g ofthe correlation results (i.e., that al bas the 1Il0st il1fluence and So
has the least). For example, the overall magnitude of difference in HMA thickness fi'om
changing the layer coefficient is approximately 12 inches (Figure 3.1), while changing
the variability only creates an overall difference of 1 inch in HMA lhickness (Figure 3.6).
30
Pelel's-Davis and Tillllll
18
&12
16
14 '""'" ~
.,'"'" ~
1-
~
J:
1:
:c'"
U
10
6
----- ---
4
o
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
a,
Figure 3. 1 General Trend of HMA Thiclmcss with Layer Coefficient (al)'
18
16
14
----
;[ 12
.,'"'"
------
1: 10
:c'"
----
U
1-
8
~ 6
~
J:
2
:.-----
o
1.E+06 1.E+07 1.E+08 1.E+09
ESALs
31
Pelers-Davis and Tillllll
18 , ---------- ---------------------------------,
16 \
14 '"
~ 12 "'-....
w ~
~ 10 ____
+-------=--~= - -- - -- - - -- - - - - - - ---j
~ 8t-------------===~~~~~~==::::======i
~
:I: 6 +----------------------------------------~
4 +-----------------------------------------~
2 +---------------------------------------------~
O +-------.--------~------_,~------_.--------r_~
18
16
14
:5 12
'"
w
e 10
...
CI>
o
:E
.... 8
oC(
:!ii 6
:I:
O
50 60 70 80 90 100
Reliabilily (%)
32
Pelel's-Davis and Timm
18.0
16.0
14.0
. 12.0
''e""
el> 10.0
-"
:c"
1-
8.0
<1:
:;; 6.0
X
4.0
2.0
0.0
1.0 1.3 1.5 1.8 2 .0 2.3 2.5
DPSI
18.0
16.0
14.0
~ 12.0
'"'e"" 10.0
.l<
:c"
1-
8.0
<1:
:;; 6.0
x
4.0
2.0
0.0
0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
S.
Figure 3.6 General Trend ofHMA Thiclmess with Variability (So).
33
Pefel's-Davis and Timm
Input Dependellcy
Dllring Ihe analysis, it became apparent Ihal all ofthe inpllls are dependen! llpon Ihe
Iraffic leve!. For example, Ihe layer coefficien! value has more of an influence on the
resultillg HMA thickness as Ihe Iraffic level increases, as illuslrated in Figure 3.7. This
ehar! shows how Ihe HMA thickllesses change on average for Ihe different traffic levels
speeified whell Ihe layer coefficient is varied. This process was repeated for all Ihe inpuls
(Figures 3.8 Ihrough 3.11) exceptthe traffic level because Ihe relative influence of each
olher input changed only as Ihe traffic level changed.
40
35 r-
'2 30
~
''"~" 25
'"u 00.2
:c 00.3
... 20
:;; 00.44
:I: r- r"" 00.6
8, 15 I-f--
.
I!!
>
10 C- -
1- - r-
f-
!',. r- r-
~h
5 -
1-, r- _ -
!... 1-
O
!,: ...
1,000,000 10,000,000 100,000,000 1,000,000,000
ESALs
34
Pefers-Davis and Timm
40 ,------------------------------------------------,
35+------------------------------------------------;
.E 30 +--- -- - - - - - - - - - - - - - - - --1
.,.,
:g 25 +--- - - - - - - - - - - -- - - - -- - -1
><
u 03000
:;;
r 20+-- - - - - - - - - - - ---;
____-1 010000
~ 020000
:I: 030000
~ 15 +-- - - - - --;
f
~
..: 10
40 , -----------------------------------------------,
35+--------------------------------------------------1
.E 30+--------------------------~
'""
~ 25+-------------------------~1_1
Jiu 050
:;; 080
r 20+-- - - - - - - - - - - - - - - - -- - - --1
~ 090
:I: 099
~15+----------------.~
f
~
..: 10 +----------------....--1
5 +-..----1.
35
Pelers-Davis and Tillllll
40 ,-----------------------------------------------,
35+-------------------------------------------------~
1 304----------------------------------------------------4
!l
" 25 t -----------------------------------------, -If-------~
.l1u 01
:c
~ m 01~
~ 02
~ OU
~15+---------------------------~
I!
~
~ 10+--------------1
40,-------------------------------------------------,
35+_--------------------------------------------~
~30+-------------------------------------------------~
lil
~ 25t---------------------------------------------------~
,. 00.2
:E 00.3
~ 20 00.4
~
~
00.5
~ 15 il O.6
I!
~
- 10+-------------~~~
36
Pelel's-Davis and Tilll/ll
It is important to note several h'ends from tbese data. First, the layer coefficient
(Figure 3.7) and resilient modulus (Figure 3.8) seem to follow similar h'ends; namely,
they have a greater effect on the HMA Ihickness as the traffic level increases. For
example, at the 1 million ESAL traffic level, going fi'om a layer coefficient ofO.2 lO 0.3
causes a tbiclmess decrease ofapproximalely 4 inches, whereas Ihat same shift at Ihe 100
million ESAL traffic level causes a decrease ofapproximately 9 inches (Figure 3.7). This
Irend can also be seen in Ihe reliability (Figure 3.9) and change in serviceability (Figure
3. 10) chalts, Ihough not lo Ihe same exten!. It is also slightly apparent in the variability
chart (Figure 3.11), but it is even less obvious than tbe other charts. By looking at these
graphs, it is evident that the traffic level does have a significant impact on the calculated
HMA thickness, and that the otber inpuls are dependent upon it to varying degrees. The
layer coefficient and resilient modulus (along with the traffic level) have the greatest
influence on the resulting HMA thiekness; the reliability, variability, and ehange in
serviceability are not as significan!. This is apparent in Ihe grapbs, and further iIlustrates
Ihe con'elation results found previously (Table 3.2), as well as the general Irends seen in
Figures 3.1 through 3.6.
SUMMARY
A sensitivity analysis was perfoJ'lned on Ihe variables ofthe 1993 AASHTO Design
Guide flexible pavemenl design equation lo see which inpuls had Ihe greatest influence
on the resulling HMA Ihickness. A eorrelation between the inputs was performed lo
analyze the signifieanee of Ihe Pearson sample R values. The eorrelation results showed
the layer eoeffieienl was Ihe most influential on the HMA Ihiekness, and Ihe variability
was the least influential. Between those inpuls (in order fi'om more influential to lea sI)
were Ihe traffie, resilient modulus, reliability, anel ehange in servieeability. Plots were
ereated for each input versus the resulting HMA thiekness to observe general hends.
Input dependeneies were fo und in the sensitivity analysis. The relative influence of all
inputs was dependent upon the traffie leve\.
Tbe sensitivity analysis results proved the HMA layer coeffieient to be the most
influential parameter on the resulting design HMA thickness. The two parameters with
similar magnitudes of influence, the traffie level and resilient lllodulus, are both generally
sel parameters for a given pavemenl designo Therefore, this increases the impOltance of
being able lo aceurately charaelerize the HMA layer coefficient. This input is the only
one ofthe tluee mosl influential Ihat can be changed; consequently, better
characterizalion of it would provide the greatest potential savings in HMA thickness.
37
Pelers-Davis and Timm
INTRODUCTION
This investigation used pavement perfonnance and detailed traffic data collected from lhe
Natiollal Center for Asphalt Teclmology (NCAT) Pavement Test Track 1iom the 2003
and 2006 test cycles to recalibrate the HMA layer coefficient (a/) used in the 1993
AASHTO Design Guide flexible pavemellt design equation as illustrated in the flowchall
in Figure 4.1.
As shown in Figure 4.l , there were two data sets needed for calibration: the traffic
and slll'face performance data. Both data sets were needed to calculate the ESALs
applied al the Test Track (an estimation ofthe actual ESALs), and only the performance
data set was needed to calculate the predicted ESALs. Before the perfol1llance data cOllld
be used, it first had to be converted from IRI (in.lmile) to serviceability (PSI). This was
achieved using the AI-OmarilDarter eqllation. Once convelted, these dala were used to
create plots for each test section that illushaled the change in PSI over time. Points were
selected from these chal1s to obtain terminal serviceability levels (P,) for recalibration
purposes. lllese PI values were used to find the calculated ESALs as iIlustrated in the
Figure 4.l.
The PSI versus time charts were al so used to find the change in serviceability
(L/PSI) needed to calculate the predicted ESALs. This was achieved by taking the
terminal serviceability points (P,) just described, and subtracting them fiom the initial
serviceability level (Po) for each test seelion.
Thc other primary input, the structural number (SN) was necessary fol' the
calculation ofboth the calculated and predicted ESALs. lllis number was originally
caleulated using thiekness data fiom each test section as well as an assumed HMA layer
coefficient (a/) of0.44. This layer coefficient comes from the AASHO Road Test and is
currently used by ALDOT fol' flexible pavement designs.
To find the ealeulated ESALs, the traffie data set, terminal serviceability, and
slmetural number were used as inputs for the traffic equations derived from the AASHO
Road Test as iIlustrated in Figure 4.1. Tite predicted ESALs were calculated using the
change in serviceability and structuralnumber as inputs to the AASHTO flexible
pavement design equation. Once found, the calculated and predicted ESALs were
compared and a simple lineal' least squares regression was performed. The error was
minimized between the two data sets by changing only the HMA layer coefficient. Tltis
process was repeated for each sltuctural section ofthe 2003 and 2006 Test Track cycles,
resulting in a new regressed layer coefficient for each section. A more detailed
discussion ofthis process is provided later in this chapter.
38
Pelers-Davis and Tillllll
I
(AI-Omari/Dalter eqllation)
(AASHTO
'-----1
1
SN 11--- flexible pavement
design equation)
al
T
j
Calculated ESALs I Predicted ESALs
(W..,jWtl8 equation) I (logW 18 equation)
I
l'
*Minimize error between ealeulated and
predieted by ehanging only al
Caleulated ESALs
39
Pelel's-Davis and Timm
TEST FACILlTY
Overview
The recalibration ofthe layer coefficients was performed llsing data Ii'om Allbllrn
University's NCAT Pavement Test Track located in Opelika, Alabama. The Test Track
is a 1.7 mile oval that is divided into 46 sponsored sections that are 200 ft long as shown
in Figure 4.2. 111e Test Track provides sponsors with a facility that SllppOrtS hot mix
asphalt (HMA) research by applying live traffic in an accelerated testing envirollment.
The test sections consist ofvarying pavement cross sections and materials based llpon
each sponsors' needs. Live traffic is applied 16 hOllrs a day for 5 days a week, which
adds up to approximately 10 million ESALs over a 3-year test cycle. The data used in
this investigatioll were from the 2003-2006 and 2006-2009 test cycles. A full description
ofthe 2003 and 2006 Test Track cycles are beyond the scope ofthis report, but have been
documented elsewhere (Timm et al., 2004; Timm, 2009).
North Tang~nt
4
Trame N
ir
!
~
SI S2 S3 S4 ss s.
Soulh Tangeot
S7 S8 s. SIO Sil
+SI2 S13
~
Structural Expcl'iment
In the 2003 Test Track cycle, there were eight sections that comprised the structural
experiment. The structural experiment inclllded sections with embedded insh'umentation
to more accurately characterize pavement response under h'affic loadings. These sections
were designed with varyillg thicknesses and materials as shown in Figure 4.3. 111e
thicknesses were varied to provide a wide array of dish'esses to analyze at the elld of the
test cycle. As seen in Figure 4.3, each section shared the same subgrade material that
was already present at the Test Track. This soil, commonly termed the "Track soil", can
be classified as an AASHTO A-4(0) soil type (Ti.mm, 2009). The test sections also
shared the same 6 inch crushed aggregate base course. Sections NI, N4, and N5 used
modified PG 76-22 HMA layers of 5, 9, and 7 inches, respectively. Sectiol1s N2, N3, and
N6 were also 5, 9, and 7 inches thick; however, they were unmodified PG 67-22 HMA
layers. Section N7 consisted of6 inches ofunmodified PG 67-22 HMA placed under a l
inch thick layer ofPG 76-22 stone matrix asphalt (SMA). Finally, section N8 had a 2
inch thick rich bottom PG 67-22 layer with an additional 0.5% binder, which was placed
under 4 inches 6funmodified PG 67-22, and then topped with I inch ofSMA (Timm et.
al,2004).
40
Pelers-Davis and TiIllI/1
Test Section
N1 N2 N3 N4 N5 N6 N7 N8
o
5
10
15
.S
E
a.
20
ID
o
25
30
_ Modmed HMA (PG 76-22)
35 c:::::J Unmodifed HMA (PG 67-22) ~ Unmodified HMA (PG 67-22), Opl +0.5%
1,,,,,,,1 Improved Roadbed (A-4(O)) Soil W...!~?4 Crushed Aggregale Base Course
Figure 4. 3 2003 Test T..ack Structural Sections (fimm et. al, 2004).
For the 2006 Test Track, five ofthe original eight structural sections (N3 through
N7) were left in place fi'om the 2003 cycle; however N5 was milled and inlaid with a 2
inch asphalt layer to control top-dowll cracking that was present throughout the sectioll.
Three ofthe sections were reconstructed (NI, N2, and N8), and three new sections were
added to the structural experiment (N9, NIO, and Sil). The cross sections ofthe 2006
Test Track str'uctural sections are shown in Figure 4.4 along with their respective
sponsors. As seen in the figure, all sections other than N8 and N9 still utilized the Track
soll as subgrade material. The subgrade material used in sections N8 and N9 can be
c1assified as an AASHTO A-7-6 soll, and is known as Seale subgrade material since it
was impOlted fi'om a bOlTOW pit in Seale, Alabama (Taylor, 2008).
In sections NI and N2, 10 inches ofFlorida Iimerock material was used as the
base layer. Both sections were topped with 7 inches ofHMA: unmodified PG 67-22 for
I alllifts ofsection NI , and a 3 inch liftofthe same unmodified binder followed by two 2
inch lifts with modified PG 76-22 binder for section N2. For sections N3, N4, N5, N6,
N7, and SIl, 6 inches of granite aggregate base material supplied fi'om Vulcan materials
was used as the base layer. Sections N3 and N4 both consisted of 9 inches of HMA. N3
./ used unmodified PG 67-22 binder in all HMA Iifts, and N4 used modified PG 76-22 in
alllifts. N5, N6, and N7 contained 7 inches ofHMA in each section: section N5 had a
HMA layer created with unmodified PG 67-22 binder placed over modified PG 76-22
layers, section N6 was comprised ofunmodified PG 67-22, and section N7 contained 6
inches ofunmodified binder topped with a I inch lift ofPG 76-22 SMA. Section N8
consisted of approximately 6 inches ofTrack soil as the base layer, followed by one layer
ofHMA with PG 64-22 binder designed with 2% air voids, one layer ofHMA with PG
41
Pelel's-Davis and Tillllll
64-22 binder, one layer ofHMA with PG 76-28 binder, and finally topped \Vith PG 76-28
binder SMA for a totalHMA thickness of 10 inches. Seclion N9 used approximalely 9
inches ofthe Track soil as base material, follo\Ved by one layer \Vith PG 64-22 compacted
to 2% air voids, t\Vo layers \Vith PG 64-22, a layer with PG 76-28, and finally topped with
PG 76-28 SMA for a total HMA thickness of 14 inches. Section NIO used 4 irlches ofa
dolomitic limestone base material, tenned Missouri Type 5 base. Above Ihe base was
one HMA layer with PG 64-22 topped wilh l\Vo HMA layers wilh PG 70-22 binder for a
total ofapproximately 8 inches ofHMA. Sectioll SIl consisted 8 inches ofHMA: t\Vo
upper layers \Vilh modified PG 76-22 binder over two layers with unmodified PG 64-22.
These layers were placed atop the 6 inches of granite aggregate base materialutilized in
sections N3 through N7.
20..0083Hl=~~
SIl
4.0
6.0
.s 8.0
::i
~
~
1-
'5
al
12.0
10.0
14.0
:2 16.0
18.0
20.0
22.0
24.0 .ill..!..W'-!.IJ"-"_'_'_'-"'''--'-~''''''_'_'_W_'_'_!..W=..w..!_!_ "-''''-'-'!..J.
=
_
PG 67-22
PG76-28
= PG 76-22 == PG 76-22 (SMA) ~ PG 76-28 (SMA)
PG 64-22 _ PG 64-22 (2% Air Voids) """'" PG 70-22
Vl2'Z/'Ll.a Limerock Base EZ:3 Granite Base lII11lII Type 5 Base ~ Track Soil ~Seale Subgrade
Performance Monitol"ing
During each Test Track cycJe, the condition of each section is monitored to determine the
accumulation of distresses such as fatigue cracking and rutting over lime. l1lese
distresses conlribute lo an increase in roughness on the pavement surface, which
adversely affects ride quality. NCAT uses an Automatic Road Analyzer(ARAN) Ineltial
Profiler (shown in Figure 4.5) at the Test Track lo measure the small wavelengths in Ihe
longitudinal profile in the pavement surface at high speeds for each wheel path. It
achieves this by using an accelerometer, lasers, a speedometer, and a computer. The
accelerometer is used lo measure the acceleration ofthe vehicle, which is processed
through data algorithms lo establish an inertial reference. This reference is used lo
determine the instantaneous height ofthe accelerometer in the vehicle. High fi'equency
42
Pefers-Davis and Timm
lasers are used to determine the distance between the accelerometer and the ground, and
the speedometer is used to determine the distance between the measurements taken by the
laser (Sayers and Karamihas, 1998). These distance measurements are continually stored
in an on-board computer, and are later equatcd to a standard measurement known as the
I international roughness index (IR!).
I
I
METHODOLOGY
IRIData
The IRI data collected with the ARAN van over the course of each Test Track cycle were
used for the recalibration process. The IRI data (recorded in inches per mile) were
converted to uuits of present serviceability iudex (PSI), a term that is a direct input into
the AASHTO flexible pavement design equatiou. This conversion was made usillg the
following relationship (AI-Omari and Dalter, 1994):
Once the data were convelted, plots were created for each section thal showed the
decrease in PSI over time for the right (RPSI) and left wheel paths (LPSI), as well as the
average ofthe two (AvgPSI). From each ofthese plols, various poinls were selected for
calibration ofthe flexible pavement desigll equation. If considerable deterioration was
present over time, multiple points were used per section to provide a belter fit. Figure 4.6
shows a plot ofPSI over time for section NI durillg the 2003 Test Track cycle. As seell
in the figure, there is more deterioration in the rigbt wheel path than in the left wheel
path. This was not the case for all the sections; therefore the wheel path with the most
deterioration (Ihe lower PSI values) was used in the recalibration procedure lo be
conservative. Five points are shown in Figure 4.6 as ch'cles on the right wheel path PSI
line. These five points were chosen lo represent tbe trend of decreasing PSI over time. It
is imporlant to note Ihat Ihere is ajump in the PSI leve! around Jauuary 2005, which is
when Ihe section was milled and inlaid. When cases such as this occurred, poil1ts were
not selecled beyond Ihe jump in PSL
43
Pelel's-Davis and Tillllll
~~
--- RPSI
4 -+- AvgPSI
PI
, "'"
3.5
3 \ \~\ / ~ I
-\ V \
1/ ~ :~
~ 2.5
//~\ ..-.. r-1 "...'\11
...- ...
2
PI calibralion poinls V \. V
1.5
1
~ \"
0.5 ~.
o
28-Jun-03 14-Jan-04 01-Aug-04 17-Feb-05 05-Sep-05 24-Mar-06
Date
Figure 4. 6 PSI Data Derivcd fmm IRI Data fmm Seetion NI (2003 Test Traek).
While Ihe deterioration in Figure 4.6 is quite apparent, other seclions did not show
similar trends. For example, Figure 4.7 shows section N3 during both the 2003 and 2006
Tesl Track cycles (as menlioned previously, Ihis seetion was left in place after Ibe 2003
eycle). From the figure, it is seen that there is no major deterioration in this section over
the eourse of6 years and 20 million ESALs oftraffic; therefore, point seleetion in this
section was not possible. In sections such as this, an altificial terminal PSI was assigned
to the section to perform the recalibration. Each value assigned was lower than the actual
PSI ofthe section to be conservative. For this particular case, a value of3.5 was IIsed,
and the actual final PSI measurement was 4.3. This value was chosen to be conservative,
and because the regression procedure could not be perfonned unless there was a
sllbstantial difference between the initial and fmal PSI values. The other PSI versus time
graphs for each section can be found in Appendix A. If paint selection was possible, the
points were denoted on the figures in the Appendix as they are in Figure 4.6.
44
Pele/'s-Davis and Tillllll
4.5
4 'V
f r -
~.,
~ 2.5
2
1.5
0.5
o
28-Jun-03 09-Nov-04 24-Mar-06 06-Aug-07 18-Dec-08
Date
Figure 4. 7 PSI Data fl'ODl Section N3 (2003 and 2006 Test Track Cycles).
Predicted ESALs
The AASHTO Design Guide flexible pavement design equation was used to predict (he
amoullt of applied ESALs, alld is expressed as:
l [ M'SI ]
10gW18 =Z. S.+9.3610g(SN+l) - 0.20+ og 4.21~!: (4.2)
0.4 + ---'---,-ccc
(SN + 1)' .19
+ 2.3210gM R - 8.07
Where ZR is the normal deviate for a given reliabilty, So is the standard deviation,
JPSI is the expected loss of serviceability over the lifetime of the pavement, MR is the
resilient modulus of the subgrade, and SN is the structural number, as defil1ed by Ihe
followil1g eqllation:
Where al, a2, and a3 are the layer coefficients for the hot mix asphalt, base, and
subbase layers, respeetively; D is the thiekness of each respective layer in inches, and 111
is the drainage coefficient for the base and sllbbase layers. For this analysis, there were
no sllbbase layers for the sections at the Traek, so the final term was dropped from the
equation.
45
Pele/'s-Davis and Timm
For Equalion 4.2, Ihe reliability was set al 50%. This value was chosen because
higher reliabililies are used lo mtificially increase Ihe predicted Iraffic lo accounl fOl'
unceltainty in Ihe design process. Since all Ihe inpuls necessary for use in Ihe equation
were known, il was not necessary lo provide Ihis arlificial traffic increase, and therefore
the reliability was se! at 50%. This reliability reslllted in a normal deviate (ZR) ofzero;
Iherefore, the first lerm in the design eqllation was zero.
For the calculation ofthe structural number (SN), the thicknesses were oblained
fi'om construction record s for each section, the drainage coefficient for the base layer (m2)
was assumed to be equal to one, and the layer coefficient for the base layer (a2) was
assumed lo be 0.14. This layer coefficien! is a recommended value for a crushed stone
base course (AASHTO, 1993). The layer coefficient for the HMA (a,) was set a! a seed
value of0.44, which is the curren! value recommended in Ihe 1993 AASHTO Design
Guide amI also commonly used by ALDOT. The Ihicknesses IIsed for the analysis are
shown in Table 4.1. These Ihicknesses represent section-wide averages based on
sllrveyed depths.
To oblain the change in serviceability (L1PSl) for each test section, the
serviceability data was IIlilized. The initial PSI (Po) va lile for each section was set to the
first PSI reading taken after the section's construction. The final PSI value (P,) was
selecled based IIpon Ihe point seleclion procedure descl'ibed previously. The difference
between these two values was Ihe L1PSI used in the eqllalion.
The resilient modulus (MR) was calclllaled for each section based on
backcalclllated falling weigbt deflectometer (FWD) data collected over the COUl'se of each
Tesl Track cycle. An average was taken of Ihe backcalculaled data for each section and
then divided by three as recommended in the AASHTO Design Guide (1993). As
mentioned previously, Ihis value is divided by Ihree lo accollnt for differences in testing
procedures uscd to find the sllbgrade modul i. Table 4.2 shows the resulting M R values
used in Equation 4.2.
46
Pefe/'s-Davis and Timm
Afier identifying all the inputs, the predieted traffie (lOgW'8) was found for eaeh
test seetion. The estimated ESALs were found by solvingIO'OgW". For example, for
section NI ofthe 2003 Test Track, Table 4.3 shows the serviceability data that were used
in the predicted ESALs ealculation (also shoWII graphieally in Figure 4.6).
Using these data in conjunction with the resilient modulus and thickness data
mentioned previously, Ihe predicted ESALs based upon this damage were calculated and
are shown in Table 4.4.
Calculated ESALs
The AASHTO Design Guide quantifies pavement damage using Equivalent Axle Load
Factors (EALFs), which are used to find the number ofESALs. An EALF is used to
describe Ihe damage done by an axle per pass relative lo the damage done by a standard
axle (typically an 18-kip single axle) per pass. This equation comes from Ihe resulls of
the AASHO Road Test, and is expressed as follows according to Huang (2004):
47
Pelel's-Davis and TiIllIl1
Where 111 is Ihe nlllllber ofaxle load groups, EALF; is the EALF for Ihe th axle
load grollp, and n; is Ihe nmnber ofpasses ofthe Ih axle load grollp dllring Ihe design
periodo
These basic eqllations were used in conjunclion with the following equalions to
that were developed from the AASHO Road Test to characterize the traffic for the
reealibralion proeess (Hllang, 2004).
G =10
, g 4.2-1.5
[4.2 -P,] (4.6a)
Where G, is the logarithm of the ratio of loss of selvieeability at some time 1 to the
potentialloss of serviceability at terminal servieeability (P,) = 1.5, and the initial
servieeability is assllmed to be 4.2; this value was typical for flexible pavements at the
AASHO Road Test, and is used as the initial value for ALDOT pavement designs as
wel\. Px is a function of design and load variables, Lx is the axle group load in kips, ~ is
the axle code (1 for single, 2 for tandem and 3 for tridem), SN is the strlleturalnumber,
W" is the number of x axle load applications at time 1, W'I8 is the number of 18 kip axle
load applications at time 1, and PI 8 is the value ofP-, whell Lx is equal to 18 and ~ is equal
to one.
Most of the inputs needed for these equations were obtained fiom traffic data
eolleeted during the 2003 and 2006 Test Track eyc\es. In the 2003 eyele, six different
trueks were used to apply traffie to the test seetions. Five ofthe trueks (tenned "triple
trailers") were eomprised of one steer axle, a drive tandem axle, and five trailing single
axles as shown in Figure 4.8. The sixth truck (telmed the "box trailer") eonsisted of a
steer axle and two tandem axles as shown in Figure 4 .9. The weights ofthese axles were
48
Pelers-Davis and Tillllll
Table 4. 5 AxIe Weights Cor 2003 Test Trae\< (arter Priest and TimJII, 2006)
Sleer, lb Drive Tandem, lb Single Axle, lb
Truck ID 1 1 2 1 2 3 4 5
1Triple 10150 19200 18550 21650 20300 21850 20100 19966
2-Triple 11000 20950 20400 20950 21200 21000 20900 20900
3Triple 10550 20550 21050 21000 21150 21150 21350 20850
10500 21050 20700 21100 21050 21050 20900 21050
::::::=;~~~1~12~0~0:]~1~9~85~0~~2~0~7J50~~2~0~3~50~~2~0~140~0
19500 20300
21500
For the 2006 Test Traek eyele, only five h"iple h"ailer h"ueks were used to apply
traffie, and their axle weights are shown in Table 4.6"
49
PeteJ's-Davis and Timm
Using these ax1e weights, averages were compllted for each axle type (steer,
single, and tandem), for each truck type (triple and box), and for each Test Track cyc1e.
For the tandem axles, averages were compllted for each tire set (1 and 2), and then
sllmmed to obtain the total average tandelll axle weight. For example, the 2003 weights
resulted in the averages shown in Table 4.7. The cOlllbined values are the reslllt ofthe
average of steer axles, the Slllll ofthe average tandem axle weights, and the average ofthe
single axle average weights. Table 4.8 shows the average axle weights for the 2006 Test
Track cycle.
These averages were used in Equatioll 4.6 (as the Lx terlll) to compute the P., and
W,./W,18 telms, which were then used to compute the amount ofESALs applied per ax1e.
The values chosen forp,alld the calclllations for SN were discllssed in the previolls
section. To provide an example ofthe calclllations performed for the calculated ESALs,
the values shown in Table 4.9 were computed for section NI ofthe 2003 Test Track
llsing the average axle weights shown in Table 4.7.
50
Peters-Davis anel Timm
Detailed data were collected for the amollnt oftraffic applied at the test track. For
each day that trucks were driving dllring the test cycle, the total amount of steer, single,
and tandem axle passes were recorded for both the triple and box trailers, and a sample of
those data are showlI in Figure 4.10. To calclllate the ESALs applied by each axle, these
axle passes were mllltiplied by their respective ESALs/Axle factors shown in Table 4.9 lo
obtain the total ESALs per day for each axle type. The ESALs were sllmmed across all
axle types to obtain the total amount ofESALs applied per day, and these were summed
cllmulatively to get the total ESALs applied up to a ceJ1ain day in the test cycle, as shown
the "ClImulative SlIm" column in Figure 4.10.
~
568 299 12327 3121 562340
=:g
11
rXj].
ITriple
156
59
I
296
564
592
299
11:
12300
4658
311
117
578053
584004
IU IBox
ITrple 25 129
6 O
5'
58401
58651
a
25S
'3r. IIBox 7261 7' 587499
3r. ITlinle 15' ~ 15131 :I;
r
r
IIBox
lTriple
IIBox
78i
e
31:
6531
fI
r
r
r
lTriple
:ox
124~
=11
~
~
65:
r
r
r ~~
'004 ",,- 11!:
~ ~
766
1549
JI;? 67
67:
1121
r"pre 157:
39E
FIgure 4.10 Sample ofDelatlcd Axle Dala.
7jl621
01
1:
t 1230.
O 8661
Using these axle data, Ihe tota l amollnt of applied ESALs could easily be obtained
for various dates. Using section NI from the 2003 Test Track as an example, the
calculated ESALs showlI in Table 4.1 O were recorded for the same dates that the ESALs
were predicted for in Table 4.4. It is imp0l1ant to note thal the calculated ESALs shown
in Table 4.10 are the result of not only the number ofaxle passes applied up to thal date
in the cycle, but also fmm a unique set ofESAL/axle factors for each date. This is
because each date represents a different pr vallle and, consequently, a different set of
ESAL factors for each axle type.
51
Pete/'s-Davis and Tillllll
Regl'essioll
When comparing the predicted and calclllated ESALs for section NI ofthe 2003 Test
Track, it is apparent that there are some sllbstantial differences when using 0.44 for the
lIMA layer coefficient. Table 4.11 shows the calculated and predicted ESALs for section
NI ofthe 2003 Test Track, as well as the difference between the two. In general, the
current Sh'llctural coefficient (0.44) reslllts in gross llnderpredictions ofthe ESAL-
capacity ofthe pavement sbllcrure. This is likely due to the newer and more advanced
lIMA materials that are used at the track, which implies that the layer coefficient shollld
be higher. The large error percentages show a need to bring these vallles closer together
for a more realistic ESAL prediction, and a least sqllares regression was perfonned to
accomplish that task.
Table 4. 11 ESAL Differences Assu = 0.44 for Section NI (2003 Test Tmck)
,
1.126,574 2,837,091 1,710,517 -60%
1,270.712 2,963,064 1,692,352 -57%
1.638,661 3,212,141 1,573,480 -49%
2,340,290 4,321,771 1,981,481 -46%
To perform Ihe regression, firsl the differences between calclllaled and predicled
were sqllared. These vallles were sllmmed to oblain the error Sllm of squares (SSE),
which is defined by Ihe following eqllation:
The mean was oblained for the calclllaled ESALs, and the difference between that
mean and each predicted ESAL level was laken and then sqllared. TIlese vallles were
sllmmed lo oblain Ihe total Sllm of sqllares (SS1), as seen in the following equation:
(4.8)
To perform the regression, Microsoft Excel Solver was utilized. The solver was
set to minimize the SSE telm while only changing the I-IMA layer coefficient (al) . This
process is inherently iterative in narure: every time the layer coefficient changes (i.e.
from 0.44 to a new regressed value), both Ihe calculated and predicted ESALs change.
This is because both oflhese values are calculaled USillg the strllcturalnllmber (SN), and
that is calculated using Ihe layer coefficient (al)'
52
Pelers-D((Vis and Timm
For sections that did not exhibit considerable damage at the end ofthe Test Track
cycle, an aJtificial terminal serviceability was selected for use in the regression
procedure. These values were always equal to or lower than the actual serviceability
level ofthe test section, and therefore provide conservative estimates ofthe layer
coefficient. For example, in section N3 ofthe 2003 and 2006 Test Track cycles (refer to
Figure 4.7), the final serviceability value measured was 4.27. The initial reading was
4.36, resulting in a L1PSI value of only 0.09. In cases such as this, the solver function was
used to create a difference in calculated and predicted ESALs of zero by only changing
the layer coefficient. For each section where artificial terminal serviceabilities were used,
the first PI used was the actual terminallevel fOllnd in the section. Ifthe solver could not
converge on a solution for the layer coefficient, then the P, was decreased by one tenth
and then the process was repeated until the solver converged on an answer. In the case of
N3, an artificial lerminallevel of3.9 was assiglled to creale a large enough L1PSlfor
convergence lo a solution for the layer coefficient, which resulted in a layer coefficient of
0.62. Figure 4.11 shows the calculated versus predicted ESALs for each test section afier
the regression was performed. The data in Figure 4.11 illustrate section-specific
regression results.
53
I
Pelers-Davis and TiJIIJII
I
I 2.0E+07
/
x N22006
j x N3 2003-2006
..:
Ih
w N4 2003-2006
"C 1.0E+07 + N5 2006
./
$
o N6 2003-2006
:;
~ x N7 2003-2006
"- + N82003
5.0E+06 N82006
I~
.+ N9 2006
1> x N10 2006
X .. ~ x S11 2006
O.OE+OO
O.OE+OO 5.0E+06 1.0E+07 1.5E+07 2.0E+07
Calculated ESALs
Table 4.13 pl'Ovides a sllmmary ofthe stmetmal eoefficients fOllud for each
2 2
seetion, as well as the R vallles. Ifthere is no assoeiated R , then those seetions were
assigned attifieial terminal serviceability levels as diseussed previously.
Although it is apparen! thal the models do not describe 100% ofthe data used in
this study, the coefficienl of detennination (R 2) values are stillreasonably high
considering the highly variable natme of research involving pavements and live traffic.
Even the AASHO Road Test results had R 2 values that were no! optimum; Ihe overall
average R 2 was appl'Oximately 0.85. Additionally, for those analyses, many more test
seclions (284) were used lo arrive al a layer coefficient.
54
Pele/'s-Davis and TilllllJ
Figure 4.12 shows the individual sectionlayer coefficients and the average layer
coefficient graphically. As seen in Table 4.13 and Figure 4.12, most ofthe valnes found
were higher than the current recommended valne of 0.44. lt is impOltant to note that
although the sections assigned artificial terminal serviceability levels do provide
conservative estimates ofthe HMA layer coefficient, the pavements did /lot actually
reach that level of serviceability in most cases. There were five sections that were
assigned a terminallevel ofserviceability, and ifthese sections are not included in the
analysis, the resulting average HMA layer coefficient is 0.51; slightly lower than the
overall average value shown in Table 4.13.
55
Pelers-Davis and Timm
0.80
0.68
0.70
0.63 0.62
0.59 0.58 0.59 0.58
-r::
CD
' 0.50
0.60
0.50
r- ~
-
-
- - 0.48 ~
0.43
0.48
0.44 0.41
-0:54
-
!i:
--r-
CD
O
o 0.40 r- - - r- r- - - r-- r- r- e-- -
~
~ 0.30 ~ - - - r- r- r- - r- r- r- r- ~
-
j
0.20 - - -
r- r- r- ~ - r- r- r- r- - -
'" '"
.... '" '".....
ID
'"
Z Z Z Z
There were two seelions thal resulted in coefficients lower than 0.44: N8 ofthe
2003 Track, and NlO ofthe 2006 Track. Afier the end ofthe 2003 Test Track cycle,
section N8 showed considerable signs of fatigue cracking damage which was not
expected given the pavement cross section and material s used. A forensic investigation
of the section indicated that debonding had occurred between the HMA lifts, making Ihe
pavement mllch more prone lo damage. A full report of Ihe investigation of debonding in
seclion N8 has been documenled elsewhere (Willis and Timm, 2006). Section NlO ofthe
2006 Test Track also displayed more damage than expeeted at the end of the lest cycle.
Considerable sllrface distOltion was present throughollt the sectioll, which warranted
another forensie invesligation. Trenches were cut in the section to view Ihe damage
throllghout the cross section, and the individual HMA lifts were easily delaminated fi'om
one another lIsing a backhoe. This phenomenon \Vas not observed in any other sections
that had trenches cut on the same day (January 29, 2009). While no reports have been
offieially published on this topie yet, the following piclures (Figures 4.13 and 4. 14)
illustrate the likelihood Ihal debonding was oceurring. Based on the published repOlt for
seetion N8 and the forensie photos ofsection N 10, it is possible tllat the HMA layer
eoefficients fOllnd for these sections were lower than the recommended value of 0.44 due
to their probable debonding issues.
56
Pefe/'s-Davis and Timm
I _
Figure 4. 13 Trench Showing Sigus ofDeho:ndiine in Section NI0.
57
Pelers-D([vis ([nd Tillllll
After the regression was perfonned, the average value for the layer coefficienl.
(0.54) was used to create a graph of calculated versus predicted ESALs with all test
sections, and is shown in Figure 4.15. As seen in the figure, the data are fairly evenly
distributed around the line of equality. A similar graph was also created using the
recoIIIlllended value of0.44, and is shown in Figure 4.16. From Figure 4.16 it is apparen!
that using the recomlllended layer coefficient results in an under-prediction of calculated
ESALs using the flexible paveIllent design equation for 89% ofthe sections at the Test
Track, which would result in overly conservative designs.
2.0E+07 ,-- - - - - - - - - - -- - - - - - -- 71
/r
~ N12003
N1 2006
1.5E+07 .I- -- - - - - -- - - - - - -cr'-- - - - I ' N2 2003
~
x N2 2006
x N3 2003-2006
x + / .
N4 2003-2006
13 ~ + N5 2006
~ 1.0E+07 ~_-----/
-"-----,L----,;:--.---='-------I . N6 2003-2006
~ +0. ~ N7 2003-2006
_ x . N8 2003
N8 2006
5.0E+06 -1--- - - . - -7"-- -- - - - -- - - - -- -1 . N9 2006
o:.
/f
x N10 2006
x S11 2006
o
i D x
O.OE+OO *"----,....'--- - --,------,--- - ---1
O.OE+OO 5.0E+06 1.0E+07 1.5E+07 2.0E+07
Calculated ESALs
Figure 4. 15 Calculated VS. Predicted ESALs Using al = 0.54.
58
Pefers-Davis and Tillllll
'~N12003
1.5E+07 -\--- /r
- - - - - - - -- - - --/-- - ----1. N2 2003
N1 2006
x N22006
j
c(
rn
w
~ 1.0E+07
:;
~
/ N3 2003-2006
N4 2003-2006
+ N5 2006
N6 2003-2006
~ N7 2003-2006
Q. N8 2003
N8 2006
x N10 2006
x S112006
I
O.OE+OO 5.0E+06 1.0E+07 1.5E+07 2.0E+07
Calculated ESALs
Figure 4. 16 Calculated VS. Predicted ESALs Usiug al = 0.44.
Trends
No trends were apparent in the regression analysis. Pavement cross section, HMA
thickness, and binder type all had no obvious effect on the resulting layer coefficient. For
example, sections NI and N2 from the 2003 Test Track both had very similar cross
sections of 5 inches HMA over a 6 inch granular base. The resulting layer coefficients
were 0.50 and 0.56, respectively. In Ihe 2006 Test Track, the same sections were rebuilt
with similar cross sections of7 inches HMA over a lO inch granular base, and the layer
coefficients were 0.59 aud 0.63, respectively. Other sections with similar cross sections
and HMA layer lhicknesses were compared, and uo trend was found. Sections N3 and
N4 ofthe 2006 Test Track both consisted of 9 inches of HMA, N3 being unmodified PG
67-22, and N4 being modified PG 76-22. The resulting layer coefficients were opposite
ofwhat was expected: 0.62 and 0.58, respectively. The highest layer coefficient was
found for section SIl, which consisted 8 inches ofHMA: two upper layers ofmodified
PG 76-22 binder over two layers of umnodified PG 64-22. There are several sections
wilh Ihicker cross sections than SIl, as well as higher PG grades. From these data and
other similar comparisons, it was concluded that there were no trends found relative to
overall cross section, HMA thickness, or binder type. This could be due to other factors
that can affect the layer coefficient within the pavement stmcture. Debonding between
HMA lifts Iikely caused lower layer coefficient values in sections N8 and NIO as
discussed previollsly. Other factors such as compaction, binder content, air voids or
pave1l1ent age might also have significant (and perhaps confounding) i1l1pacts 011 the
differences in layer coefficients, which 1l1ay be why no trends are apparent in the data.
Conversely, it could be that a, is generally insensitive to these factors . This is evidenced
59
Pefers-Davis and Tillllll
by many states, ALDOT included, by using a single al value in sh'uctUl'al design for a
wide variety of asphalt mixhlres.
Minimum Thiclmess
It is important to note that the regressed layer coefficient ofO.54 was calibrated to
sections witl! HMA thick.nesses no less than 5 inches. Therefore, this coefficient should
not be Ilsed for designs that result in pavements will! a HMA Ihickness ofless Ihan 5
inches. It is recommended thal for designs resuIting in thiclOlesses of less than 5 inches,
the AASHTO recommended coefficienl of0.44 be used, or the minimum thickness be set
lo 5 inches.
14
12
10
g
-"
J:
Q
Q,
8
oC( 6
::;; -+- al = 0.44
:I:
=
___ al 0.54
4
O
1,000,000 10,000,000 100,000,000 1,000,000,000
ESAls
60
Pelel's-Davis and Tillllll
SUMMARY
Pavement performance and traffic data were collected over the course ofthe 2003 and
2006 Test Track cycles. These data were used to perform a recalibration ofthe HMA
layer coefficient (al) that is used in the current ALDOT flexible pavement design
methodology. The recalibration was achieved by calculating the predicted ESALs using
the 1993 AASHTO Desigll Guide flexible pavement design equation, and also calculating
the ESALs applied at the Test Track during the two test cycles. The pavemellt
performance data (IRlmeasurements) were used to find a challge in serviceability (iJPSI)
01' a terminal serviceability level (PI) for use in the design equations. Some sections did
not exhibit an increase in roughness over the course ofthe Test Track cycle(s), and
therefore were assiglled artificial terminal serviceability levels to provide a iJPSI term
large enough for convergence to a layer coefficien!. The values chosen were always less
than 01' equal to the actual final serviceability leve! at the end ofthe cycle, and therefore
are conservalive. The predicted and calculated ESALs were compared, and a least
squares regression \Vas performed lO minimize the difference belweell the t\Vo values
while ol1ly changing the HMA layer coefficient. This resulted in an average layer
coefficient of0.54 \Vith a standard deviation ofO.08 for all the test sections analyzed. fr
the sections assigned a tel1l1inal serviceability leve! were not included in the analysis, the
average layer coefficient was 0.51. These values can be compared to the current value
used for the HMA layer coefficient of0.44, which comes fi'om the AASHO Road Tes!.
There were no trends in the data regarding the pavement cross section, HMA thickness,
01' binder type. The impacl of changing the layer coefficient to the average value fouud
in this analysis would result in a reduced HMA thickness, regardless of designo The
amount of redllced thickness is dependent upon other design inputs such as resilient
modulus, traffic, etc.
61
Pelel's-Davis and Tilllll/
SUMMARY
This investigation was performed to determine the relative sensitivity ofHMA thickness
to the inputs ofthe AASHTO Design Guide flexible pavement design equation, and to
recalibrate the HMA layer coefficient used for flexible pavement designs. For the
recalibration procedure, Iraffic alld performance data from the 2003 ane! 2006 NCAT
Test Track cycles were used in conjunction with flexible pavement design and traffic
equations developed fiom the AASHO Road Test. Calculated and predicted ESALs
were computed and compared, and a least squares regression was performed to
determine new layer coefficients for each test section.
CONCLUSIONS
From !he Iiterature review, it was apparent that research \Vas needed to determine the
relative sensitivity ofthe inputs to the AASHTO flexible pavement design equation
since no Iiterature conld be fonnd on this topic. It was also concluded that based upon
the Iimited parameters ofthe AASHO Road Test, the recommended default HMA layer
coefficient of0.44 shollld be reanalyzed to ensllre accllracy for CIllTent Illaterials. While
several studies have been conducted to determine the layer coefficients ofnew materials,
there is no recommended hied-and-hlIe procedure that all researchers can agree upon.
In addition, the results found fiom many layer coefficient stlldies tended to be highly
variable, and a specific layer coefficient was typically not recolllmended for a palticlllar
material.
The input sensitivity analysis showed that the layer coefficient, resilient
modlllus, and traffic are by far the most inflllential parameters on the resulting HMA
thickness. Since the resilient modulus and traffic are typically given parameters for a
particular pavement design, it was concluded that an accurate characterization ofthe
layer coefficient is extremely importan!. Input dependencies were found in the
sensitivity analysis; all inputs had an increasing influence on the resulting HMA
thickness as the traffic level increased.
The recalibration procedure resulted in an average HMA layer coefficient ofO.54
with a standard deviation ofO.08 from the 14 pavement sections stlldied. Five sections
that did not exhibit considerable deterioration were assigned artificially low, yet
conservative, terminal serviceability levels to obtain a LJPSI term large enough for
convergence to a layer coefficient. Ifthese sections are not included in the calculations,
the result is an average layer coefficient ofO.51. The only two test sections that resulted
in regressed layer coefficients lower than Ihe AASHTO recol11lllended value of0.44 had
probable slippage failures, as found in forensic investigations. No hends were observed
in the resulting layer coefficients when cOl11paring binder type, HMA layer thickness, or
overall pavement cross section. The impact of changing the layer coefficient to the
average value found in Ihis analysis would result in a reduced design HMA Ihickness,
regardless ofthe values of other inputs to the designo
62
Pelers-Davis and Timm
RECOMMENDATIONS
Based upon Ihe results ofthis investigalion, it is recommended Ihat ALDOT use a layer
coefficient ofO.54 for flexible pavement designs using the AASHTO design
methodology. This coefficient is larger Ihan the recommended value of0.44; however,
it was expected to increase due lo Ihe numerous advancements in HMA materials and
conslruction since the AASHO Road Test was completed in 1961. Using a layer
coefficienl of 0.54 would result in a lIMA Ihickness savings of approximalely 18%.
This coefficient shou ld not be used for desigus that result in pavements wilh a HMA
thickness of less than 5 inches. It is recommended that for designs resultiug in
thicknesses of less than 5 inches, the AASHTO recommended coefficient of 0.44 be
used, or the minimum thickness be set to 5 inches. Care should be taken when applyiug
the new regressed coefficient to otller states. The layer coefficient of0.54 is the result
ofthe environmental conditions and materials used in this sludy.
63
Pelers-Davis and Tillllll
REFERENCES
ALDOT. ALDOT Procedllre 390: Procedure for Conducling Soil Surveys and Preparing
Malerials Reporls. ALDOT Bureau ofMatcrials and Tests, 2004.
AI-Omari, B. and M.l. Darter, Relalionships beflveen lnlernalional ROlIglmess lndex and
Presenl Serviceabilily Raling. Transportation Research Record 1435, TranspOltation
Research Board, Washington, D.C. 1994.
Coree, BJ. and T.D. White, The Synlhesis ofMixlure Slrenglh Paralllelers Applied lO Ihe
Delerminafion ofAASHTO Layer Coefficient Dish"ibulions. Asphalt Paving Technology,
vol. 58. 1989.
Hossain, M., A. Habib and T.M. LaTorella. Slruclural Layer Coefficienls ofCmmb-
Rllbber Modified Asphalt Concrele Mixtures. Transportation Research Record 1583,
Transportation Research Board, Washington, D.C. 1997.
HRB. The AASHO RoadTesl. Special RepOlts 61A, 61C, 61E. Highway Research Board,
1962.
Huang, Y.H., Pavement Analysis and Design. 2 nd ed. New Jersey: Prentice Hall, 2004.
Jess, J.C. and D.H. Timm, Struclural Coefficienlsfor Nell' Asphall Mixtures, ALDOT
Project No. 930-559, National Center for Asphalt TeclUlology, Auburn University, 2005 .
64
Pelel's-Davis and Timlll
Priest, A.L. ami D.H. Timm, Melhodology and Calibralion ojFaligue Transjel' Funclions
jol' Mechanistic-Empil'ical Flexible Pavemenl Design, Report No. 06-03, National Center
for Asphalt Technology, AlIbllrn University, 2006.
Sayers, M. W alld S.M. Karamihas, The Lil/le Book ofProfilijng: Basic lnformation abolll
Measuring and lntelpl'eting Road Pl'ofiles. University ofMichigall, 1998.
Smith, K.D., K.A. Zimmerman alld F.N. Fm, The AASHO Road Test: The Living Legacy
jol' Highway Pavements. TR News No. 232, TranspOltation Research Board, 2004.
Timm, D.H., A.L. Priest and T.V. McEwen, Design and Inslrumenlation ojlhe Slruc/ural
Pavemenl Expel'imenl al/he NCATTesl Tl'ack, ReportNo. 04-01, National Center for
Asphalt Technology, Aubllrn Un iversity, 2004.
Timm, D.H., Design, Construclion, and lnsh'ulllentalion of Ihe 2006 Tesl Tl'ack
Slruclul'al SllIdy, Report No. 09-01, National Center for Asphalt Technology, Auburn
Ulliversity, 2009.
Van Wyk, A., EJ. Yoder and L.E. Wood, Delel'minalion ofSlruclllral Equivalency
Faclol's ofRecycled Layel's by Using Field Data. TranspOltation Research Record 898,
TranspOltatioll Research Board, Washington, D.C., 1983.
Von Quintus, H.L., Evalllalion ojPl'ocedul'e lo Assign Slrucll/ral Layel' Coefficienls fol'
Use in Flexible Pavelllenl Design. Kansas Depaltmellt ofTransportation RepOlt No. KS-
07-9.2007.
Willis, J .R. and D.H. Tinun, FOl'ensic Inves/iga/ion oj a Rich Bollolll Pavemen/, Report
No. 06-04, National Center for Asphalt Technology, Auburn University, 2006.
65
Pete/'s-Davis and Tillllll
APPENDIXA
66
Pefe/'s-Davis and Tim/ll
4.5
~
"- ~
--
3.5 -"-
3 ~ lf"I ~ ~
..,.,
rYV"" r;Y ~
2.5 -+- LPSI -
i ...... RPSI
o..
2 -+- AvgPSI -
PI
1.5
0.5
o
10-0cl-06 18-Jan-Ol 28-Apr-Ol 06-Aug-07 14-Nov-07 22-Feb-08 01-Jun-OB 09-Sep-08
Date
Figure Al: PSI Data from Seetion Nl ofthe 2006 Test Traek.
4.5
-+- LPSI
4
..-.... ...... RPSI
3.5
"'l
;"''' '->'<\' /
.
-..- AvgPSI
PI ~
3
"'-.
2.5
\
\"-\ ~ ~
i
o.. \ 1 m JI
~
'lo!
2
1.5 \..~
'"
0.5
o
28-Jun-03 14-Jan-04 01-Aug-04 11-Feb-05 05-Sep-05 24-Mar-06
Date
FigUl-e A2: PSI Data from Seetion N2 of the 2003 Test Trael(.
67
Pelers-Davis and Til/llll
2.5
........ LPSI
; ...... RPSI
D.. 2 r---
....... AvgPSI
1.5
0.5
o
10-Gel-06 lB-Jan-07 2B-Apr-07 06-Aug-07 14-Nov-07 22-Feb-OB 01-Jun-OB 09-Sep-OB
Date
Figure A3: PSI Data from Seetion N2 of the 2006 Test Traek.t
4.5
....
~~
..JW.
4
3.5
-- ........ LPSI
3 ...... RPSI
......-AvgPSI
; 2.5
D..
1.5
0.5 -
O
20-Mar-03 01-Aug.04 14-Dee-05 2B-Apr-07 09-Sep-OB
Date
Figure A4: PSI Data from Seetion N4 oftbe 2003 and 2006 Test Trael, Cycles. t
68
Pele/'s-Davis and Tillllll
4.5
4
.. _. .... . -+- LPSI
...... RPSI
....... AvgPSI
W
3.5
" PI
3
2.5
D..
~ Lt
1.5
0.5
o
20-Mar-03 01-Aug-04 14-Dec-05 2B-Apr-07 09-SepOB
Date
Figure AS: PSI Data from Section NS oftbe 2006 Test Tmck.
4 .5
4
J ..... .... ... T"
~
~
3.5
~~
3 "-
-+- LPSI
~ 2.5 RPSI
....... AvgPSI
2
PI
1.5
0.5
o
20-Mar-03 01 -Aug-04 14-Dec-05 2B-Apr-07 09-SepOB
Dale
Figure A6: PSI Data from Section N6 oftbe 2003 and 2006 Test Tl'ack CycIes.
69
Pelers-Davis and Til/ll/l
4.5
4
~
3.5
""~
~ ~!
3
2.5
,t
o..
-+- LPSI
2 -+- RPSI
-+- AvgPSI
1.5
PI
1
0.5
o
01-Sep-02 14-Jan-04 28-May-05 10-0cl-06 22-Feb-08 06-Jul-09
Dale
Figure A7: PSI Data from Section N7 of the 2003 and 2006 Test Tracl, Cycles.
4.5
\ 1.~.,
~
3.5
~---.
3
~r.".
2.5
o..
2
.... ~
....f'-..
I-+- LPSI
1.5 I-+- RPSI
1 I ........ AvgPSI
PI
0.5
o
28-Jun-03 14-Jan-04 01-Aug-04 17-Feb-05 05-Sep-05 24-Mar-06
Date
Figure A8: PSI Data fl'ODl Section N8 ofthe 2003 Test Traek
70
Pelers-Davis and Tilllll/
4.5 ,----------------------------------------------,
4 . ~~~~=t~~----------------~
O+------r-----.,-----,------.------.-----.-----~
Figure AH: PSI Data from Section NIO of tbe 2006 Test Tracl{.
4.5
3.5
.~ ---. ~~
3 ~ LPSI f--
-- RPSI
2.5 r--
; ~ AvgPSI
D..
2
1.5
0.5
O
10-0el-06 lB-Jan-07 2B-Apr-07 06-Aug-07 14-Nov-07 22-Fe b-OB 01 -Jun-OB 09-Sep-OB
Date
Figure A1 2: PSI Data from Section Sl1 of tbe 2006 Test Tmc!{. t
t Point selectionnot possible. No visible decrease in serviceability.
72
Pele/'s-Davis and Timm
~tb.
3.5
3
~ - -<00
....
2.5 ~
-+-- LPSI
i 2 r-
o.. ----- RPSI
--+-- AvgPSI
1.5 PI r-
1
0.5
O
10-0cl-06 18-Jan-07 28-Apr-07 06-Aug-07 14-Nov-07 22-Feb-08 01-Jun-08 09-Sep-08
Date
Figure A9: PSI Data from SectioIl N8 of the 2006 Test Track.
3.5
~ p ::-..,.....,... . ~
3
-..-
- ..--
2.5 .... LPSI
----- RPSI
~ 2
---'>--- AvgPSI
-
1.5
0.5
O
10-0cl-06 18-Jan-07 28-Apr-07 06-Aug-07 14-Nov-07 22-Feb-08 01-Jun-08 09-Sep-08
Date
Figure AtO: PSI Data fmm SectioIl N9 ofthe 2006 Test Track. t
71
Peters-Davis and Timm
APPENDIXB
73
Pelers-Davis and Ti/ll/ll
Table B4 Re "essioll Statistics for Seetion N6 of the 2003 and 2006 Test Tracl<s
Predicted ESALs Calculated ESALs Difference % Error Diff STOi
5,880,826 176,433 5704393 97% 3.25E+13 9.72E+13
14,187,307 13,768,047 419260 3% 1.76E+11 1.39E+13
Average 10,034,067 $lIm 3.27E+13 1.1 1E+14
R' 0.706
Table BS Re "ession Statistics for Section N7 of the 2003 and 2006 Test Tracl<s
Predicted ESALs Calculated ESALs Difference % Error Di STDi
6,004,943 1,044,458 4960485 83% 2.46E+13 7.08E+13
9,317,670 5,160,057 4157613 45% 1.73E+13 1.85E+13
13,051,495 14,008,131 956635 7% 9.15E+11 2.07E+13
Average 9,458,036 $lIm 4.28E+13 1.1 0E+14
R' 0.611
74
Peters-Davis and Tillllll
Table B8 Regression Statistics fol' Section N10 of the 2006 Test Track
Predicted ESALs Calcu lated ESALs Oifference % Error Oiff' STOiff'
1,853,463 1,261,542 591921 32% 3.50E+11 6.77E+12
3,553,293 3,793,560 240266 7% 5.77E+10 4.75E+09
6,180,789 6,099,393 81396 1% 6.63E+09 5.00E+12
Average 3,862,515 SlIm 4.15E+11 1.18E+13
R' 0.965
75