AASHTO EMPIRICAL DESIGN

METHODOLOGY
 Design Methodologies

 Flexible pavements
 Empirical method without or with soil strength test

 Limiting shear failure method

 Limiting deflection method

 Regression methods based on pavement

performance or road Tests

 Mechanistic-Empirical method

 Rigid pavements
 Empirical solution

 Analytical solution

 Numerical solution
 Empirical Methods (Flexible Pavement)

 Without strength test

 Soil classification system (Terzaghi 1929)

 Uniform conditions (A1 to A8)

 Non-uniform conditions (B1 to B3)
 Highways research board, US (1945)
 Soil classification (A1 to A7)
 Group index to differentiate soils into groups
 With strength test
 CBR - California highway department (1950)

 Disadvantages
 Applied to only specific conditions, trial and error for

new conditions
 Limiting Shear Method

 Limiting shear failure method

 Based upon that no shear failure occurs.
 Major properties: cohesion and angle of internal
friction
 Terzaghi’s bearing capacity formula (1943)
 Not popular since the pavement should be designed
for riding comfort rather than barely preventing shear
failures
 Limiting Deflection Method

 Limiting Deflection Methods

 Vertical deflection will not exceed the allowable limit
 Deflection measured by layer elastic methodologies
 Subgrade deflection < 0.1 inch (Boussinesq: Kansas
Highway Dept)
 Surface deflection < 0.25 (Burmister: US Navy- 1953)
 Advantage: Surface deflection can be measured in
field
 Disadvantage: Pavement failure caused by
excessive stresses and stains instead of deflections
 Regression/Empirical Method

 Regression method based on pavement

performance
 Example: AASHTO method
 Method can be applied to specific conditions
 Mechanistic Empirical Method

 Mechanistic-Empirical method
 Based upon mechanics of materials
 Input: wheel load, material type, climatic conditions
 Output: pavement response such as stress and
strain
 Response used to predict pavement performance
(dependence on field performance is necessary)
 Typical response
 Vertical compressive strain above 
the subgrade
t
 Tensile strain under the
Hot Mix Asphalt (HMA) c
 Elastic layer programs (Flexible)

 CHEV: Chevron Research Company (1963)

 Modified CHEV (DAMA) Non-linear elastic granular
material (1979)
 BISAR: Shell Oil Company for vertical and
horizontal loads
 ELSYM5 : California, Berkeley - 5 layer system with
multiple wheel loading (1986)
 PDMAP: Probabilistic Distress Model for Asphalt
Pavements (1986): Predicts rutting and cracking
 Rigid Pavement Design

 Major design factor: flexural stress in concrete

 Empirical solution: AASHTO (1993) design
methodology
 Analytical solutions (slab-subgrade full contact)
 Westergaard’s analysis based on liquid/winkler

foundations
 Stresses and deflections in concrete pavement
 Temperature curling; loading (corner, edge, interior)
 Pickett formulation (corner)
 Modified Westergaard solution (part of slab not in contact
and 20 percent load transfer)
 Used in PCA (1951)
 Pickett’s analysis based on solid foundation
 Rigid Pavement Design (Cont’d)

 Numerical solutions (accounts for pumping,

temperature curling, and moisture warping – when
slab and subgrade are not in contact)
 Consideration of both liquid and elastic foundation

 Applicable to jointed slabs.

 Computer Solutions

 WESLIQID (Chou 1981)

 WESLAYER (Chou 1981)
 ILLI-SLAB (University of Illinois, 1979)
 AASHTO 2002 design methodology (ERES and ASU 2006)
 Applicable approaches
 Discrete element method
 Finite element method
AASHTO Design Approach
AASHO Road Test (late 1950’s)
 History of AASHTO Guide

 Empirical design methodology based on AASHO

Road Test in the late 1950’s
 Several versions:
 1961 (Interim Guide), 1972, 1986

 1986 version included refined material characterization

 1993
 More on rehabilitation
 Consistency between flexible, rigid designs
 Current version for flexible design procedures
 1998
 Supplemental Guide for rigid pavement
design (current version)
One Subgrade Type (A-6/A-7-6, Poor
Drainage)
Controlled Construction Methods

Rigid Pavement

Flexible
Pavement
 1950s’ Vehicle Loads

(And on a specific stretch of

roadway!)
Two Years of Testing...

(not 10 to 20 years of
service)
 Limited Set of Materials

 One asphalt concrete

 ¾-inch surface course

 1-inch binder course

 One portland cement concrete (3500 psi @ 14 days)

 Four base materials
 Well-graded crushed limestone (main experiment)

 Well-graded uncrushed gravel (special studies)

 Bituminous-treated base (special studies)

 Cement-treated base (special studies)

 One uniform sand/gravel subbase

 Current AASHTO Design Inputs and
Outputs
 Inputs
 Performance criterion (PSI)

 Traffic (W18) and design life (analysis period)

 Foundation (Resilient Modulus)

 Material of Construction

 Environmental

 Drainage coefficient

 Reliability (ZR, So)

 Outputs
 Required pavement capacity (SN) for AC
 PSR and PSI

 Serviceability rating scales (0 to 5)

 PSR: Present serviceability rating (subjective)
 PSI: Present serviceability index (objective)
PSI  5.03  1.9 log(1  SV )  0.01 C  P  1.38 RD 2
PSI  5.41  1.80 log(1  SV )  0.09 C  P
where
 SV is slope variance
 C is linear feet of cracking per 1000 ft2 area
 P is bituminous patching in ft2 per 1000 ft2 area
 RD is rut depth in inches (average of both wheel tracks) measured
with 4 foot straightedge
 Slope Variance, SV

 Roughness is defined by the longitudinal

profile.
 Slope Variance (SV) is a quantitative measure
of roughness.
 Slope measured at 1.0 foot interval in both
wheel paths and SV  
Y 2 is Y /n
computed individually.
2

SV 
n 1

 Examples
 Slope profilometer (AASHO Road Test)
 Roughness meter
 The Road meter
Performance Criterion (DPSI)
 Performance Criterion (DPSI), Cont’d

Decrease in Serviceability
 Current AASHTO Design Inputs and
Outputs
 Inputs
 Performance criterion (PSI)

 Traffic (W18) and design life (analysis period)

 Foundation (Resilient Modulus)

 Material of Construction

 Environmental

 Drainage coefficient

 Reliability (ZR, So)

 Outputs
 Required pavement capacity (SN) for AC
 Materials of construction (Subbase), a3

Structural number Use CBR, R-value, or Mr

of the subbase, a3 to find a3 values
 Materials of construction (base), a2

Structural Use CBR, R-value, or

number of the Mr to find a2 values
base course, a2
Equation for Layer Coefficient, a2 and a3

a2  0.249(log 10 MrBS )  0.977

and
a3  0.227(log 10 MrSB )  0.839

a2 = layer coefficient for base

a3 = layer coefficient for subbase
MrBS = elastic (resilient) modulus of the base
MrSB = elastic (resilient) modulus of the
subbase
Materials of construction (AC surface), a1

0.44
 Design considerations for the
AASHTO Flexible Pavement Design
 Inputs
 Performance criterion (PSI)
 Traffic (W18) and design life (analysis period)
 Foundation (Resilient Modulus)
 Material of Construction
 Environmental
 Drainage coefficient
 Reliability (ZR, So)
 Outputs
 Required pavement capacity (SN) for AC
Environment
Temperature and rainfall
affect the level of strength of
the subgrade, reflected on
Step the value of resilient
1 modulus. AASHTO developed
a chart that helps you to
estimate the effective
roadbed soil resilient
modulus using the
serviceability criteria (in
terms of “relative damage,
uf.”)
Determine the average uf.
value and obtain Mr from the
chart or the equation of uf. .
Step 2
 The bar on the right is
used twice: Once to read uf
Step 3 value for each month’s
sample Mr, then to read
annual average Mr using the
average u value.
 Design considerations for the
AASHTO Flexible Pavement Design
 Inputs
 Performance criterion (PSI)
 Traffic (W18) and design life (analysis period)
 Foundation (Resilient Modulus)
 Material of Construction
 Environmental
 Drainage coefficient
 Reliability (ZR, So)
 Outputs
 Required pavement capacity (SN) for AC
 Drainage

 The effect of drainage on the performance of

flexible pavements is considered with respect
to the effect water has on the strength of the
base material and roadbed soil.
 This effect is expressed by the drainage
coefficient, mi. This value is dependent on the
drainage quality and the percent of time
pavement structure is exposed to moisture
levels approaching saturation.
Drainage Quality and Recommended “mi”
Values

Step 1

If “Fair” and
30% exposure,
then mi is 0.80.

Step 2
 Design considerations for the
AASHTO Flexible Pavement Design
 Inputs
 Performance criterion (PSI)
 Traffic (W18) and design life (analysis period)
 Foundation (Resilient Modulus)
 Material of Construction
 Environmental
 Drainage coefficient
 Reliability (ZR, So)
 Outputs
 Required pavement capacity (SN) for AC
 Definition of Reliability

 Reliability is the probability that serviceability

will be maintained at adequate levels from a
user’s point of view, throughout the design life
of the facility.
 Reliability is the probability that the load
applications a pavement can withstand in
reaching a specified minimum serviceability
level is not exceeded by the number of load
applications that are actually applied to the
pavement.
 Reliability is the probability that the pavement
system will perform its intended function over
its design life and under the conditions
 Selection of Reliability Level (AASHTO
1993)
Functional Classification Recommended Level of Reliability

Urban Rural

Interstate and Freeways 85-99.9 80-99.9

Principal Arterials 80-99 75-95

Collectors 80-95 75-95

Local 50-80 50-80

Standard Deviation Values
• Flexible Pavement: 0.4 to 0.5
• Rigid Pavement: 0.3 to 0.40
Reliability Table (ZR – Values)

Reliability, R Standard Normal

(percent) Deviate, ZR

50 -0.000

60 -0.253

70 -0.524

75 -0.674

80 -0.841

85 -1.037

90 -1.282

95 -1.645

98 -2.054

99 -2.327
 Design considerations for the
AASHTO Flexible Pavement Design
 Inputs
 Performance criterion (PSI)
 Traffic (W18) and design life (analysis period)
 Foundation (Resilient Modulus)
 Material of Construction
 Environmental
 Drainage coefficient
 Reliability (ZR, So)
 Outputs
 Required pavement capacity (SN) for AC
Structural Number, SN

SN  a1 D1  a2 D2 m2  a3 D3 m3

SN = surface structure number

ai = layer coefficient for layer i
Di = layer thickness (inches) for layer i
mi = drainage coefficient for layer i
 Required Structural Number, SN
 PSI 
log10  
 4.2 1.5 
log10 (W18 )  Z R S o  9.36 log10 ( SN  1)  0.20  1, 094
0.4 
( SN 1) 5.19

 2.32 log10 ( Mr )  8.07

W18 = predicted traffic load in ESAL (18,000 lb loads)
ZR = standard normal deviate for specified reliability
R
S0 = combined standard error of the traffic prediction
and performance prediction (typically about 0.45 for
flexible pavements)
PSI = pi – pt = difference between initial serviceability
index and terminal serviceability index
Mr = resilient modulus (psi)
SN = structural number (layer thickness)
Flexible Pavement Nomograph
 AASHTO Flexible Pavement Example # 1

The expected design ESAL is 13x106. The pavement structure is to

consist of asphalt concrete with an elastic modulus of 400,000 psi, a
granular base with a resilient modulus of 26,000 psi, and a granular
subbase with a resilient modulus of 12,000 psi. The resilient modulus of
the subgrade is 8,000 psi. It is estimated that it will take a week for water
to drain from the pavement and that the pavement will be saturated about
15 percent of the time.

Use the nomograph to determine the structural numbers for the various
layers. This requires values for R, So, W18, Mr, and PSI. These are as
follows:
R = 90% (given)
So = 0.45 (typical for flexible pavements)
W18 = 13 x 106 (given)
PSI = pi – pt = 4.2 -2.5 = 1.7
Layer Coefficient for Asphalt Layer (a1)
AASHTO Example # 1
 Example Solution #1

For SN1, the structural number of the asphalt surface, it is the resilient
modulus of the base, which is next lowest layer; for SN2, it is the
resilient modulus of the subbase, etc. Thus for SN1, MR= 26,000 psi, for
SN2, MR= 12,000 psi, and for SN3, MR= 8,000 psi.
From the nomograph,
SN1 = 3.3; SN2 = 4.4; SN3 = 5.0
Determine layer coefficients:
a1= 0.42
a2= 0.249 (log10MRBS) – 0.977 = 0.249[log10(26,000)] – 0.977 = 0.12
a3= 0.227 (log10MRSB) – 0.839 = 0.227[log10(12,000)] – 0.839 = 0.09
Determine drainage coefficients:
Time to drain is 1 week. This would be characterized as “Fair.”The
pavement is expected to be saturated about 15 percent of the time.
m2= m3= 0.90
 Minimum Thickness Requirement

SN  a1 D1  a2 D2 m2  a3 D3 m3

SN1  a1 D1
SN 2  a1 D1  a2 D2 m2
Proceed in SN 3  a1 D1  a2 D2 m2  a3 D3 m3
this direction
 Final Design Thicknesses

Solve for layer thicknesses:

D1= SN1/a1 = 3.3/0.42 = 7.9 in. Round to 8.0 in

SN1* = a1D1*= (0.42)(8 in) = 3.36

D2= (SN2- SN1*)/(a2m2) = (4.4 – 3.36)/(0.12 x 0.90) = 9.6 in Round to 9.5 in

SN2*= SN1*+ a2m2D2* = 3.36 + (0.12)(0.90)(9.5) = 4.39

D3= (SN3- SN2*)/(a3m3) = (5.0 – 4.39)/(0.09 x 0.90) = 7.5 in

The Pavement will consist of 8.0 in of asphalt concrete surface, 9.5 in. of
granular base, and 7.5 in of granular subbase.
 AASHTO Example #2
Given:
ESAL = 2 x 106
One week for water to be drained
Saturation level moisture exposure = 30% of the time
AC’s Mr at 68oF = 450,000 lb/in2
CBR of base course =100, Mr or E = 31,000 lb/in2
CBR of subbase =22, Mr or E = 13,500 lb/in2
CBR of subgrade = 6, Mr or E = 1500*CBR= 6*1500 = 9000
lb/in2
Parameter values:
Reliability level (R ) = 99%
Standard Deviation (So) = 0.49,
Initial serviceability, pi = 4.5
Terminal serviceability, pt = 2.5
Drainage mi values = 0.8 for “Fair” category” and “Greater than
25%”
Home Problem #2

For subbase,
Mr=13,500 For base
course,
Mr=31,000

SN1= 2.6

This line is for the SN3= 4.4

subgrade: Mr=1500*6=9000
SN2= 3.8
Required Structural Number for
Individual Layers
AASHTO Design Procedure

RIGID PAVEMENTS
 Rigid Pavement Design Equation
 PSI 
log1 0  
log(Wt18 )  Z R S o  7.35 log10 ( D  1)  0.06   3.0 
1.624107
0.4 
( D 1)8.4 6
 
 
 S cCd ( D 0.75
 1.132) 
 (4.22  0.32 pt ) log10  
 215.63 J  D 0.75  18.42  
 0.25  

   E c / k  

S’c = modulus of rupture (psi) for Portland cement concrete

J = load transfer coefficient
Cd = drainage coefficient
Ec = modulus of elasticity (psi) for Portland cement concrete
k = effective modulus of subgrade reaction (pci)
Design Chart for Rigid Pavements
Design Chart
for Rigid
Pavements
(Cont’d)
 Modulus of Subgrade Reaction (k)

 Without subbase
 Modulus of subgrade reaction (k) is directly obtained

from plate loading test. p

Stiffness k
w
 Plate load test: use 30 inch diameter plate for k (“P”
is the pressure and “w” is deflection).
 Correlation between subgrade reaction (k) and
resilient modulus (MR)
MR
Stiffness k
19.4
 With base\Subbase layer
 Requires determination of composite modulus

 Modulus obtained based upon infinite subgrade

 Composite Modulus of Subgrade Reaction

 For each combination

of MR and ESB and
assume thickness of
subbase  determine
kc (k∞)
 kc (k∞) composite
modulus of subgrade
reaction
Effect of Rigid Foundation on K-Value
Chart for Estimating Relative Damage
 Loss of Support (LS)

 To account for potential loss of support by

foundation erosion or differential soil
movement.
 Loss of support value varies from 0 to 3.
 LS = 0: when slab and subgrade are in full contact
 LS = 3: when a significant portion of the slab is not
in contact with the subgrade
 Typical values of LS for design purposes
Loss of Support Correction for K-Value
Example:
Effective
Modulus of
Subgrade
Reaction
 Load Transfer Coefficient (J)

 Factor to account for the ability of a concrete pavement structure to

transfer load across joints and cracks.
 Increase in load transfer will decrease the load transfer coefficient.
 AASHO road test J = 3.2 (dowel joint with no tied PCC).
Drainage Coefficient (Cd)
 Selection of Reliability Level

Functional Classification Recommended Level of Reliability

Urban Rural

Interstate and Freeways 85-99.9 80-99.9

Principal Arterials 80-99 75-95

Collectors 80-95 75-95

Local 50-80 50-80

Standard Deviation Values
• Flexible Pavement: 0.4 to 0.5
• Rigid Pavement: 0.3 to 0.40
Layer Equivalency Factor for Single Wheel
(Rigid)
 Rigid Pavement Design Example

Using the AASHTO rigid design procedure, design a pavement

for a motorway design. The expected design ESAL is 26x106.
The pavement structure is to consist of Portland cement
concrete with an elastic modulus of 5.0x106 psi and a modulus
of rupture of 650 psi, and a cement-treated granular subbase
with an elastic modulus of 1,000,000 psi (insensitive to
season). The pavement is to be plain jointed concrete with an
asphalt concrete shoulder. The elastic modulus of the
subgrade is 5,500 psi during the dry season (May to Oct) and
4,500 psi during the wet season (Nov to Apr). There is no
bedrock or other rigid material pavement and that the
pavement will be saturated about 25 percent of the time with
good quality drainage. Assume that a reliability level of about
95 percent is required and that the initial serviceability index is
4.2 and the final serviceability index is 2.5.
 Example Solution

Determine the effective modulus of subgrade reaction:

Assume a subbase thickness of 12 in. and a slab thickness of 10
in.
With DSB= 12 in., ESB= 1,000,000 and MR= 4,500, the composite k
value k= 1,000 pci. For MR = 5,500, k= 800 pci.
Since there is no rigid material within 10 ft. of the surface, these
k values do not need to be modified to account for a rigid
foundation.

for k = 1,000 and slab thickness of 10 in., relative damage ur = 0.67.

for k = 800, ur = 0.75.
From the table below, the sum of ur is 8.52, and the average
value is 8.52/12 = 0.71. From Figure the effective modulus of
subgrade reaction k = 925. Assuming LS = 1.0, the corrected
effective modulus of subgrade reaction is 260 pci.
 Solution
Trial subbase: Type Cement-treated granular Depth to rigid foundation (feet) >10
Thickness (inches) 12 Projected slab thickness (inches) 10
Loss of support, LS 1.0
(1) (2) (3) (4) (5) (6)

Month Roadbed Modulus MR (psi) Subbase modulus ESB (psi) Composite k value K value (pci) on rigid Relative damage ur
(pci) foundation

Jan. 4,500 1,000,000 800 - 0.75

Feb. 4,500 1,000,000 800 - 0.75

Mar. 4,500 1,000,000 800 - 0.75

Apr. 4,500 1,000,000 800 - 0.75

May. 5,500 1,000,000 1,000 - 0.67

June 5,500 1,000,000 1,000 - 0.67

July 5,500 1,000,000 1,000 - 0.67

Aug. 5,500 1,000,000 1,000 - 0.67

Sept. 5,500 1,000,000 1,000 - 0.67

Oct. 5,500 1,000,000 1,000 - 0.67

Nov. 4,500 1,000,000 800 - 0.75

Dec. 4,500 1,000,000 800 - 0.75

ur = 8.52
Loss of Support
 Solution for the Problem
Average ur = 0.71
Effective modulus of subgrade support, k (pci) = 925
Corrected for loss of support; k (pci) = 260
Determine slab thickness:
Enter the nomograph with the following values:
k = 260 pci; EC = 5 x 106 psi
SC = 650 psi J = 3.2
S0 = 0.35 R = 95 % (ZR= -1.645)
PSI = pi – pt = 4.2 – 2.5 = 1.7 Cd = 1.00
W18 = 26 x 106
From the nomograph, the required slab thickness = 12 in.
The k value was determined assuming a slab thickness of 10 in.,
so this must be recalculated. Assuming a slab thickness of 12
in., ur= 1.9 for k = 1,000 and ur= 2.1 for k = 800. Recalculating the
corrected effective modulus of subgrade support (see table
AASHTO Design Chart
Nomographic
Solution (Cont’d)
 Second Trial
Trial subbase: Type Cement-treated granular Depth to rigid foundation (feet) >10
Thickness (inches) 12 Projected slab thickness (inches) 12
Loss of support, LS 1.0
(1) (2) (3) (4) (5) (6)

Month Roadbed Modulus MR (psi) Subbase modulus ESB (psi) Composite k value K value (pci) on rigid Relative damage ur
(pci) foundation

Jan. 4,500 1,000,000 800 - 2.10

Feb. 4,500 1,000,000 800 - 2.10

Mar. 4,500 1,000,000 800 - 2.10

Apr. 4,500 1,000,000 800 - 2.10

May. 5,500 1,000,000 1,000 - 1.90

June 5,500 1,000,000 1,000 - 1.90

July 5,500 1,000,000 1,000 - 1.90

Aug. 5,500 1,000,000 1,000 - 1.90

Sept. 5,500 1,000,000 1,000 - 1.90

Oct. 5,500 1,000,000 1,000 - 1.90

Nov. 4,500 1,000,000 800 - 2.10

Dec. 4,500 1,000,000 800 - 2.10

ur = 24.00
 Final Solution
Average ur = 2.00
Effective modulus of subgrade support, k (pci) = 900
Corrected for loss of support; k (pci) = 250
Recalculating the slab thickness for k = 250 results in a slab
thickness of 12 in., which agrees with that assumed. Use a 12
in. PCC slab with a 12 in. cement-treated subbase.