pavement thickness design methods

Pavement Thickness Design Methods

Municipal Streets SeminarNovember 12, 2014

Ames, Iowa

Eric Ferrebee - ACPA

U.S. JPCP Roadway Design Methods

AASHTOWarePavement ME (previously known as DARWin-ME and MEPDG)

AASHTO 93 (software as ACPA WinPAS)

ACPA StreetPave

325 & 330

Design Method Basis

Mechanistic – Purely scientific and based on measured, defendable scientific rules and laws

Empirical – Based on observations or experimentation and requires a lot of tests to connect all the relationships

𝜖𝜖 = 𝜎𝜎/𝐸𝐸 ∆L = α ∗ ∆𝑇𝑇 ∗ 𝐿𝐿𝑜𝑜

𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪 = 𝒍𝒍𝒍𝒍𝑪𝑪𝒍𝒍𝒍𝒍 + 𝒆𝒆𝑪𝑪𝒆𝒆𝑪𝑪𝑪𝑪𝒍𝒍𝑪𝑪𝒆𝒆𝒆𝒆𝑪𝑪𝒆𝒆 + 𝒆𝒆𝑪𝑪𝒆𝒆𝒆𝒆𝑪𝑪𝑪𝑪𝑪𝑪𝒍𝒍

AASHTO 93 / ACPA WinPAS

acpa.org/winpas

AASHO Road Test (1958-1960)

Wholly empirical Included 368 concrete and 468 asphalt sections | focus was highway pavement

Subgrade = Clay Soil

Necessary Thickness was Guessed!

Max Single Axle

Max Tandem Axle

Sections Loaded for 2 Yrs | 1.1 Mil Reps

Performance Estimated Subjectively

Present Serviceability Index (PSI)4.0 – 5.0 = Very Good3.0 – 4.0 = Good2.0 – 3.0 = Fair1.0 – 2.0 = Poor0.0 – 1.0 = Very Poor

“Failure” at the Road Test considered @ 1.5

Typical U.S. state agency terminal serviceability in practice = 2.5

Note on Inference Space of ‘93

• •• •• •

•

Data Limits(AASHO Road Test)

CurrentDesigns>100 million

1.1mil

AXLE LOAD REPETITIONS

PAVEM

ENT

THIC

KN

ESS

Current design traffic is far beyond AASHO road test limits

Don’t Just Take My Word…“The current design guide and its predecessors were largely based on design equations empirically derived from the observations AASHTO’s predecessor made during road performance tests completed in 1959-60. Several transportation experts have criticized the empirical data thus derived as outdated and inadequate for today’s highway system. In addition, a March 1994 DOT Office of Inspector General report concluded that the design guide was outdated and that pavement design information it relied on could not be supported and validated with systematic comparisons to actual experience or research.”

…this is why Pavement ME exists!

++

−∆

+−++=

46.8

7

)1(10*624.11

5.15.406.0)1(*35.7*)(

D

PSILogDLogsZESALLog oR

−

−−+

25.075.0

75.0'

)/(42.18**63.215

)132.1(***)*32.022.4(

kEDJ

DCSLogp

c

dct

Standard Normal Deviate

OverallStandard Deviation Thickness

Change in Serviceability

Terminal Serviceability

DrainageCoefficient

Load Transfer

Modulus ofRupture

Modulus of Elasticity

Modulus ofSubgrade Reaction

1986-93 JPCP AASHTO 93 Equation

Traffic

MEPDG / DARWin-ME /AASHTOWare Pavement ME

Pavement ME Design

Not “perfect” & not intended to be a “final” productComplex and relatively costlyFor highways and NOT street, road, parking lot, etc.

=+Mechanistic Calculation

of Responses

Empirical Tie to

Ground

Pavement Performance Prediction

JPCP Calibration – BIG INF. SPACE!

LTPP GPS-3 & RPPR JPCP Sections LTPP SPS-2, MnROAD, & AASHO JPCP Sections

AASHTO 93 vs. ME

AASHTO Pavement ME

AASHTO 93

50+ million load reps

1.1 million load reps

Wide range of structural and rehabilitation designs

Limited structural sections

1 climate/2 years

All climates over 20-50 years

1 set of materials

New and diverse materials

Sounds Easy Enough, Right?

INPUTS, INPUTS, INPUTS!!!!

OUTPUTS, OUTPUTS, OUTPUTS!!!

Some Agencies Trying to Simplify

Note the inputs that have been deemed worth

varying…

…designers have a new

idea of “what

matters”!

ACPA StreetPave

acpa.org/streetpave

ACPA StreetPave

Roots date back to the 1960s PCA MethodTailored for streets and roadsFailure modes are cracking and faulting

Total trucks in design lane over the design life… calculated from trucks/day (2-way), traffic growth rate (%/yr), design life (yrs), directional distribution (%) and design lane distribution (%)

Traffic Spectrum + CountsSingle Axles

Axle Load (kip) Axles/1,000 Trucks34 0.1932 0.5430 0.6328 1.7826 3.5224 4.1622 9.6920 41.8218 68.2716 57.07

Tandem AxlesAxle Load (kip) Axles/1,000 Trucks

60 0.5756 1.0752 1.7948 3.0344 3.5240 20.3136 78.1932 109.5428 95.7924 71.16

Equivalent stress at the slab edge:

Me = equivalent moment, psi; different for single, tandem, and tridem axles, with and without edge support - func on radius of relative stiffness, which depends on concrete modulus, Poisson’s ratio, and thickness and the k-value

hc = pavement thickness, in.f1 = adjustment for the effect of axle loads and contact areaf2 = adjustment for a slab with no concrete shoulderf3 = adjustment to account for the effect of truck (wheel) placement at the slab edgef4 = adjustment to account for approximately 23.5% increase in concrete strength

with age after the 28th day and reduction of one coefficient of variation (COV) to account for materials variability

Traffic Loads Generate Stresses

𝜎𝜎𝑒𝑒𝑞𝑞 =6 ∗ 𝑀𝑀𝑒𝑒

ℎ𝑐𝑐2 ∗ 𝑓𝑓1 ∗ 𝑓𝑓2 ∗ 𝑓𝑓3 ∗ 𝑓𝑓4

Stress Ratio (SR) = Stress / Concrete Strength

Limit Stress Ratio to Allow Design Reps

0.4

0.5

0.6

0.7

0.8

0.9

1

1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 1.E+10

Stre

ss R

atio

Repetitions

Fatigue Data

StreetPave R=95%

Inference space normalized to SR

StreetPavemakes slab thicker to limit stress ratio low enough to achieve the design traffic repetitions

A Conservative Approach!

StreetPave fatigue calculation should be conservative relative to ME Design because:

Size Effects – Slabs have a greater fatigue capacity than beamsSupport – The beam test has a k-value for support of 0 psi/in.!

L/3Span Length = L

d=L/3

…versus…

Faulting Design in StreetPave

If dowels used, faulting mitigated & fails by cracksNo faulting data collected at the AASHO road test so model developed in 1980s using field performance data from WI, MN, ND, GA, and CASimilar to cracking models, the pavement is made thicker, as necessary, until faulting model predicts that the pavement will not fail by faulting during the design lifeStreetPave’sweak point

StreetPave Accepted in MN

http://www.dot.state.mn.us/stateaid/admin/memos/12-sa-03.pdfhttp://www.dot.state.mn.us/research/documents/201210.pdf

And Its Use is Growing!

Also “approved” in VA and many other state, city, and county engineers are using it in the U.S.StreetPave used in design tables in:

ACI 325 and 330 documentsDr. Norb Delatte’s textbook Concrete Pavement Design, Construction, and Performance

Internationally, used in Australia, Portugal, Mexico, Uruguay, Argentina, Chile, etc.

StreetPave | Project Details

StreetPave | Traffic Details

StreetPave | Design Details | General

StreetPave | Design Details | Concrete

StreetPave | Design Results

StreetPave | Design Report

Comparison of Results and U.S. Trends

100

125

150

175

200

225

250

275

300

325

350

4

5

6

7

8

9

10

11

12

13

14

- 10,000,000 20,000,000 30,000,000 40,000,000 50,000,000

Requ

ired

Thic

knes

s (m

m)

Requ

ired

Thic

knes

s (in

.)

Design Lane ESALs

AASHTO 93 (ACPA WinPAS)

AASHTOWare Pavement ME @ ORD

AASHTOWare Pavement ME @ PHX

ACPA StreetPave

Doweled JPCP Thickness Comparison

remember AASHTO 93 limit?

Top 10 ME Design Most Sensitive

1. Concrete Flexural Strength at 28-Days2. Concrete Thickness3. Surface Shortwave Absorptivity (SSA)4. Joint Spacing5. Concrete Modulus of Elasticity at 28-Days6. Design Lane Width with a 14 ft (4.3 m) Widened Slab7. Edge Support via Widened Slab8. Concrete Thermal Conductivity9. Concrete Coefficient of Thermal Expansion (CTE)10. Concrete Unit Weight

Red = only ME Design input… the VALUE of the software!Blue + Bold = common for all

100

125

150

175

200

225

250

275

300

325

350

4

5

6

7

8

9

10

11

12

13

14

400 450 500 550 600 650 700 750 800

Requ

ired

Thic

knes

s (m

m)

Requ

ired

Thic

knes

s (in

.)

Concrete Flexural Strength (psi)




ACPA StreetPave

Flexural Strength Sensitivity

100

125

150

175

200

225

250

275

300

325

350

4

5

6

7

8

9

10

11

12

13

14

3,000,000 3,500,000 4,000,000 4,500,000 5,000,000 5,500,000 6,000,000

Requ

ired

Thick

ness

(mm

)

Requ

ired

Thick

ness

(in.

)

Modulus of Elasticity (psi)




ACPA StreetPave

Modulus of Elasticity Sensitivity

… in reality, need to change strength too…(20.7 GPa) (41.4 GPa)

0

25

50

75

100

0

0.5

1

1.5

2

2.5

3

3.5

4

- 10,000,000 20,000,000 30,000,000 40,000,000 50,000,000

Thic

knes

s Red

uctio

n w

/Edg

e Su

ppor

t (m

m)

Thic

knes

s Red

uctio

n w

/Edg

e Su

ppor

t (in

.)

Design Lane ESALs




ACPA StreetPave

Thickness Reduction w/ Edge Support

100

125

150

175

200

225

250

275

300

325

350

4

5

6

7

8

9

10

11

12

13

14

50% 60% 70% 80% 90% 100%

Requ

ired

Thic

knes

s (m

m)

Requ

ired

Thic

knes

s (in

.)

Reliability




ACPA StreetPave

Reliability Sensitivity

100

125

150

175

200

225

250

275

300

325

350

4

5

6

7

8

9

10

11

12

13

14

0 100 200 300 400 500

Requ

ired

Thic

knes

s (m

m)

Requ

ired

Thic

knes

s (in

.)

Static k-value (psi)




ACPA StreetPave

k-value Sensitivity

Very few designed for < 100

psi/in. (27

MPa/m)?

(136 MPa/m)

U.S. Agencies Quickly Changing

Summary of State Agency practice in 2005:

At the end of 2013, 41 state agencies had performed ME Design calibration and implementation efforts, indicating a relatively quick shift from AASHTO 93.

Design Method Used

Percent of Responding Agencies State Agency

AASHTO 72/86/93 85%

AR, AZ, DE, FL, ID, IN, IA, KS, MD, MI, NV, NC, OH, OK, SC, SD, TN, UT, VA, WA, WV, WI, WY

AASHTO MEPDG 4% MOPCA Method 11% HI, IN, IA

State-Developed 7% IL, MT

U.S. Roadway Length (lane miles)

State Agency,

19%

County, 44%

Town, 32%

Other, 1% Federal, 3%

Source: HM-10, 2012 FHWA Highway Statistics

AASHTO tools are being developed for these owners…

City, county, and other local engineers need

to decide what to use locally because

Pavement ME will not trickle down due to its

cost and complexity!

Overlay Design – Bonded Concrete on Asphalt

Bonded over Asphalt/Composite

AASHTO 93: not applicableAASHTO ME: limited to a joint spacing > 10 ft (3 m) and not applicable to thinner (2-6 in. [50-150 mm]) concrete overlays; refers people to ACPAACPA BCOA and StreetPave: account for bond to asphalt and short slab size, fibers, etc.


Not an extension of an existing design method; a completely custom-built design methodFor ACPA BCOA and StreetPave

based primarily on FHWA-ICT-08-016

Other methods include CDOT 6x6x6 and recently released FHWApooled fund TPF-5(165)

FHWA pooled fund TPF-5(165)

Pitt BCOA ME

FHWA Pooled Fund Study TPF 5-165

BCOA ME failure modes5 to 7 ft

Long. & DiagCrack

Positive ΔT Negative ΔT

< 4.5 ftCorner Break

Positive ΔT

10 x 12 ft12 x 12 ft12 x 15 ft

Trans. Crack

BCOA-ME

Inputs

Stress for corner cracks

Stress for long. & diag. cracks

Fatigue model hpcc

Jt. Spacing < 4.5 ft

Jt. Spacings5 to 6 ft

Pitt Model

PCA Model

ACPA Model

Stress for trans. cracks

CDOT Model

Jt. Spacings10 x 12 ft12 x 12 ft15 x 12 ft

Wheel wander considered

Variation in HMA stiffness

Seasonal variation

Daily variation

Stiff HMASoft HMA

Stiff HMASoft HMA

Effective temp. gradient

Negative ΔT

Positive ΔT

0%2%4%6%8%

10%12%14%16%

-6.5

-5.5

-4.5

-3.5

-2.5

-1.5

-0.5 0.5

1.5

2.5

3.5

Rel

ativ

e Fr

eque

ncy

Equivalent linear temperature gradient, ºF/in

Design input required: Effective temp. gradient (ETG)

Regional variation in temp. grad.

University of Pittsburgh Department of Civil & Environmental Engineering

0

0.02

0.04

0.06

0.08

0.1

-3 -2 -1 0 1 2 3

Rel

ativ

e Fr

eque

ncy

Pavement temperature gradient (F˚/in)Pittsburgh Phoenix Miami

Vandenbossche and Mu (2010)

Calibration - Stress Adj. Factors

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

Cal

cula

ted

stre

ss a

djus

tmen

t fa

ctor

Generated stress adjustment factor

Stress adjustment factor (Joint spacing > 4ft x 4ft)Linear (Series2)

Pitt BCOA-ME Web App

Pitt BCOA-ME Web App

ACPAk-valueweb app

Annual Mean Daily Average Temp

Results of the BCOA-ME

• BCOA ME is <1/3 ACPA BCOA app, <1/3 CDOT procedure and the remainder was developed at Pitt

• Comparison to ACPA BCOA app:• JS ≤ 4.5 ft, the model is similar but thicknesses

possibly greater because BCOA ME was calibrated to field performance, accounting for loading directly in the wheel path

• JS > 4.5 ft, it will not compare because failure mode changes (e.g., no longer corner cracking); BCOA ME will almost always be thinner than ACPA BCOA for a 6’ x 6’ JS

• Comparison to CDOT design procedure:• Depends on failure mode

Design considerations

Overlaythickness

Load carrying layer

HMA

PCC

Bond is essential

Bond enhances performance

PCC = 4 in or lessHMA

PCC = 5 in or more PCC

PCC = 4 in or less

PCC = 5 in or more


10 in HMA

4 in PCC6 in HMA

6 in PCC4 in HMA


10 in HMA

4 in PCC6 in HMA

6 in PCC4 in HMA

Critical Stress for 6ft x 6ft slab, psiPCC Thickness 3 in 4 in 6 in

HMA Thickness4 in 352 339 2746 in 246 234 2118 in 198 191 177

Thank you.Questions? FEEDBACK!

Main Website | acpa.orgConcrete Wiki | wiki.acpa.org

App Library | apps.acpa.orgDesktop Software | software.acpa.org

Resources | resources.acpa.orgOn-Demand Training | ondemand.acpa.org

Live Online Training | webinars.acpa.orgYour Local Contact | local.acpa.org

Eric Ferrebee Technical Services Engineer

American Concrete Pavement [email protected] | 847.423.8709

Download StreetPave:www.acpa.org/streetpave

pavement thickness design methods

Documents