Pavement Design Task Group Feedback

Copyright @ CSIR 2005 www.csir.co.za

Pavement Design Task Group Feedback

RPF meeting

23 November 2005


Sequence of events

Scientific Research Council Act (Act No 46 of 1988, as amended by Act 71/1990)

• RPF May 2005– Establish task group

• 1st Meeting– Discussion of role and operation of the task group– Presentation on potential improvements to mechanistic-

empirical (ME) design method– Unpacking of ME design method core

• 2nd Meeting– Presentation on a project performance based design method– Presentation of simulation models– Discussion of the deliverable from the process


Sequence of events


• 3rd Meeting– Integration of the performance based information system

and the ME-design component into a single design method– Design sensitivity analysis based on the current ME design

method– Details of the performance based information system

• 4th Meeting– Unpacking the details of the ME-design component– Development of a R&D matrix and rating of the R&D areas


Task group members

• Prof M de Beer (CSIR);• Prof E Horak (UP);• Mr M Henderson

(PAWC);• Prof K Jenkins (US);• Dr F Jooste (MAS);• Dr G Jordaan

(Tshepega);• Mr L Kannemeyer

(SANRAL);• Dr F M Long (MAS);

• Mr P Olivier (J&G);• Mr B Perrie (C&CI);• Mrs E Sadzik

(Gautrans);• Mr A Taute (Vela VKE);• Mr H L Theyse (CSIR);• Dr I van Wijk (Africon);

and• Dr H Wolff (Nyeleti).


General recommendations:Deliverable from the task group


• The task group will deliver a research and development (R&D) plan that will guide the development of a pavement design method for southern Africa that – allows for a range of design applications of different

importance and – caters for a user group with different levels of design

experience• R&D plan to be executed under the auspices of the

Road Pavement Research Advisory Committee (RPRAC)– Project management may be referred back to sub-groups of

the task group


General recommendations


• The R&D plan is aimed at developing an improved design method based on– A performance based information system– Mechanistic-empirical design concepts

• Initial intention to include rigid and block pavement design– Details of design sufficiently different to treat

separately– Urgent need to develop single design guide for all

types of pavements




• The R&D plan only covers R&D activities to develop improved design method– Dissemination of individual research results and design method as

a whole– Not descriptive in terms of delivery mechanisms such as software

packages– RPRAC may considered the development of a software package as

a form of delivery mechanism• The improved design method should close the gap between

ME-design and day-to-day engineering as far as possible• Conventional ME-design focus almost exclusively on the effect

of the imposed stresses– Recent research on unbound material shows the effect of field

variables such as density and saturation is equally important– The improved design method must incorporate the effect of field

variables




• Initially a separate task group for design traffic– Design traffic estimation is an integral part of

pavement design– Design traffic task group combined with pavement

design task group– The improved design method should allow for

gradual transition from the E80 concept to mixed traffic design

• Revision of guideline documents containing sound engineering practice excluded form the R&D plan– Close integration required between guidelines and

the design method


Integration of guidelines and design method

PavementDesign

Knowledge base

Sou

nd e

ngin

eerin

g pr

actic

e

Pav

eme

nt d

esig

n m

etho

ds

PavementDesign

Knowledge base

Sou

nd e

ngin

eerin

g pr

actic

e

Pav

eme

nt d

esig

n m

etho

ds


Performance based information system


• Based on performance data collected from – Rehabilitation investigation documentation – APT test results

• Statement of fact rather than interpretation of data and modelling


Design Method

Integration of performance based and ME-design components

Entry level

Intermediate level

Advanced level

ME-design component

Design scenario:• Routine design• Low design experience• Use only known materials• Conventional material classification

Design scenario:• Important design • Seasoned professional designer• Test input parameters for materials

Design scenario:• Very important design• Special investigations • Specialist designer• Unusual materials• Test input and performance of materials

Performance based

information system


Design Option

Click here for Design Report

Click here for Situation Summary

Click here for Rehab. Report


Click here…


Click here…


Click here…

• 8 Years to Resurface

• 1278 H/day

• 8 MESA to Rehab

• Terminal Rut = 16 mm

• Not Resurfaced

• 987 H/day

• 5 MESA to Rehab



• 1250 H/day

• 9 MESA to Rehab



• 1400 H/day

• 12 MESA to Rehab


Performance Database

Performance based information system


Mechanistic-empirical design component


• Basic ME procedure– Fully determines the accuracy of the ME-design

component– Requires engineering knowledge and

understanding

• Various implementation schemes– Allows for spatial and time-based variation– Enhances the interpretation of results– Requires mathematical and computer

programming skills


Temperature

Moisturecontent

Density

Grading

Atterberglimits

Bindercontent

Binderproperties

Geometry

Axle loadhistogram

Other?

Other?

Contact stresshistogram

Grading

Fixedload

Fixedcontact stress

Mr = Constant

Mr = f (Temp)

Mr = f (Dens,saturation)

Mr = f (Bulk and shear

stress)

Mr = f (Strain)

Linearvisco-elastic

Other?

UCS

Stress and strain at break

Time/previousloading

HM

AU

nbou

ndS

tabi

lized

Mat

eria

l dat

a

Trafficdata

Resilientresponseanalysis

Input layers

CurrentFuture

Resilientresponse models

Damage models

Fatigue

Plastic strain/permanent deformation

Other?

Thermal cracking

HM

AU

nbou

nd


Other?

Other?

Sta

biliz

ed Stiffness reduction


Other?

Crushing

Sub

grad

e Plastic strain/permanent deformation

Other?

Structuralcapacityestimate

Com

pute

rso

lutio

n


Basic me-design

module

Trafficloading

Spatial variationof field variables

Axle load

Frequency

Contact stress

Frequency

T1

RD1, S1

1b, 1

b

T2

RD2, S2

2b, 2

b

T3

RD3, S3

3b, 3

b

Tn

RDn, Sn

nb, n

b

Layer thicknessTemperatureBinder contentRelative densitySaturationGradingetc.

Pre-processing Post-processing

Extent of fatigue

Time Fatigue life

Frequency

Rut

Time Rut life

Frequency

Terminal rut

Riding quality

Time Riding quality life

Frequency

Unacceptable


Research & Development Plan


• Main components of the R&D plan– Integration level project– Performance based information system– Demand analysis

• Traffic and natural environment– Material response models– Pavement response models– Damage models– Probabilistic and recursive implementation schemes

• R&D needs formulated for each component• Short, medium and long-term actions formulated for

each R&D need• Each action rated in term of benefit/impact and

effort/cost


Execution of R&D plan:Integration of activities

Integration level project:• Integrate output from individual projects into a single design system• Develop documentation on the use of the system• Present seminars and workshops on the use of the design method`

Dem

and

an

alys

is:

•T

raff

ic•

Env

ironm

ent

Mat

eria

l res

ilien

t re

sp

on

se

mo

del

s

Pa

vem

ent

res

ilien

t re

spo

nse

m

od

els

Da

mag

e m

od

els

Imp

lem

enta

tio

n s

chem

es

Integration level project:• Integrate output from individual projects into a single design system• Develop documentation on the use of the system• Present seminars and workshops on the use of the design method`

Dem

and

an

alys

is:

•T

raff

ic•

Env

ironm

ent

Mat

eria

l res

ilien

t re

sp

on

se

mo

del

s

Pa

vem

ent

res

ilien

t re

spo

nse

m

od

els

Da

mag

e m

od

els

Imp

lem

enta

tio

n s

chem

es


Execution of R&D plan:Stress effect theme

Pavement resilient response model:• Correct modelling of the complex tyre/pavement contact patch in terms of shape and stress distribution

Demand analysis:• A tyre/pavement contact stress information system• A traffic volume and axle load information system• Vehicle dynamics

Contact stress and axle load histograms

and response histograms

Damage models:• Calibrated for the effect of field variables and stress regime

Implementation scheme:• Probabilistic, linear recursive and non-linear recursive simulation of cumulative damage

Damage per individual load cycle


Execution of R&D plan:Field variable effect theme

Pavement resilient response model:• Modelling the vertical variation in material response caused by vertical variation in field variables and stress condition

Material resilient response models:• Improved material response models calibrated for time dependent deterioration and field variable variation

Demand analysis:• An environmental information system• A field variable and engineering parameter information system including spatial and time variation

Design environment

Environment specific material response models

Damage models:• Calibrated for the effect of field variablesand stress regime

and response histograms

Implementation scheme:• Probabilistic, linear recursive and non-linear recursive simulation of cumulative damage

Damage per individual load cycle


Rating of short-term R&D objectives


• Integration level project– Integrate output from R&D projects into single

design method– Reality checks and validation– Develop manual on use of method

Integration project

0

2

4

6

8

10

1 2 3 4 5 More

Bin

Fre

qu

ency

Expected benefit

Integration project

0

2

4

6

8

10

50%

100%

150%

300%

500%

Mor

e

Bin

Fre

qu

ency

Benefit/Effort


Rating of short-term R&D objectives (continued)


• Material response models– Agreement between different deflection methods

Material response models

0

2

4

6

8

10

1 2 3 4 5 More

Bin

Fre

qu

ency

Expected benefit

Agreement between deflection methods

Material response models

0

2

4

6

8

10

50% 100% 150% 300% 500% More

Bin

Fre

qu

ency

Benefit/Effort

Agreement between deflection methods


Rating of short-term R&D objectives Summary


• Integration project and performance based information system– Average benefit ratings between 3,8 and 3,9– Average benefit/effort between 114 and 190 %

• Demand analysis– 3 out of 6 items had average benefit ratings between

3,5 and 3,9 the other items were between 2,8 and 3,2– Average benefit/effort between 96 and 159 %

• Material resilient response models– Average benefit ratings between 3,6 and 4,1– Average benefit/effort between 99 and 141 %


Rating of short-term R&D objectives Summary


• Pavement resilient response models– Average benefit ratings between 2,4 and 3,4– Average benefit/effort between 70 and 98 %

• Damage models– Average benefit ratings between 3,6 and 4,4– Average benefit/effort between 112 and 146 %

• Probabilistic and recursive implementation schemes– Average benefit ratings between 3,5 and 3,9– Average benefit/effort between 132 and 152 %


Recommendation to RPRAC

• Select projects based on– Ratings– Recent and ongoing research that may

fast-track certain projects– The achievement of the objective of the

R&D plan to developed an improved design method for flexible pavements

Documents

Pavement Design Task Group Feedback