fieldassessmentandspeciﬁcationreviewforroller-integrated ... · 2019. 7. 31. · time. if...

Hindawi Publishing CorporationAdvances in Civil EngineeringVolume 2011, Article ID 783836, 15 pagesdoi:10.1155/2011/783836

Research Article

Field Assessment and Specification Review for Roller-IntegratedCompaction Monitoring Technologies

David J. White, Pavana K. R. Vennapusa, and Heath H. Gieselman

Department of Civil, Construction, and Environmental Engineering, Center for Earthworks Engineering Research,Iowa State University of Science and Technology, Ames, IA 50011-8664, USA

Correspondence should be addressed to Pavana K. R. Vennapusa, [email protected]

Received 1 May 2011; Accepted 31 July 2011

Academic Editor: Sai K. Vanapalli

Copyright © 2011 David J. White et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Roller-integrated compaction monitoring (RICM) technologies provide virtually 100-percent coverage of compacted areas withreal-time display of the compaction measurement values. Although a few countries have developed quality control (QC) andquality assurance (QA) specifications, broader implementation of these technologies into earthwork construction operations stillrequires a thorough understanding of relationships between RICM values and traditional in situ point test measurements. Thepurpose of this paper is to provide: (a) an overview of two technologies, namely, compaction meter value (CMV) and machinedrive power (MDP); (b) a comprehensive review of field assessment studies, (c) an overview of factors influencing statisticalcorrelations, (d) modeling for visualization and characterization of spatial nonuniformity; and (e) a brief review of the currentspecifications.

1. Introduction

Roller-integrated compaction monitoring (RICM) technolo-gies refer to sensor measurements integrated into com-paction machines. Work in this area was initiated over 30years ago in Europe for smooth drum rollers compactinggranular soils and involved instrumenting the roller with anaccelerometer and calculating the ratio of the fundamentalfrequency to the first harmonic [1, 2]. Modern sensor tech-nologies, computers, and global positioning system (GPS)technologies now make it possible to collect, transmit, andvisualize a variety of RICM measurements in real time. Asa quality assessment tool for compaction of earth materials,these technologies offer tremendous potential for fieldcontrolling the construction process to meet performancequality standards. Recent efforts in the United States (US)have focused attention on how RICM technologies can beused in road building [3–5] and relating selected RICMparameters to mechanistic pavement design values.

Several manufactures currently offer RICM technologieson smooth drum vibratory roller configurations for com-paction of granular materials and asphalt, and nonvibratoryroller configurations for compaction of cohesive materials.

The current technologies calculate: (1) an index valuebased on a ratio of selected frequency harmonics for a settime interval for vibratory compaction [1, 2], (2) groundstiffness or dynamic elastic modulus based on a drum-ground interaction model for vibratory compaction [6–8],or (3) a measurement of rolling resistance calculated frommachine drive power (MDP) for vibratory and nonvibratorycompaction [9]. When the accelerometer-based measure-ment system provides automatic feedback control for rollervibration amplitude and/or frequency and/or roller speedcontrol, it is referred to as “intelligent” compaction and offersthe advantage of reducing the potential for drum “bouncing”.The MDP approach has the advantage of working in bothvibratory and static modes [9] and has its origin in thediscipline of terramechanics. Recent finding from the MarsExploratory Rover (MER) mission demonstrated that theMDP approach can be applied to determine Martian regolithcohesion and friction angle by monitoring the electrome-chanical work expended [10]. Future RICM technologiesmay provide information on soil mineralogy and moisturecontent but are currently only a subject of research.

Regardless of the technology, by making the compactionmachine a measuring device and insuring that compaction

2 Advances in Civil Engineering

requirements are met during construction, the compactionprocess can be better controlled to improve quality, reducerework, maximize productivity, and minimize costs [11].Recent advancements with global positioning systems (GPS)add a significant benefit with real time spatial viewing of theRICM values. Some of these technologies have recently beenimplemented on full-scale pilot earthwork and asphalt con-struction projects in the US [12–18], and its use is anticipatedto increase in the upcoming years. Effective implementationof this technology needs proper understanding of the rela-tionships between RICM values and traditional in situ pointtest compaction measurements (e.g., static or dynamic plateload test modulus, density, etc.). This builds confidence inthe technology and provides insight into the key parametersaffecting the machine measurement values.

The purpose of this paper is to provide: (a) an overviewof two technologies—compaction meter value (CMV) andmachine drive power (MDP); (b) a summary of fieldevaluation studies, (c) an overview of factors influencing thestatistical correlations, (d) modeling for visualization andcharacterization of spatial nonuniformity, and (e) a briefreview of the current specifications.

2. Overview of CMV and MDP Technologies

Compaction meter value (CMV) is a dimensionless com-paction parameter developed by Geodynamik that dependson roller dimensions, (i.e., drum diameter and weight)and roller operation parameters (e.g., frequency, amplitude,speed) and is determined using the dynamic roller response[19, 20]. CMV is calculated using (1), where C is a constant(used as 300 for the results presented in this paper), andA2Ω = the acceleration of the first harmonic componentof the vibration, AΩ = the acceleration of the fundamentalcomponent of the vibration [8]:

CMV = C · A2ΩAΩ

. (1)

According to Geodynamik [21], CMV at a given pointindicates an average value over an area whose width equalsthe width of the drum and length equal to the distance theroller travels in 0.5 seconds. At least two manufactures haveused the CMV technology as part of their RICM systems(Figure 1). The Geodynamik system also measures theresonant meter value (RMV) which provides an indicationof the drum behavior (continuous contact, partial uplift,double jump, rocking motion, and chaotic motion). RMVis not discussed in detail here, but it is important to notethat the drum behavior affects the CMV measurements [6]and therefore CMV must be interpreted in conjunction withRMV [22].

Machine drive power (MDP) technology relates themechanical performance of the roller during compaction tothe properties of the compacted soil. The use of MDP asa measure of soil compaction is a concept originated fromstudy of vehicle-terrain interaction [23]. The basic premiseof determining soil compaction from changes in equipmentresponse is that the efficiency of mechanical motion pertainsnot only to the mechanical system but also to the physical

properties of the material being compacted. More detailedbackground information on the MDP system is provided in[9]. The basic formula for MDP is

MDP = Pg −WV(

sinα +a

g

)− (mV + b), (2)

where Pg = gross power needed to move the machine (kJ/s),W = roller weight (kN), a = machine acceleration (m/s2),g = acceleration of gravity (m/s2), α = slope angle (rollerpitch from a sensor), V = roller velocity (m/s), and m (kJ/m)and b (kJ/s) = machine internal loss coefficients specific toa particular machine [9]. The second and third terms of (2)account for the machine power associated with sloping gradeand internal machine loss, respectively. MDP is a relativevalue referencing the material properties of the calibrationsurface, which is generally a hard compacted surface (MDP =0 kJ/s). Positive MDP values therefore indicate material thatis less compact than the calibration surface, while negativeMDP values would indicate material that is more compactedthan the calibration surface (i.e., less roller drum sinkage).In some recent field studies [13], the MDP output value hasbeen scaled to MDP80 or MDP40 depending on the modifiedsettings which are recalculated to range between 1 and 150using (3) and (4), respectively,

MDP80 = 150− 1.37 (MDP), (3)MDP40 = 150− 2.75 (MDP). (4)

For the MDP80 calculation, the calibration surface withMDP = 0 kJ/s is scaled to MDP80 = 150, and a soft surfacewith MDP = 108.47 kJ/s (80000 lb-ft/s) is scaled to MDP80 =1. For the MDP40 calculation, the calibration surface withMDP = 0 kJ/s is scaled to MDP40 = 150 and a soft surfacewith MDP = 54.23 kJ/s (40000 lb-ft/s) is scaled to MDP40 =1.

Effective use of RICM technologies is aided by theintegration of GPS position information and an on-boardcomputer monitor (Figure 1) which displays the rollerlocation, machine measurement values (i.e., CMV or MDP),vibration amplitude and frequency, and roller speed. Thus,the technology enables a roller operator to make judgmentsregarding the condition of the compacted fill material in realtime. If real-time kinematic (RTK) GPS systems are used,those systems reportedly have position accuracies of about±10 mm in the horizontal plane and ±20 mm in the verticalplane [24].

3. Field Evaluation of CMVand MDP Technologies

Field evaluation studies beginning in about 1980 havedocumented correlations between RICM measurements andvarious traditionally used point measurements. A summaryof key findings from these different studies, types of rollersused, and materials tested is provided in Table 1. A varietyof QC/QA measurements have been used in the documentedcorrelation studies, which include:

Advances in Civil Engineering 3

(a) (b)

Figure 1: Photographs of on-board display units by Caterpillar (a) and Dynapac (b).

(a) nuclear gauge (NG), electrical soil density gauge(SDG), water balloon method, sand cone replace-ment method, radio isotope method, “undisturbed”shelby tube sampling, and drive core samples todetermine moisture content and dry unit weight,

(b) light weight deflectometer (LWD), soil stiffness gauge(SSG), static plate load test (PLT), falling weightdeflectometer (FWD), Briaud compaction (BCD),dynamic seismic pavement analyzer (D-SPA), andClegg hammer to determine stiffness or modulus,

(c) dynamic cone penetrometer (DCP), cone pene-tration testing (CPT), “undisturbed” shelby tubesampling, and rut depth measurements under heavytest rolling to determine shear strength or Californiabearing ratio (CBR).

Most of the field studies involved constructing andtesting controlled field test sections for research purposes andcorrelation development, while a few studies were conductedon full-scale earthwork construction projects where RICMwas implemented as part of the project specifications [12,13].

Based on the findings from a comprehensive correlationstudy conducted on 17 different soil types from multipleproject sites as part of the National Cooperative HighwayResearch Program (NCHRP) 21-09 project [25], the factorsthat commonly affect the correlations are as follows:

(1) heterogeneity in underlying layer support conditions,

(2) high moisture content variation,

(3) narrow range of measurements,

(4) machine operation setting variation (e.g., amplitude,frequency, speed, and roller “jumping”),

(5) nonuniform drum/soil contact conditions,

(6) uncertainty in spatial pairing of point measurementsand roller MVs,

(7) limited number of measurements,

(8) not enough information to interpret the results,

(9) intrinsic measurement errors associated with theRICM and in situ point measurements.

In general, results from controlled field studies indicatethat statistically valid simple linear or simple nonlinearcorrelations between RICM values and compaction layerpoint-MVs (e.g., modulus or density) are possible when thecompaction layer is underlain by a relatively homogenousand stiff/stable supporting layer. For example, Figure 2presents simple linear regression relationships between CMVand in situ LWD modulus and dry density point-MVsobtained from a calibration test strip with plan dimensionsof 30 m × 2 m. The test strip consisted of silty sand withgravel base material underlain by a very stiff fly ash stabilizedsubgrade layer. For this case, correlations between CMV andboth LWD modulus and dry density measurements showedR2 > 0.8.

On the contrary, many field studies summarized inTable 1 indicate that modulus- or stiffness-based measure-ments (i.e., determined by FWD, LWD, PLT, etc.) generallycorrelate better with the RICM measurements than com-paction layer dry unit weight or CBR measurements. Thisis illustrated in Figures 3 and 4. Data presented in Figure 3was obtained from several calibration and production testareas with lean clay subgrade, recycled asphalt subbase,recycled concrete base, and crushed limestone base materialscompacted with a vibratory smooth drum roller. Datapresented in Figure 4 was obtained from several calibrationand production test areas with lean clay subgrade compactedusing a nonvibratory padfoot roller. CBR measurementspresented herein are obtained from DCP tests using empir-ical correlations between DCP index values and CBR [38].Figures 3 and 4 clearly indicate that CMV correlates betterwith LWD modulus point-MVs compared to dry unitweight or CBR point-MVs. One of the primary reasons forthis is that modulus measurements represent a compositelayered soil response under an applied load which simulatesvibratory drum-ground interaction. The density and CBRmeasurements are average measurements of the compactionlayer and do not directly represent a composite layered soilresponse under loading. Although DCP-CBR measurementsdid not correlate well in the two cases presented in Figures3 and 4, many field studies [13, 25, 34] have indicated thatDCP tests are effective in detecting deeper “weak” areas(at depths > 300 mm) that are commonly identified by


Ta

ble

1:Su

mm

ary

offi

eld

corr

elat

ion

sst

udi

eson

CM

Van

dM

DP

tech

nol

ogie

sdo

cum

ente

din

the

liter

atu

re.

Ref

eren

ce;p

roje

ctlo

cati

onR

olle

rdr

um

type

;RIC

M;

soil

type

sQ

C/Q

Apo

int-

MV

sK

eyfi

ndi

ngs

and

com

men

ts

Fors

sbla

d[1

];Sw

eden

Dyn

apac

SD;C

MV

;fin

ean

dco

arse

rock

fill.

Wat

erba

lloon

,PLT

,FW

D,a

nd

surf

ace

sett

lem

ent

Lin

ear

corr

elat

ion

sar

eob

serv

edbe

twee

nC

MV

and

poin

t-M

Vs.

Moi

stu

reco

nte

nt

shou

ldbe

con

side

red

inco

rrel

atio

ns

for

fin

e-gr

ain

edso

ils.R

olle

rre

sult

sin

aco

mpo

site

valu

ein

ala

yere

dso

ilco

ndi

tion

.C

MV

isaff

ecte

dby

rolle

rsp

eed

(hig

her

spee

dsre

sult

inlo

wer

CM

V)

Han

sbo

and

Pra

mbo

rg[2

0];S

wed

en

Dyn

apac

SD;C

MV

;gr

avel

lysa

nd,

silt

ysa

nd,

and

fin

esa

nd

San

dco

ne,

pres

sure

-met

er,P

LT,

CP

T,an

dD

CP

Com

pact

ion

grow

thcu

rves

show

edim

prov

emen

tin

CM

Van

dot

her

mec

han

ical

prop

erti

es(i

.e.,

mod

ulu

san

dco

ne

resi

stan

ce)

wit

hin

crea

sin

gpa

ss.R

elat

ive

perc

ent

com

pact

ion

(or

den

sity

)w

asn

otse

nsi

tive

toch

ange

sin

CM

V

Flos

set

al.[

26];

Mu

nic

h,

Ger

man

yD

ynap

acdu

alSD

;CM

V;

san

dyto

silt

ygr

avel

fill

Wat

erba

lloon

and

san

dco

ne,

PLT

,an

dD

CP

Cor

rela

tion

sw

ith

mod

ulu

san

dD

CP

mea

sure

men

tsar

ege

ner

ally

bett

erth

ande

nsi

ty.C

MV

mea

sure

men

tsar

ede

pen

den

ton

spee

d,vi

brat

ion

freq

uen

cyan

dam

plit

ude

,soi

ltyp

e,gr

adat

ion

,wat

erco

nte

nt,

and

stre

ngt

hof

subs

oil

Bra

nd

lan

dA

dam

[6];

not

prov

ided

Bom

agSD

;CM

VP

LT

Cor

rela

tion

betw

een

CM

Van

dP

LTm

odu

lus

(in

itia

l)sh

owed

diff

eren

tre

gres

sion

tren

dsfo

rpa

rtia

lupl

ift

and

dou

ble

jum

pop

erat

ing

con

diti

ons.

Reg

ress

ion

sin

part

ialu

plif

tan

ddo

ubl

eju

mp

con

diti

ons

yiel

dedR

2=

0.9

and

0.6,

resp

ecti

vely

Noh

seet

al.[

27];

Tom

ei,

Japa

nSa

kaiS

D;C

MV

;cla

yey

grav

elR

adio

-iso

tope

Dry

den

sity

and

CM

Vin

crea

sed

wit

hin

crea

sin

gro

ller

pass

ona

calib

rati

onte

stst

rip.

Lin

ear

regr

essi

onre

lati

onsh

ips

wit

hR

2>

0.9

are

obse

rved

for

corr

elat

ion

sbe

twee

ndr

yde

nsi

tyan

dC

MV

Wh

ite

etal

.[9,

28],

Edw

ards

,Ill,

USA

Cat

erpi

llar

PD

;MD

P;l

ean

clay

NG

,Dri

veco

re,D

CP,

and

cleg

gh

amm

er

Cor

rela

tion

sbe

twee

nM

DP

and

insi

tute

stm

easu

rem

ents

usi

ng

sim

ple

and

mu

ltip

lere

gres

sion

anal

yses

are

pres

ente

d.M

DP

corr

elat

edbe

tter

wit

hdr

yde

nsi

ty(R

2=

0.86

)th

anw

ith

DC

P(R

2=

0.38

)or

Cle

ggim

pact

valu

e(R

2=

0.46

).In

clu

din

gm

oist

ure

con

ten

tvi

am

ult

iple

regr

essi

onan

alys

isim

prov

edth

eR

2va

lues

for

DC

Pan

dC

legg

impa

ctva

lue

(R2>

0.9)

.Res

ult

sar

eba

sed

onda

taav

erag

ed20

mlo

ng

stri

pp

erpa

ss

L.Pe

ters

enan

dR

.Pe

ters

on[2

9];T

H53

,D

ulu

th,M

inn

,USA

Cat

erpi

llar

SD;C

MV

and

MD

P;fi

ne

san

dLW

D,D

CP,

and

SSG

Wea

kco

rrel

atio

ns

are

obta

ined

ona

poin

t-by

-poi

nt

basi

sco

mpa

riso

nbe

twee

nin

situ

test

mea

sure

men

tsan

dro

ller

mea

sure

men

ts,l

ikel

ydu

eto

the

dept

han

dst

ress

depe

nde

ncy

ofso

ilm

odu

lus

and

the

het

erog

enei

tyof

the

soils

.Goo

dco

rrel

atio

ns

are

obta

ined

betw

een

CM

Vva

lues

and

DC

Pm

easu

rem

ents

for

dept

hs

betw

een

200

and

400

mm

dept

h

Wh

ite

etal

.[11

,30]

;E

dwar

ds,I

ll,U

SAC

ater

pilla

rSD

;MD

P;

wel

l-gr

aded

silt

ysa

nd

NG

and

DC

P

Ave

rage

MD

Pva

lues

show

eda

decr

easi

ng

tren

don

alo

gsc

ale,

and

dry

un

itw

eigh

tan

dD

CP

inde

xva

lues

show

edan

asym

ptot

icde

crea

sew

ith

incr

easi

ng

rolle

rpa

ss.C

orre

lati

ons

betw

een

MD

Pan

dp

oin

t-M

Vs

show

edgo

odco

rrel

atio

ns

(R2=

0.5

to0.

9).I

nco

rpor

atin

gm

oist

ure

con

ten

tin

toan

alys

isis

crit

ical

toim

prov

eco

rrel

atio

ns

for

dry

un

itw

eigh

t


Ta

ble

1:C

onti

nu

ed.

Ref

eren

ce;p

roje

ctlo

cati

onR

olle

rdr

um

type

;RIC

M;

soil

type

sQ

C/Q

Apo

int-

MV

sK

eyfi

ndi

ngs

and

com

men

ts

Th

omps

onan

dW

hit

e[3

1];E

dwar

ds,I

ll,U

SAC

ater

pilla

rP

D;M

DP

;silt

and

lean

clay

NG

,DC

P,C

legg

Ham

mer

,an

dLW

D

Cor

rela

tion

sbe

twee

nM

DP

and

poi

nt-

MV

sar

epr

esen

ted

usi

ng

sim

ple

and

mu

ltip

lere

gres

sion

anal

ysis

.Ave

ragi

ng

the

data

alon

gth

efu

llle

ngt

hof

the

test

stri

p(p

erpa

ss)

impr

oved

the

regr

essi

ons.

Mu

ltip

lere

gres

sion

anal

ysis

byin

corp

orat

ing

moi

stu

reco

nte

nt

asa

regr

essi

onpa

ram

eter

furt

her

impr

oved

the

corr

elat

ion

s

Wh

ite

etal

.[12

];T

H64

,A

ckle

y,M

inn

,USA

Cat

erpi

llar

SD;C

MV

;po

orly

grad

edsa

nd

and

wel

l-gr

aded

san

dw

ith

silt

LWD

,DC

P,an

dN

G

Pro

ject

scal

eco

rrel

atio

ns

byav

erag

ing

data

from

diff

eren

tar

eas

onth

epr

ojec

tar

epr

esen

ted,

wh

ich

show

edR

2va

lues

ran

gin

gfr

om0.

52fo

rde

nsi

tyan

d0.

79fo

rD

CP

inde

xva

lue.

Cor

rela

tion

sw

ith

LWD

show

edpo

orco

rrel

atio

ns

due

toth

eeff

ect

oflo

ose

mat

eria

lat

the

surf

ace.

Th

eva

riab

ility

obse

rved

inth

eC

MV

data

was

sim

ilar

toD

CP

and

LWD

mea

sure

men

tsbu

tn

otto

den

sity

mea

sure

men

ts

Wh

ite

etal

.[32

],E

dwar

ds,I

llin

ois,

USA

Cat

erpi

llar

PD

;MD

P;

san

dyle

ancl

ayN

Gan

dD

CP

Bas

edon

aver

age

mea

sure

men

tsov

erth

ele

ngt

hof

the

test

stri

p( ∼

20m

);co

rrel

atio

ns

betw

een

MD

Pan

dpo

int-

MV

ssh

owed

R2=

0.87

for

den

sity

and

0.96

for

DC

Pin

dex

valu

es

Ven

nap

usa

etal

.[33

],E

dwar

ds,I

ll,U

SAC

ater

pilla

rP

D;M

DP

;cr

ush

edgr

avel

base

DC

Pan

dLW

D

Cor

rela

tion

sw

ere

obta

ined

ona

test

bed

wit

hm

ult

iple

lifts

plac

edon

aco

ncr

ete

base

and

aso

ftsu

bgra

deba

se.C

orre

lati

ons

betw

een

MD

Pan

dpo

int-

MV

syi

elde

dR

2=

0.66

to0.

85fo

rsp

atia

llyn

eare

stp

oin

tda

ta,a

nd

R2=

0.74

to0.

92fo

rav

erag

edda

ta(o

ver

the

len

gth

ofco

ncr

ete

orso

ftsu

bgra

de)

Wh

ite

etal

.[13

,34]

,T

H60

,Big

elow

,Min

n,

USA

Cat

erpi

llar

PD

-MD

P80

and

SD-C

MV

;san

dyle

ancl

ayto

lean

clay

wit

hsa

nd

Hea

vyte

stro

ller,

DC

P,LW

D,a

nd

PLT

Cor

rela

tion

sar

epr

esen

ted

from

mu

ltip

leca

libra

tion

test

stri

psan

dpr

odu

ctio

nar

eas

from

the

proj

ect.

MD

P80

and

LWD

mod

ulu

sco

rrel

atio

nsh

owed

two

diff

eren

ttr

ends

(R2=

0.35

and

0.65

)ov

erth

era

nge

ofm

easu

rem

ents

asth

eM

DP

80re

ach

edan

asym

ptot

icva

lue

ofab

out

150

wh

ich

isth

em

axim

um

valu

eon

the

calib

rati

onh

ard

surf

ace.

CM

Vco

rrel

atio

nw

ith

LWD

mod

ulu

spr

odu

cedR

2=

0.70

,an

dw

ith

rut

dept

hpr

odu

cedR

2=

0.64

Wh

ite

etal

.[13

],T

H36

,N

orth

St.P

aul,

Min

n,

USA

Cat

erpi

llar

SD;C

MV

;gr

anu

lar

subb

ase

and

sele

ctgr

anu

lar

base

DC

P,SS

G,C

legg

Ham

mer

,LW

D,

PLT

,FW

D,a

nd

CP

T

Cor

rela

tion

sbe

twee

nC

MV

and

poi

nt-

MV

sfr

omca

libra

tion

and

prod

uct

ion

test

area

sba

sed

onsp

atia

llyn

eare

stp

oin

tda

taar

epr

esen

ted.

Posi

tive

tren

dsar

ege

ner

ally

obse

rved

wit

hR

2>

0.5

(for

LWD

,FW

D,

PLT

,SSG

,an

dC

legg

)w

ith

exce

ptio

nof

one

test

bed

(FW

D,L

WD

,an

dC

PT

)w

ith

limit

ed/n

arro

wra

nge

ofm

easu

rem

ents


Ta

ble

1:C

onti

nu

ed.

Ref

eren

ce;p

roje

ctlo

cati

onR

olle

rdr

um

type

;RIC

M;

soil

type

sQ

C/Q

Apo

int-

MV

sK

eyfi

ndi

ngs

and

com

men

ts

Wh

ite

etal

.[13

],U

S10,

Stap

les,

Min

n,U

SA

Cat

erpi

llar

SD;C

MV

;po

orly

grad

edsa

nd

wit

hsi

ltto

silt

ysa

nd

LWD

,PLT

,an

dD

CP

Cor

rela

tion

sbe

twee

nC

MV

and

poi

nt-

MV

sfr

omca

libra

tion

and

prod

uct

ion

test

area

sba

sed

onsp

atia

llyn

eare

stp

oin

tda

taar

epr

esen

ted.

Cor

rela

tion

sbe

twee

nC

MV

and

poi

nt-

MV

ssh

owed

R2

valu

era

ngi

ng

from

0.2

to0.

9.T

he

prim

ary

fact

ors

con

trib

uti

ng

tosc

atte

rar

eat

trib

ute

dto

diff

eren

ces

inm

easu

rem

ent

infl

uen

cede

pth

s,ap

plie

dst

ress

es,a

nd

the

loos

esu

rfac

eof

the

san

dyso

ilson

the

proj

ect.

Cor

rela

tion

sbe

twee

nC

MV

and

LWD

orD

CP

mea

sure

men

tsim

prov

edu

sin

gm

easu

rem

ents

atab

out

150-

mm

belo

wth

eco

mpa

ctio

nsu

rfac

e

Wh

ite

etal

.[13

],C

SAH

2,O

lmst

edC

oun

ty,

Min

n,U

SA

Cat

erpi

llar

PD

;MD

P80

;sa

ndy

lean

clay

LWD

MD

P80

valu

esar

ein

flu

ence

dby

the

trav

eldi

rect

ion

ofth

ero

ller

due

tolo

caliz

edsl

ope

chan

ges

and

rolle

rsp

eed.

Cor

rela

tion

sbe

twee

nM

DP

80

and

LWD

gen

eral

lysh

owed

R2>

0.6

(wit

hex

cept

ion

ofon

eca

se)

wh

enre

gres

sion

sar

epe

rfor

med

byse

para

tin

gda

tase

tsw

ith

diff

eren

ttr

avel

dire

ctio

ns

and

spee

d.D

ata

was

com

bin

edby

per

form

ing

mu

ltip

lere

gres

sion

anal

ysis

inco

rpor

atin

gtr

avel

spee

dan

ddi

rect

ion

wh

ich

show

edco

rrel

atio

ns

wit

hR

2=

0.93

Moo

ney

etal

.[25

];M

inn

esot

a,C

olor

ado,

Mar

ylan

d,N

orth

Car

olin

a,Fl

a,U

SA

Cat

erpi

llar

PD

-MD

Pan

dSD

-CM

V,D

ynap

acSD

-CM

V;t

wo

typ

esof

coh

esiv

eso

ils,e

leve

nty

pes

ofgr

anu

lar

soils

NG

,DC

P,LW

D,F

WD

,PLT

,C

legg

ham

mer

,SSG

Sim

ple

and

mu

ltip

lere

gres

sion

anal

ysis

resu

lts

are

pres

ente

d.Si

mpl

elin

ear

corr

elat

ion

sbe

twee

nR

ICM

and

poin

t-M

Vs

are

pos

sibl

efo

ra

com

pact

ion

laye

ru

nde

rlai

nby

rela

tive

lyh

omog

enou

san

da

stiff

/sta

ble

supp

orti

ng

laye

r.H

eter

ogen

eou

su

nde

rlyi

ng

con

diti

ons

can

adve

rsel

yaff

ect

the

corr

elat

ion

s.A

mu

ltip

lere

gres

sion

anal

ysis

appr

oach

isde

scri

bed

that

incl

ude

spa

ram

eter

valu

esto

repr

esen

tu

nde

rlyi

ng

laye

rco

ndi

tion

sto

impr

ove

corr

elat

ion

s.M

odu

lus

mea

sure

men

tsge

ner

ally

capt

ure

the

vari

atio

nin

RIC

Mva

lues

bett

erth

andr

yu

nit

wei

ght

mea

sure

men

ts.D

CP

test

sar

eeff

ecti

vein

dete

ctin

gde

eper

“wea

k”ar

eas

that

are

com

mon

lyid

enti

fied

byR

ICM

valu

esan

dn

otby

com

pact

ion

laye

rpo

int-

MV

s.H

igh

vari

abili

tyin

soil

prop

erti

esac

ross

the

dru

mw

idth

and

soil

moi

stu

reco

nte

nt

con

trib

ute

tosc

atte

rin

rela

tion

ship

s.A

vera

gin

gm

easu

rem

ents

acro

ssth

edr

um

wid

th,a

nd

inco

rpor

atin

gm

oist

ure

con

ten

tin

tom

ult

iple

regr

essi

onan

alys

is,c

anh

elp

mit

igat

eth

esc

atte

rto

som

eex

ten

t.R

elat

ivel

yco

nst

ant

mac

hin

eop

erat

ion

sett

ings

(i.e

.,am

plit

ude

,fre

quen

cy,a

nd

spee

d)ar

ecr

itic

alfo

rca

libra

tion

stri

psan

dco

rrel

atio

ns

are

gen

eral

lybe

tter

for

low

ampl

itu

dese

ttin

gs(e

.g.,

0.7

to1.

1m

m)


Ta

ble

1:C

onti

nu

ed.

Ref

eren

ce;p

roje

ctlo

cati

onR

olle

rdr

um

type

;RIC

M;

soil

type

sQ

C/Q

Apo

int-

MV

sK

eyfi

ndi

ngs

and

com

men

ts

Wh

ite

etal

.[35

];FM

156,

Roa

nok

e,Te

x,U

SA

Dyn

apad

SD;C

MV

;gr

anu

lar

base

and

lime

stab

ilize

dsu

bgra

deN

G,L

WD

,PLT

,FW

D,D

-SPA

CM

Vm

easu

rem

ents

show

edgo

odre

peat

abili

tybu

tar

ein

flu

ence

dby

vibr

atio

nam

plit

ude

.Hig

ham

plit

ude

(i.e

.,>

1.5

mm

)ca

use

ddr

um

bou

nci

ng

and

affec

ted

the

CM

Vm

easu

rem

ents

.In

crea

sin

gam

plit

ude

gen

eral

lysh

owed

anin

crea

sein

CM

V.R

esu

lts

show

edth

atFW

Dm

odu

lus

poin

tm

easu

rem

ents

trac

ked

wel

lwit

hva

riat

ion

sin

CM

Vin

som

eca

ses

and

inso

me

case

sit

did

not

.Th

ere

ason

for

poo

rco

rrel

atio

ns

wit

hFW

Dm

easu

rem

ents

inso

me

case

sis

attr

ibu

ted

toth

epo

ssib

lein

flu

ence

ofh

eter

ogen

eity

obse

rved

inth

em

ater

iala

cros

sth

edr

um

wid

thdu

eto

moi

stu

rese

greg

atio

n.T

he

CM

Vm

easu

rem

ents

how

ever

wer

ew

ellc

orre

late

dw

ith

vari

atio

ns

inm

oist

ure

con

ten

tas

evid

ence

dby

ade

crea

sein

CM

Vw

ith

incr

easi

ng

moi

stu

reco

nte

nt.

D-S

PA,P

LT,a

nd

DC

Pm

easu

rem

ents

trac

ked

wel

lwit

hth

eva

riat

ion

sin

CM

V

Wh

ite

etal

.[36

];U

S219

,Sp

rin

gvill

e,N

Y,U

SAC

ater

pilla

rSD

;CM

Van

dM

DP

40;w

ell-

grad

edgr

avel

DC

P,LW

D,F

WD

,PLT

,BC

D,

NG

,an

dSD

G

Non

linea

rpo

wer

,exp

onen

tial

,an

dlo

gari

thm

icre

lati

onsh

ips

are

obse

rved

betw

een

RIC

Man

dpo

int-

MV

s.C

orre

lati

ons

betw

een

RIC

Mva

lues

and

diff

eren

tp

oin

t-M

Vs

are

gen

eral

lyw

eak

wh

enev

alu

ated

inde

pen

den

tly

for

each

test

bed

due

ton

arro

wra

nge

ofm

easu

rem

ents

.W

hen

data

are

com

bin

edfo

rsi

tew

ide

corr

elat

ion

sw

ith

aw

ide

mea

sure

men

tra

nge

,th

eco

rrel

atio

ns

impr

oved

.RIC

Mva

lues

gen

eral

lyco

rrel

ated

bett

erw

ith

mod

ulu

s/st

iffn

ess

and

CB

Rpo

int-

MV

sth

anw

ith

dry

den

sity

poin

t-M

Vs.

Cor

rela

tion

sbe

twee

nR

ICM

valu

esan

dFW

Dm

easu

rem

ents

show

edth

est

ron

gest

corr

elat

ion

coeffi

cien

ts

Wh

ite

etal

.[37

];U

S84,

Way

nes

boro

,Mis

s,U

SAC

ater

pilla

rP

D;M

DP

40;

poor

lygr

aded

tosi

lty

san

dD

CP,

LWD

,FW

D,N

G,a

nd

PLT

Not

es:S

D:s

moo

thdr

um

,PD

:pad

foot

dru

m.


40 60 80 100 120 1400

20

40

60

80C

MV

CMV = 1.2 (LWD modulus) − 70.4n = 30R2 = 0.83

300 mm dia. plate LWD modulus (MPa)

(a)

0

20

40

60

80

CM

V

Dry unit weight (kN/m3)

15.6 16.2 16.8 17.4 18

n = 30R2 = 0.81

CMV = 31.55 (dry unit weight) − 490.8

(b)

Figure 2: Simple linear regressions between CMV (amplitude = 1.00 mm) and in situ point measurements (LWD modulus and dry unitweight)—silty sand with gravel underlain by fly ash stabilized subgrade.

RICM measurements and not by point-MVs obtained onthe surface. This is primarily because of the differences inmeasurement influence depths. Accelerometer based rollermeasurements have measurement influence depths rangingfrom 0.8 m to 1.5 m depending on soil layering, drum mass,and excitation force [25, 39–41], while machine drive powerbased measurements range from 0.3 to 1.3 m depending onthe heterogeneity in subsurface conditions [33]. On the otherhand, most point-MVs have influence depths


0 10 20 30 40 50 60

CM

V

0

5

10

15

20

25

30CMV = 1.62 exp0.05 (LWD modulus)R2 = 0.86n = 104


(a)

CM

V

0

5

10

15

20

25

30


10 12 14 16 18 20 22

R2 = 0n = 85

(b)

CM

V

0

5

10

15

20

25

30

R2 = 0.18

0 5 10 15 20 25 30

Lean clay subgrade

Recycled asphalt subbase

Recycled concrete base

Crushed limestone subbase

CMV = 0.4 CBR + 4.5

n = 50

DCP-CBR of compaction layer (%)

(c)

Figure 3: Relationships between CMV (theoretical amplitude = 1.50 mm) and in situ point measurements (LWD modulus, dry unit weight,and CBR determined from DCP).

construction where the backfill may not be as compactas the neighboring unexcavated materials. After subgradepreparation, the area was mapped using a smooth drummachine in seven lanes using a vibration amplitude = 2.1 mmand frequency = 29 Hz. DCP tests were performed at 144locations in the upper 200 mm (shown in gray circles onFigure 6) following roller mapping passes. The DCP testlocations were strategically spaced such that the boundariesof compacted and uncompacted areas were captured duringKriging interpolation.

CMV and MDP spatial data along with experimentaland theoretical variogram models are shown in Figure 6.Log transformation of CMV was required to detrend theexperimental variogram (details on detrending is explained

in detail in Vennapusa et al. [22]). Kriged surface map andvariogram model generated for the DCP index values are alsopresented in Figure 6. The univariate statistics (mean (μ),standard deviation (σ), and coefficient of variation (COV))of the measurement values are also provided on the figurefor reference. The compacted and uncompacted areas weregenerally well captured by the CMV/MDP and DCP indexmeasurements; however, they were more clearly delineatedby the DCP Index measurements.

Univariate statistics show that the COV of the MDP(89%) and DCP index (86%) measurements are comparableand are also significantly higher compared to that of CMV(39%). Similarly, the exponential variogram models of MDPand DCP index exhibit significantly lower range values than


MD

P40

60

80

100

120

140

160

0 2 4 6 8 10 12 14

MDP40 = 68.3 (LWD modulus)0.25R2 = 0.63n = 76


(a)

MD

P40

60

80

100

120

140

160


13 14 15 16 17 18 19 20

R2 = 0n = 76

(b)

MD

P40

60

80

100

120

140

160

Moisture content, w (%)

12 14 16 18 20 22 24 26 28 30

R2 = 0.15n = 76

MDP40 = −1.8 (w) + 134.6

(c)

MD

P40

60

80

100

120

140

160


0 1 2 3 4 5 6 7

R2 = 0.51n = 76

MDP40 = 100.7 (CBR)0.08

(d)

Figure 4: Example 3: Relationships between MDP40 (obtained in static mode) and in situ point measurements (LWD modulus, dry unitweight, moisture content, and CBR determined from DCP)—Sandy clay to silty clay subgrade.

CMV. This suggests that the MDP and DCP index mea-surements have less spatial continuity and higher variabilitycompared to CMV measurements. This high variability islikely due to differences in the measurement influence depthsof these measurements as discussed earlier in this paper. DCPindex values presented in Figure 6 are based on average valuesin the top 300 mm of the surface.

In addition to using spatial analysis for visualizationpurposes, analysis from some field studies by the authors [36]has indicated that experimental semivariograms of RICMvalues sometimes show nested structures with distinctivelydifferent long- and short-range components. The nestedstructures are very likely linked to spatial variation in theunderlying layer support conditions. These observationsare new, have not been fully evaluated, and warrant moreresearch. Further, a field study by White et al. [35] reported

that variograms developed for two different spatial areaswith similar univariate statistics (i.e., mean and standarddeviation) showed distinctly different shapes of variogramswith different spatial statistics, which illustrate the impor-tance of spatial modeling to obtain better characterizationof “nonuniformity” compared to using univariate statistics.This emphasizes the importance of dealing with “nonuni-formity” in a spatial perspective rather than in a univariatestatistics perspective.

5. Implementation of RICM Technology

A few countries and governmental agencies have devel-oped specifications to facilitate implementation of RICMtechnologies into earthwork and hot mix asphalt (HMA)construction practices [44]. The international society of soil


MD

P40

60

80

100

120

140

160

0 2 4 6 8 10 12 14

ELWD-Z3 (MPa)

R2 = 0.71n = 76

MDP40 = 82.3 (LWD modulus)0.25 − 0.8 (w)

(a)

MD

P40

60

80

100

120

140

160

R2 = 0.51


0 1 2 3 4 5 6 7

n = 76

MDP40 = 103.8 (CBR)0.08 − 0.2 (w)

(b)

Figure 5: Example 4: Multivariate non-linear regression analysis between MDP40 and in situ point measurements (LWD modulus and CBRdetermined from DCP) using moisture content—Sandy clay to silty clay subgrade.

mechanics and geotechnical engineering (ISSMGE) [39],Minnesota department of transportation (DOT) [45, 46],Austrian [47], German [48], and Sweden [49] specificationsrequire performing either static or dynamic plate load testson calibration strips to determine average target values(typically based on 3 to 5 measurements) and use the samefor QA in production areas. The ISSMGE, Austrian, andGerman specifications suggest performing at least three staticplate load tests in locations of low, medium, and highdegree of compaction during calibration process. Further,it is specified that linear regression relationships betweenroller compaction measurement values and plate load testresults should achieve a regression coefficient, R ≥ 0.7.Although it is not clear yet what the right number of testmeasurements is to develop a field calibration, the experienceof the authors shows that by increasing the number ofmeasurements to 10–15 points, this substantially increasesthe statistical significance of the predictions.

One of the major limitations of the existing RICMspecifications is that the acceptance requirements (i.e.,percent target value limits, acceptable variability, etc.) aretechnology specific and somewhat based on local experience.This limitation hinders widespread acceptance of thesespecifications into practice as there are currently at leastten different RICM technologies. Significant efforts arebeing made in the US in developing widely acceptableand technology-independent specifications [3–5]. Based onfeedback obtained from various state and federal agencypersonnel in recent national level workshops conductedon this topic, White and Vennapusa [4] documented thefollowing as the key attributes of RICM specification:

(1) descriptions of the rollers and configurations,

(2) guidelines for operations (speed, vibration frequencyand amplitude, and track overlap),

(3) records to be reported (time stamp, operations/mode, soil type, moisture content, layer thickness,etc.),

(4) repeatability and reproducibility measurements forRICM values,

(5) ground conditions (smoothness, levelness, isolatedsoft/wet spots),

(6) calibration procedures for rollers and selection ofcalibration areas,

(7) regression analysis between RICM values and pointmeasurements,

(8) number and location of quality control (QC) andquality assurance (QA) tests,

(9) operator training/certification, and

(10) acceptance procedures/corrective actions based onachievement of minimum RICM target values andassociated variability.

Several new and innovative specification concepts havebeen proposed by researchers in the past few years [4, 13,25]. These concepts primarily vary in the way the itemnumber 10 listed above is dealt in the specification, thatis, in the required level of upfront calibration work anddata analysis which consequently leads to differences inthe level of confidence in the quality of the completedwork. A few of those concepts have been beta tested ondemonstration level projects [25] but have never been fullyevaluated on full scale projects to explore their limitationsand advantages. More coordination between researchersand practitioners is needed to carefully evaluate theseconcepts and is a high-priority step forward to successfullyimplement the RICM technologies in practice. Further,integrating advanced analytical methods discussed in this


10203040

Transverse length (m)

0 4 8 12 16

Lon

gitu

din

alle

ngt

h(m

)

4

8

12

16

20

24

28

32

CMV

μ = 20.4σ = 7.9COV = 39%

(a)

Transverse length (m)0 4 8 12 16

4

8

12

16

20

24

28

32

−50510

μ = 3.7σ = 3.3COV = 89%

MDP(kJ/s)

(b)

Transverse length (m)

0 4 8 12 16

4

8

12

16

20

24

28

32

Scarifiedzone

Compactedzone

Spottests

20

406080

DCP index(mm/blow)

μ = 43σ = 37COV = 86%

(c)

Lag distance (m)

Var

iogr

am(C

MV

)2

0

0.02

0.04

0.06

0.08Parameter log CMV

0 5 10 15 20

Experimental

Exponential

Nugget—0Sill—0.038Range—2.8Cressie “goodness”—0.02

(d)

Lag distance (m)

0 5 10 15 20

Var

iogr

am(k

J/s)

2

Experimental

Exponential

0

4

8

12

16

20Parameter MDP

Nugget—0Sill—11.2Range—0.8Cressie “goodness”—0.002

(e)

Lag distance (m)

0 5 10 15 20

Var

iogr

am(m

m/b

low

)2

Parameter DCP index

0

500

1000

1500

2000

2500

Experimental

Exponential

Nugget—450

Sill—1000Range—1.2Cressie “goodness”—0.006

(f)

Figure 6: Spatial comparison of dynamic cone penetration (DCP) index and roller-integrated CMV and MDP measurements on compactedand scarified Edwards glacial till subgrade soil.

paper (such as simple and multiple regression analysis todevelop correlations and target values and spatial analysisto address nonuniformity) into a specification along withdevelopment of simple and ready-to-use software tools ismuch necessary to help advance the technology.

6. Concluding Remarks

RICM technologies with real-time display capabilities and100% coverage of compaction data offer significant advan-tages to the earthwork construction industry. Integration ofthese technologies into practice requires proper understand-ing of the correlations between RICM values and compactionmeasurements (e.g., density, modulus) and factors that influ-ence these correlations. This paper provided an overviewof two technologies (i.e., CMV and MDP) and a review of

field correlation studies documented in the literature. Reviewof the literature revealed correlations between the RICMmeasurements and various in situ tests used to measuredensity, modulus or stiffness, shear strength, and CBR.Results from field studies indicated that simple linear ornonlinear correlations between any of these measurementsare possible if the compaction layer is placed over a “stable”and “homogenous” underlying layer. If the underlying layeris not stable or homogenous, correlations are adverselyaffected. For those cases, in general, relationships betweenRICM and modulus based measurements (e.g., LWD orFWD or PLT modulus) are better compared to RICM and drydensity or CBR measurements. Multiple regression analysiscan be performed incorporating properties of the underlyinglayers and moisture content to improve correlations. Otherimportant factors that affect the correlations include narrow


range of measurements, variations in machine settings,nonuniform conditions across the drum width, limited num-ber of measurements, and measurement error associatedwith both RICM and in situ point measurements.

Geostatistical spatial modeling techniques can be utilizedfor data visualization and characterization of spatial nonuni-formity using spatially referenced RICM data. To demon-strate the application of spatial analysis for visualization,an example CMV and MDP data set over a spatial areawas used in comparison with DCP index measurements.Analysis results indicated that the compacted and uncom-pacted areas over the spatial area were well captured byall the measurements. Some field studies documented inthe literature indicate that geostatistical semivariograms ofRICM measurements can be used for construction processcontrol and also to analyze variations in the underlyingsupport conditions. These observations are new, have notbeen fully evaluated, and warrant more research.

Review of current RICM specifications revealed a poten-tial limitation with the acceptance requirements (i.e., percenttarget value limits, acceptable variability, etc.) being technol-ogy specific and somewhat based on local experience. Thislimitation hinders widespread acceptance of these specifica-tions into practice as there are currently at least ten differentRICM technologies. Several new specification concepts havebeen recently documented in the literature with variationsin the way calibration work is performed and acceptancerequirements are established. These concepts require detailedfield evaluation to explore their limitations and advantages.Integration of advanced analytical methods discussed in thispaper (such as simple and multiple regression analysis todevelop correlations and target values, and spatial analysisto address nonuniformity) into a specification along withdevelopment of simple and ready-to-use software tools aremuch necessary to help advance the technology.

Acknowledgments

The results presented in this paper are from several researchprojects sponsored by the Highway Division of the IowaDepartment of Transportation (DOT), Minnesota DOT,Federal Highway Administration, NCHRP, Caterpillar, Inc.,and Iowa State University. The findings and opinionsexpressed in this paper are those of the authors and do notnecessarily reflect the views of the sponsor and administra-tors. Several research associates, graduate, and undergradu-ate students at Iowa State University provided assistance withfield testing.

References

[1] L. Forssblad, “Compaction meter on vibrating rollers forimproved compaction control,” in Proceedings of the Interna-tional Conference on Compaction, vol. 2, pp. 541–546, Paris,France, 1980.

[2] H. Thurner and A. Sandström, “A new device for instant com-paction control,” in Proceedings of the International Conferenceon Compaction, vol. 2, pp. 611–614, Paris, France, 1980.

[3] D. J. White, “Workshop on intelligent compaction for soils andHMA,” Report ER08-01, Iowa State University, Ames, Iowa,USA, 2008.

[4] D. J. White and P. Vennapusa, “Report of the workshopon intelligent technologies for earthworks,” in Proceedingsof the 2nd Annual Workshop, no. ER09-02, Iowa DOT andthe Earthworks Engineering Research Center at Iowa StateUniversity, Sheraton West Des Moines Hotel, West DesMoines, Iowa, USA, April 2009.

[5] D. J. White and P. Vennapusa, “Report of the webinarworkshop on intelligent compaction for earthworks andHMA,” in Proceedings of the 3rd Annual Workshop, no. ER10-02, Earthworks Engineering Research Center at Iowa StateUniversity and the Iowa DOT, Ames, Iowa, USA, March 2010.

[6] H. Brandl and D. Adam, “Sophisticated continuous com-paction control of soils and granular materials,” in Proceedingsof the 14th International Conference on Soil Mechanics & Foun-dation Engineering, vol. 1, Hamburg, Germany, September1997.

[7] M. A. Mooney and D. Adam, “Vibratory roller integrated mea-surement of earthwork compaction: an overview,” in Proceed-ings of the 7th International Symposium on Field Measurementsin Geomechanics (FMGM ’07), ASCE, Boston, Mass, USA,2007.

[8] A. Sandström and C. Pettersson, “Intelligent systems for QA/QC in soil compaction,” in Proceedings of the Annual Trans-portation Research Board Meeting, Washington, DC, USA,January 2004.

[9] D. J. White, E. J. Jaselskis, V. R. Schaefer, and E. T. Cackler,“Real-time compaction monitoring in cohesive soils frommachine response,” Transportation Research Record, no. 1936,pp. 173–180, 2005.

[10] R. Sullivan, R. Anderson, Biesiadecki, T. Bond, and H. Stewart,“Martian regolith cohesions and angles of internal frictionfrom analysis of MER wheel trenches,” in Proceedings of the38th Lunar and Planetary Science Conference, League City, Tex,USA, March 2007.

[11] D. J. White, M. Morris, and M. Thompson, “Power-basedcompaction monitoring using vibratory padfoot,” in Proceed-ings of the Geotechnical Engineering in the Information Tech-nology Age (GeoCongress ’06), p. 44, Atlanta, Ga, USA, March2006, CD-ROM.

[12] D. J. White, M. Thompson, P. Vennapusa, and J. Siekmier,“Implementing intelligent compaction specifications on Min-nesota TH 64: synopsis of measurement values, data manage-ment, and geostatistical analysis,” in Proceedings of the AnnualTransportation Research Board Meeting, Washington, DC,USA, January 2008.

[13] D. J. White, P. Vennapusa, J. Zhang, H. Gieselman, and M.Morris, “Implementation of intelligent compaction perfor-mance based specifications in Minnesota,” Report ER09-03,MN/RC 2009-14, Minnesota Department of Transportation,St. Paul, Minn, USA, 2009.

[14] Iowa DOT, “Special provisions for intelligent compaction—HMA SP-090048(New),” Harrison County NHSN-030-1(127)—2R-43, Iowa DOT, Ames, Iowa, USA, 2010.

[15] Iowa DOT, “Special provisions for intelligent compaction—HMA SP-090057a,” Ida County NHSN-020-2(70)—2R-47,Iowa DOT, Ames, Iowa, USA, 2010.

[16] Iowa DOT, “Special provisions for intelligent compaction—HMA roller pass mapping SP-090058 (New),” Kossuth CountySTPN-009-4(44)—2J-55, Iowa DOT, Ames, Iowa, USA, 2010.


[17] Iowa DOT, “Special provisions for intelligent compaction—Embankment SP-090063 (New),” Sac County NHSX-020-2(89)—3H-81, Iowa DOT, Ames, Iowa, USA, 2010.

[18] J. Siekmeier and M. Oman, “Intelligent compaction to deter-mine highway subcut depth during construction. Presented inthe workshop on reinventing geotechnical investigations,” inProceedings of the 90th Annual Transportation Research BoardMeeting, Washington, DC, USA, January 2011.

[19] Å. Sandström, “Numerical simulation of a vibratory roller oncohesionless soil,” Internal Report, Geodynamik, Stockholm,Sweden, 1994.

[20] S. Hansbo and B. Pramborg, “Compaction control,” in Pro-ceedings of the International conference on Compaction, vol. 2,pp. 559–564, Paris, France, 1980.

[21] Geodynamik ALFA-030. Compactometer, Compaction Meterfor Vibratory Rollers, ALFA-030-051E/0203, Geodynamik AB,Stockholm, Sweden.

[22] P. K. R. Vennapusa, D. J. White, and M. D. Morris, “Geosta-tistical analysis for spatially referenced roller-integrated com-paction measurements,” Journal of Geotechnical and Geoenvi-ronmental Engineering, vol. 136, no. 6, pp. 813–822, 2010.

[23] M. Bekker, Introduction to Terrain-Vehicle Systems, The Uni-versity of Michigan Press, Ann Arbor, Mich, USA, 1969.

[24] F. Peyret, D. Bétaille, and G. Hintzy, “High-precision applica-tion of GPS in the field of real-time equipment positioning,”Automation in Construction, vol. 9, no. 3, pp. 299–314, 2000.

[25] M. Mooney, R. Rinehart, D. J. White, P. Vennapusa, N.Facas, and O. Musimbi, “Intelligent soil compaction systems,”Final Report NCHRP 21-09, National Cooperative HighwayResearch Program, Washington, DC, USA, 2010.

[26] R. Floss, N. Gruber, and J. Obermayer, “A dynamical testmethod for continuous compaction control,” in Proceedings ofthe 8th European Conference on Soil Mechanics and FoundationEngineering, H. G. Rathmayer and K. H. O. Saari, Eds., pp. 25–30, Helsinki, Finland, May 1983.

[27] Y. Nohse, K. Uchiyama, Y. Kanamori et al., “An attemptapplying a new control system for the vibratory compactionusing GPS and CMV in the embankment construction (Part1),” in Proceedings of the 13th International Conference of theISTVS, pp. 295–300, Okinowa, Japan, 1999.

[28] D. J. White, E. Jaselskis, V. Schaefer, T. Cackler, I. Drew, and L.Li, “Field evaluation of compaction monitoring technology:phase I,” Final Report, Iowa DOT Project TR-495, Iowa StateUniversity, Ames, Iowa, USA, 2004.

[29] L. Petersen and R. Peterson, “Intelligent compaction and In-Situ testing at Mn/DOT TH53,” Final Report MN/RC-2006-13, Minnesota DOT, St. Paul, Minn, USA, 2006.

[30] D. J. White, M. Thompson, K. Jovaag et al., “Field evaluationof compaction monitoring technology: phase II,” Final Report,Iowa DOT Project TR-495, Ames, Iowa, USA, 2006.

[31] M. J. Thompson and D. J. White, “Estimating compaction ofcohesive soils from machine drive power,” Journal of Geotech-nical and Geoenvironmental Engineering, vol. 134, no. 12, pp.1771–1777, 2008.

[32] D. J. White, M. Thompson, and P. Vennapusa, Field Studyof Compaction Monitoring Systems—Tamping Foot 825 andVibratory Smooth Drum CS-533E Rollers, Center of Trans-portation Research and Education, Iowa State University,Ames, Iowa, USA, 2007.

[33] P. Vennapusa, D. J. White, and H. Gieselman, “Influence ofsupport conditions on roller-integrated machine drive power

measurements for granular base,” in Proceedings of the Interna-tional Foundation Congress and Equipment Expo (IFCEE ’09),Orlando, Fla, USA, March 2009.

[34] D. J. White, P. Vennapusa, H. Gieselman, L. Johanson, andJ. Siekmeier, “Alternatives to heavy test rolling for cohesivesubgrade assessment,” in Proceedings of the 8th InternationalConference on the Bearing Capacity of Roads, Railways, andAirfields (BCR2A ’09), Champaign, Ill, USA, June 2009.

[35] D. J. White, P. Vennapusa, H. Gieselman, L. Johanson, and R.Goldsmith, “Accelerated implementation of intelligent com-paction monitoring technology for embankment subgradesoils, aggregate base, and asphalt pavement materials TPF-5(128)—Texas IC demonstration field project,” Report, TheTranstec Group, FHWA, 2008.

[36] D. J. White, P. Vennapusa, H. Gieselman et al., “Acceler-ated implementation of intelligent compaction monitoringtechnology for embankment subgrade soils, aggregate base,and asphalt pavement materials TPF-5(128)—New York ICdemonstration field project,” Report ER10-01, The TranstecGroup, FHWA, 2010.

[37] D. J. White, P. Vennapusa, H. Gieselman, B. Fleming, S. Quist,and L. Johanson, “Accelerated implementation of intelligentcompaction monitoring technology for embankment sub-grade soils, aggregate base, and asphalt pavement materialsTPF-5(128)—Mississippi IC demonstration field project,”Report ER10-03, The Transtec Group, FHWA, 2010.

[38] ASTM D6951-03, Standard Test Method for Use of the DynamicCone Penetrometer in Shallow Pavement Applications, Ameri-can Standards for Testing Methods, West Conshohocken, Pa,USA.

[39] ISSMGE, Roller-Integrated Continuous Compaction Control(CCC): Technical Contractual Provisions, Recommendations,TC3: Geotechnics for Pavements in Transportation Infrastruc-ture, International Society for Soil Mechanics and Geotechni-cal Engineering, 2005.

[40] R. V. Rinehart and M. A. Mooney, “Measurement depth ofvibratory roller-measured soil stiffness,” Géotechnique, vol. 59,no. 7, pp. 609–619, 2009.

[41] P. Vennapusa, D. J. White, J. Siekmeier, and R. Embacher, “Insitu mechanistic characterizationsof granular pavement foun-dation layers,” International Journal of Pavement Engineering.In press.

[42] R. M. Srivastava, “Describing spatial variability using geosta-tistical analysis,” in Geostatistics for Environmental and Geo-technical Applications, R. M. Srivastava, S. Rouhani, M. V.Cromer, A. J. Desbarats, and A. I. Johnson, Eds., vol. 1283,ASTM, West Conshohocken, Pa, USA, 1996.

[43] I. Clark and W. Harper, Practical Geostatistics 2000, EcosseNorth America Llc, Columbus, Ohio, USA, 3rd edition, 2002.

[44] D. J. White, M. Thompson, and P. Vennapusa, “Field vali-dation of intelligent compaction monitoring technology forunbound materials,” Final Report MN/RC-2007-10, Min-nesota DOT, St. Paul, Minn, USA, 2007.

[45] Mn/DOT, Excavation and Embankment—(QC/QA) IC QualityCompaction (2105) Pilot specification for Granular TreatmentS.P.0301-47, Minnesota DOT, St. Paul, Minn, USA, 2007.

[46] Mn/DOT, Excavation and Embankment—(QC/QA) IC QualityCompaction (2105) Pilot Specification for Non-Granular SoilsS.P.6211-81, Minnesota DOT, St. Paul, Minn, USA, 2007.

[47] RVS 8S.02.6, “Continuous compactor integrated compac-tion—proof (proof of compaction),” Technical ContractStipulations RVS 8S.02.6—Earthworks, Federal Ministry forEconomic Affairs, Vienna, Austria, 1999.


[48] ZTVE StB/TP BF-StB, “Surface Covering Dynamic Com-paction Control Methods—German Specifications and Reg-ulations,” Additional Technical Contractual Conditions andGuidelines for Earthwork in Road Construction and TechnicalTesting Instructions for Soil and Rock in Road Construction,Research Society of Road and Traffic, Germany, 1994.

[49] ATB Väg, “Kapitel E-Obundna material VV Publikation2005:112,” General technical construction specification forroads, Road and Traffic Division, Sweden, 2005.

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2010

RoboticsJournal of

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation http://www.hindawi.com

Journal ofEngineeringVolume 2014

Submit your manuscripts athttp://www.hindawi.com

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation http://www.hindawi.com

Volume 2014

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

SensorsJournal of


Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks


fieldassessmentandspeciﬁcationreviewforroller-integrated ... · 2019. 7. 31. · time. if...

Documents