TECHBRIEF

Intelligent Compaction
CASE STUDY • SPRING 2014
SUMMARY OF INTELLIGENT COMPACTION
ON SOIL AND SUBBASE

INTRODUCTION
ABSTRACT Intelligent compaction (IC) refers to an innovative pavement
This Tech Brief represents the construction technology in which conventional rollers are
equipped with instrumentation that is used to monitor and
second in a series of three
control the material compaction process. The technology,
tech briefs that are targeted
which is applicable to both soil/subbase and hot-mix
at sharing information about asphalt/warm-mix asphalt (HMA/WMA) compaction,
intelligent compaction (IC) provides graphical information that a roller operator can use
technology and helping to to better manage his roller operations. This, in turn, ensures
promote its use throughout the that the target properties of the layer are achieved in a more
U.S. The first Tech Brief provides uniform and efficient fashion.
an Executive Summary that
IC technology was developed in Europe in the 1980s and
provides background informa-
has proven to be an effective construction practice in many
tion, a detailed description of European countries. IC arrived in the U.S. roughly 10 years
the technology, and identifies ago and, despite its proven success and demonstrated
the advantages and imple- benefits, it is only slowly being implemented into U.S.
mentation considerations. It is practice.
applicable to IC for both soil/
subbase and hot-mix asphalt/
warm-mix asphalt (HMA/WMA)
construction. Using the find-
ings of recent demonstration
projects, this Tech Brief provides
information that demonstrates
the field application of this
technology for soil and subbase
materials.

Antonio Nieves
[email protected]
(202) 366-4597
 DESCRIPTION OF IC FOR SOIL/BASE CONSTRUCTION
OVERVIEW: When conventional roller compaction equipment and
techniques are used for pavement construction, roller operators are
essentially blind to what is happening to the underlying materials. To achieve
adequate compaction, the operators must rely heavily on the application
of pre-established roller patterns and the use of portable gauges that
measure density at spot locations. In contrast, the continuous graphical (and A sin gle - drum D yn ap ac ro l l er eq ui pp ed w i t h
roller-integrated CMV m eas urem ent s ys tem .
numerical) information produced by IC-equipped rollers create a window into
the layer compaction process.

INSTRUMENTATION: The rollers used for IC on soil and subbase materials are basically the same as those used
for conventional compaction techniques. The primary difference is the instrumentation system that is attached to
measure, display, and record the compaction effects. The key components of this system are described below:

• GPS – Global positioning systems are used to accurately locate

the roller on the project. A real-time kinematic (RTK) GPS ca-
pable of providing the highest level of precision location (within
1 to 3 cm) is required for IC operations. This level of precision
is achieved through the use of single or multiple RTK on-site
GPS base stations or virtual reference stations (VRS), depending
on the terrain. If the line of sight is unobstructed, the effective
range of a GPS station on the project is approximately 2 miles.
On IC rollers, the GPS antenna is mounted on the top of the cab.
RT K u s i n g a H a nd h e l d rove r /receiver w it h a reference net wo rk

• Accelerometer – The key component of the IC system is a small device that is

mounted on the roller frame near the vibratory drum (see image to right). Double
vibratory drum rollers may have two accelerometers, one mounted near both
vibratory drums. The accelerometers measure the vertical acceleration of the roller
frame as it moves down the road project.
• Processing – The software for processing the IC data is
proprietary and manufacturer dependent. Each software product
is capable of conducting a real-time analytical process in which
the acceleration is converted to downward displacement and
then combined with other roller information (such as amplitude,
frequency, and speed) to produce a continuous profile of the
level of compaction achieved in the soil/subbase layer. The
software also uses information from the accelerometers to
indicate whether there is a problem with drum bouncing or soft
spots in either of these layers.

O ver vi ew of I CM V co m p a c t io n mo nito r ing systems

• Automatic Feedback Control – AFC refers to a feature available from some IC roller manufacturers in which the
vibration amplitude and/or frequency are automatically adjusted if “drum jumping” above a certain threshold is
detected. This phenomenon can occur when the dynamic reaction of the soil or base becomes synchronized with the
frequency of the roller. By reducing or eliminating drum jumping, the compaction process can become more efficient
and more uniform.
2
 • Visual Display – A monitor located in the cab displays real-time compaction
information in both numerical and graphical format so that roller operators can
make appropriate corrections or adjustments to the roller operation. The displayed
information includes roller amplitude, frequency, GPS location, and speed. The effects
of a roller pass are displayed in varying colors (or shades of color) so that the operator
Cate rpi l l a r on -b o a rd d is p l ay u nit can track his location along the project, monitor the increase in layer stiffness, and
with Acu G ra d e ® s o f t wa re.
ensure uniform coverage.
• Data storage – The system produces a complete digital record of the compaction process. The stored data can be
downloaded at any time for analysis and documentation.

Each of the components is part of an overall IC system.

The system does require routine inspection and calibration
checks; however, it is relatively reliable and robust considering the
construction conditions under which it must operate.

EQUIPMENT MANUFACTURERS: Ammann, Bomag, Case/

Ammann, Caterpillar, Dynapac, and Sakai are the primary
manufacturers of single-drum IC rollers for soil and subbase
compaction. Currently, all six produce IC rollers that are available
in the U.S. Of the six manufacturers listed, four of them (Ammann,
Bomag, Case/Ammann, and Dynapac) offer the AFC capability.

The primary vendors for GPS equipment including base stations and
virtual reference stations include Trimble, TopCon, and Leica. Trimble
also offers equipment and software for retrofitting IC on an existing
roller.
B om ag Vario Control System

INTELLIGENT COMPACTION MEASUREMENT VALUE: A standard measure for reporting the compaction results
of IC rollers has not been established. Accordingly, each IC equipment manufacturer has developed its own unique
method for characterizing the level of soil/subbase layer compaction (which in turn is incorporated into its own
proprietary software for data processing and display). Since no standard exists, the term intelligent compaction
measurement value (ICMV) is used to represent (collectively) the compaction measures produced by the different IC
roller equipment. Following is a brief description of the five ICMVs currently used.

• Compaction Meter Value (CMV) was the first ICMV. It was established in 1976 by Geodynamic (a Swiss company) in
cooperation with the Dynapac Research Department. CMV is an indicator of the soil/subbase layer’s stiffness/modulus
and is dimensionless. Dynapac, Caterpillar, and Trimble report CMV.
• Vibratory Modulus (EVIB) is a measure of the compacted layer’s stiffness as produced by Bomag IC rollers. Its units are
in mega pascals (MPa).
• Roller Integrated Stiffness (kb) is the measure of the compacted layer’s stiffness as reported by Ammann/Case rollers.
Its units are in mega-Newtons per meter (MN/m).

3
 • Machine Drive Power (MDP) basically defines the amount of additional power required (in kJ/s) by the roller to
compact a given soil (or subbase layer) over the power level required to compact the calibration layer. Thus, a
positive MDP means that the layer being compacted has not yet reached the level of compaction associated with
the calibration layer. Similarly, a negative MDP means that the layer being compacted has achieved a higher level of
compaction than the calibration layer because it requires less power to move the roller. MDP may also be reported
as an index property (either MDP80 or MDP40) in which the power required is rated on a scale of 1 to 150, where 150
corresponds to the compaction level of the calibration layer.
• Compaction Control Value (CCV) is a measure of the compacted layer’s stiffness as produced by Sakai’s IC rollers. Like
CMV, CCV is dimensionless.

KEY FINDINGS OF THE FHWA/TPF DEMONSTRATION IC PROJECTS

FOR SOIL AND SUBBASE
PROJECT OVERVIEW: Between 2008 and 2010, 13 states participated in an FHWA-sponsored, Transportation
Pooled Fund (TPF) study entitled “Accelerated Implementation of Intelligent Compaction Technology for
Embankment Subgrade Soils, Aggregate Base, and Asphalt Pavement Materials. “ The primary purpose of the study
was to demonstrate and evaluate IC technologies through the execution of multiple field projects. Of the 13 states,
6 participated in the soil/subbase field demonstration projects (or case studies). These included Indiana, Kansas,
Mississippi, North Dakota, New York, and Texas. Following is summary information on each demonstration project.

Lo c at i on an d L ayers Co mpac ted

H i g hway M achine(s) I n-situ Te s ts *
Year Using IC

W. L a f aye tte,
IN S andy em b ank m ent Cater p i llar ( si ngle
SR 2 5 LW D, D C P, FWD
f ill and silt y c lay dr um )

Weat hered shale and

c lean c lay f ills o n Cater p i llar and
Pl e a s a nto n , K S LW D, D C P, NG ,
US 69 sub grade and st i f f S ak ai ( si ngle - dr um
(2 0 0 8 ) F W D, PLT
weat hered shale o n p adfo o t )
c lean c lay sub grade
Granular b ase, cem ent
t reated granular b ase, Cater p i llar and
Way n e s b o ro, LW D, D C P, NG ,
US 84 granular sub grade, Case/ Am m ann
M S (2 0 0 9 ) F W D, PLT
and cem ent t reated (A F C and m anual)
granular sub grade

Cater p i llar
Silt y sub grade and LW D, D C P, CBR,
M a r m a r th , N D US 12 ( p adfo o t and
salvage b ase NG, F W D, BST
sm o o t h dr um s)

Em b ank m ent, Cater p i llar and

Sp r i n g vi l l e, NY 2 1 9 em b ank m ent ( wi t h LW D, D C P, NNG ,
B o m ag (A F C and
(2 0 0 9 ) under ly i ng t i re f ill) , F W D, PLT, BCD
m anual)
and gravel sub b ase

Case/ Am m ann and

Sub grade, lim e
Fo r t Wo r th , T X D y nap ac ( sm o o t h LW D, D C P, NG ,
FM 1 5 6 st ab i lized sub grade,
(2 0 0 8 ) and p adfo o t F W D, PLT, PSPA
and f lex b ase dr um s)

*Te st Cod e s: B S T- B o re H o l e Sh e a r Te s t, C B R- Cali fo r ni a B ear i ng R at i o, D C P- D y nam i c Co ne Penet ro m eter,

P S PA-Por t a b l e S e i s m i c Pave m e nt An a l y ze r, F W D - Falling Weight D ef lec to m eter, LW D - L ight Wei ght
D e f l e c tome te r, N G - N u cl e a r D e n s i t y G a u g e, NNG- No n- nuc lear D ensi t y G auge, PLT- Plate Lo ad Test.

4
SUMMARY OF KEY FINDINGS: The use of IC was demonstrated to be a success in all six states. The
following table provides a summary of some of the more significant findings from each state’s demonstration
project.

State M ost Signific ant Findings

• Granular embankment fill Machine Drive Power generally increased with the number of passes, but
no significant differences were detected using different amplitudes.
I ndiana
• MDP values are generally repeatable in both forward and reverse passes.
• WD modulus values correlated best with MDP values.

• Sakai CCV and Caterpillar MDP measurements are repeatable.

• Color-coded maps of the IC data with 100 percent coverage information provided the opportunity to
K ansas visualize compaction quality over a production area or at a given point.
• Linear regression analysis produced poor to good correlations between IC and point measurement
values.
• Empirical correlations between ICMVs and different in-situ measurements were weak sometimes.
• ICMVs generally correlated better with modulus-based and CBR in-situ measurements than with dry
density in-situ measurements.
M ississippi
• IC data indicated that the response distance for altering the amplitude and/or frequency was about
1 to 2 m for a roller speed of about 4 km/hr. The Case/Ammann roller on this project reported ICMVs
every 1 m. A higher sampling rate is required to accurately evaluate the response distance.

• MDP increases with pass count, and high-amplitude setting results on lower MDP than low-
Nor th Dakota
amplitude setting.

• Good correlation between ICMVs and in-situ point tests can be obtained (as long as both tests are
conducted properly).
New Yor k
• More testing is necessary to evaluate the AFC mode compaction operations with variable subsurface
conditions

• IC rollers can effectively identify soft spots (e.g., wet zones) and hard materials (e.g., box culverts).
• Padfoot rollers were demonstrated to produce excellent compaction curves.
Texas • Results of in-situ NG, DCP, and LWD did not match well with ICMVs. PLT and FWD results correlated
better with ICMVs.
• ICMVs resulting from automatic feedback control and manual control exhibited a similar trend;
however, the AFC resulted in higher ICMVs.

Additional findings from the FHWA/TPF Demonstration projects fall under three categories: compaction
curves, compaction uniformity, and correlation studies. Compaction curves refers to the two-dimensional
graphs of a layer’s level of compaction versus the number of roller passes. They are typically used to
determine the number of roller passes that are required to achieve the target level of compaction.
Compaction uniformity is important because of its impact on pavement performance. If the layers of the
pavement structure are not compacted uniformly, the pavement is more likely to perform poorly, because
of all the weak spots associated with poor uniformity. Correlation studies refer to the work done to develop
meaningful relationships between ICMVs and other in-situ measurements, such as moduli, density, and CBR.
Following are tables that summarize the key findings for these factors as they pertain to IC.

5
 Key Findin g s fo r Co m p a c t io n Cur ve s

• Generally, ICMV and in-situ measurements increase with increasing number of passes; however, the rate of increase
gradually decreases until close to being constant.
• In some cases, the ICMV may drop during the first and second pass before increasing.
• For some materials, the ICMV may increase and then decrease with each roller pass.
• Compaction curve results are material dependent.
• Compaction curve is affected by material moisture condition, with the dry and optimum moisture resulting in higher
ICMV with each roller pass.
• The grade of the compaction area may affect the compaction curve.

Key Findings fo r Co m p a c t io n Unifo rm it y

• A higher vibration amplitude usually results in a lower compaction uniformity.

• With each roller pass, the compaction uniformity may increase or decrease.
• There was no consistent proof that either manual or AFC mode achieves higher uniformity.
• Materials that require curing typically exhibit more non-uniformity.

Key Finding s f ro m Co rre l at io n St ud ie s

• ICMV increases with in-situ measurements. ELWD and EFWD had a good linear correlation; however, others (Plate Load
Test results, CBR (from DCP), and non-nuclear density) were generally poor.
• Depending on the material, the linear relation may be either direct or logarithmic.
• There was no evidence that machine settings such as frequency, amplitude, and speed will affect the correlation quality.
• Separate trends were observed for different materials.

CONCLUSIONS
1. The primary conclusion that can be drawn from the FHWA/TPR study is that IC is an effective method for achieving
the target level of soil and subbase compaction. Each of the IC rollers (and their respective instrumentation) worked
satisfactorily in the demonstration projects where it was applied. There was no attempt in the study to compare the
different IC rollers.
2. Both flat and padfoot steel wheel drum rollers were effective in their respective applications. With the exception of one
state (Texas) reporting that padfoot rollers produced excellent compaction curves, there was no indication that one roller
type was more effective for IC than the other.
3. Because of the ability to visually monitor compaction with each pass, IC rollers do increase the level of compaction
uniformity as compared to conventional roller operations. However, there was no data to indicate that changing the roller
settings (amplitude, frequency, or speed) results in improved uniformity during an individual pass.
4. Automatic feedback control was not demonstrated to improve compaction uniformity (as compared to rollers that were
not equipped with the feature).
5. IC equipment can be used effectively to develop compaction curves and to determine the optimum number of roller
passes.
6. The correlation studies showed some correlation between the ICMVs and the in-situ measurements. However, there was
generally a lot of variability. The correlations between ICMVs with deflection-based moduli produced the best results.

6
 AVAILABLE RESOURCES
The best source of information currently available on IC is
the website: www.intelligentcompaction.com. In addition
to providing valuable up-to-date information on most IC
topics, it also identifies sources of information from previous
and currently ongoing research efforts and case studies and
provides links to other useful websites.

In addition to the Transportation Pooled Fund project (TPF-

5(128)), there are two other recent studies that address
intelligent compaction of soils and subbases:

NCHRP Report 676, Intelligent Soil Compaction Systems, National

Cooperative Highway Research Program, Transportation Research
Board, National Research Council, Washington, D.C., 2010.

Evaluation of Intelligent Compaction Technology for Densification

of Roadway Subgrades and Structural Layers, Draft Final Report,
Wisconsin Highway Research Program, WHRP Project ID #0092-08-
07, June 2010.

Te ch n ical Cont ac t s
Antonio Nieves
[email protected]
(202) 366-4597

FHWA-HIF-13-052