GROUND IMPROVEMENT – STATE OF THE ART

M. Terashi1 and I. Juran2

ABSTRACT

The ground conditions of construction site became worse than ever due to the over-population in the urban
areas throughout the world. Good quality material for the construction is becoming a precious resources to be left
for the next generation. Due to these reasons and environmental restrictions on the public works, ground
improvement is becoming a necessary part of the infrastructure development projects both in the developed and
developing countries in the past decade. The basic principles of ground improvement were unchanged since the
beginning of the history of mankind. The practices, however, are changing with time due to the development of
new materials, new machinery, new technologies such as computer, soft ware and sophisticated instrumentation.
The application is diversified due to the ever changing needs of practicing engineers. The present state of the art
outlines the principles of ground improvement and focuses on the recent development of such traditional
technologies as densification, consolidation/dewatering, and admixture stabilization.

1.0 INTRODUCTION

Mitchell (1981) delivered an excellent State of the Art Report on Soil Improvement in Stockholm at the 10th
International Conference on Soil Mechanics and Foundation Engineering. This must have been the first
comprehensive state of the art paper included in a main session in an international conferences organized by the
International Society. The Eighth European Regional Conference on Soil Mechanics and Foundation Engineering
in 1983 held in Helsinki devoted all the sessions to cover a wide range of ground improvement. Mitchell’s paper
and the three volumes of the European Conference (Rathmayer and Saari, 1984) are still very good source of
information for practicing engineers to understand the traditional ground improvement technologies. Stimulated
by these publications and enhanced by the emerging needs of ground improvement, we had a number of specialty
conferences and symposia in 1990’s but each focusing on a limited topics. Later in 1997 in Hamburg at the 14th
International Conference, one of the plenary session was allocated again for the ground improvement. Schlosser
(1997) in his theme lecture, concentrated on a recent development of reinforcement and grouting. Therefore, it is
quite appropriate for the organizing committee of GeoEng 2000 to select Ground Improvement and
Reinforcement as one of the major topics for up-dating the state of art on this subject since 1981.
When we look back the development of ground improvement technologies in the last two decades, we can
raise probably five major topics; 1) the development and wide spread use of soil and rock reinforcement by such
materials as geosynthetics, anchors and micro-piles, 2) enlarged application of deep admixture stabilization
known as deep mixing which was originally developed in Japan and Sweden and recently found its application
worldwide, 3) Jet grouting also originated in Japan which has changed a former image of grouting technology, 4)
the application of ground improvement technology to remedy contaminated ground, and finally 5) ground
improvement for the liquefaction or other hazards associated with large earthquakes. According to the organizing
committee of Geo-Eng 2000, the first topic will be covered by Schlosser and Brandl in a companion keynote
lecture as well as two issue lectures by Barley & Windsor and Bruce. The third topic of grouting technology is
covered by Welsh & Burke in their issue lecture. The fourth and fifth topics will be covered by the corresponding
sessions on environment and seismic problems. Under these boundary conditions, authors are requested to
overview the ground improvement technologies.
The structure of this keynote lecture is as follows. First part is aimed to provide a classification of wide variety
of ground improvement technologies based on their principles although with brief descriptions. Second is to
provide the general concept of selecting an appropriate ground improvement technology or a combination of
them for specific applications which includes a case history as an example. Third is composed of selected recent

1
Masaaki TERASHI, Principal, Nikken Sekkei Nakase Geotechnical Institute, 212-0055, Kawasaki, Japan
2
Ilan JURAN, Professor, Polytechnic University, Brooklyn, New York, 11201, USA
topics which authors think important. Finally the authors conclude the keynote by describing the perspective of
ground improvement to the next decade.

2.0 CLASSIFICATION OF GROUND IMPROVEMENT BASED ON ITS PRINCIPLES

Mitchell (1981) described a variety of ground improvement technologies under six categories based on
principles. The basic principles of the ground improvement have been acknowledged since the mankind started
construction on and in the ground (ASCE, 1977). A principle which was not taken was the most simple technique
of “Replacement”. Authors of the current state of the art think it appropriate to classify the wide variety of soil
improvement in the following way;
1. Replacement: Replacement is the most simple in concept and reliable technique if employed properly. Soft
soil, mostly soft clay or highly organic clay under or near the expected structure is removed and replaced by a good
quality foreign materials up to the extent required to maintain stability or to avoid unfavorable settlement of the
structure. Natural sand and gravel were preferred as foreign materials initially. It is because even loosely placed
sand and gravel exhibit good performance in comparison with soft clay and organic soils as far as the static
problems are concerned. However, loose sand if saturated might cause a serious problems in the seismic region,
which was exemplified in the past large earthquakes. Due to the shortage of good quality granular materials and
due to the concern on the dynamic problems, engineered soils are becoming popular in Japan recently. A typical
example is the light-weight stabilized soils which will be described later in the paragraph of admixture
stabilization.
2. Densification: Densification of loose granular soils, heterogeneous soils, municipal wastes, liquefiable soils
is quite a common practice. The purpose of densification is to increase strength and to reduce settlement of loose
granular soils. Often, improvement of uniformity of the originally heterogeneous soil become the purpose of
densification. A group of short wooden piles driven into loose sand layer beneath the buildings may be the
forerunner of this category. Vibro-rod, vibro-flotation, sand compaction pile method, compaction grouting, and
heavy tamping are well established techniques of recent days. Development of efficient vibrating equipment
enhanced the technique and is indispensable technique to tame the liquefiable loose sandy deposit. The
technologies of this category may be described into details at the separate session which devotes on the Seismic
problems. Density after the densification process is in most case predicted preliminary based on the accumulated
experience and confirmed later by the field test.
Vibro-Compaction (Vibro-flotation, Vibro-rod)
Vibro-compaction is a technique of densification by inserting a vibratory probes into in situ soil. The probes
vibrates horizontally (Vibro-flotation) or vertically (Vibro-rods) to densify the soils at the probe location. The
granular materials are supplied from the ground surface to fill the void created by the vibration. In some machine,
the bottom feed of granular material is possible. Typically 3 to 15 m are the improvement depth. Comprehensive
overview of the technology may be found in Brown (1977), Welsh (1987) and Wightman (1991).
Heavy Tamping (Dynamic Compaction, Dynamic Consolidation)
Dropping a wooden pile repeatedly on to the ground surface was a common practice for compacting a
foundation of a small housing for long time. Menard (1975) increased the impact energy tremendously,
introduced systematic repetition of tamping and improved this as a modern technique of deep compaction.
Tamper nowadays is made of steel or concrete block and weighs 6 - 35 tons. Drop height ranges from 10 to 20
meters or more. The depth of influence is proportional to the square root of the tamper weight times the drop
height. The recent development of Heavy Tamping may be found in Mitchell and Zoltan (1984), Mitchell and
Welsh (1989), ASTM Committee on Soil and Rock (1990), Rollins and Rogers (1994), Chow et al. (1994), Lo
et al.(1990). Oshima and Takada (1997, 1998) simulate the compaction process of heavy tamping by
geotechnical centrifuge modeling.
Sand Compaction Pile Method
The equipment of the Sand Compaction Pile Method (SCP) resembles with the equipment for sand drain
installation. When the equipment reached a desired depth, the equipment is withdrawn leaving a loose sand pile of
predetermined length through its mandrel. Then with the aid of a vibrator at the top of the mandrel, the mandrel
compresses the sand pile and expand its diameter. By the sequence of this process, compacted sand piles are
created and the surrounding soils are also densified. In the alluvial plain, quite often the ground is composed of the
alternate layers of loose sand and soft clay. SCP method can create a loose sand pile for vertical drainage in the
clay layer and simultaneously densify the loose sand layer by compaction. SCP also used to create a group of
compacted sand piles even in the clay layer. In this case, the compacted sand piles functions to reduce settlement
and reinforce soft clay layer as a system. Therefore this technique is one of the most preferred ground
improvement in Japan to tame soft ground. Aboshi et al. (1990) provided a concise state of the art of this
technology.
3. Consolidation/ Dewatering: When a foundation ground is the cohesive soils with low strength and low
permeability, structures constructed on the ground will experience the stability problem and/or the long term
unfavorable settlement. These soils however increase strength and improve their compressibility with time under
the sustained loading. Applied external load causes the increase of total stress in the ground. The increment of the
total stress is sustained by the excess pore water pressure if the soil is saturated. Then excess pore water pressure
dissipates with time and results in the reduction of soil volume and increase of effective stress and increase of
strength. This is the consolidation. If time allows, preloading of the ground prior to the construction by the
embankment fill whose load intensity is equivalent to or exceeding that of the expected structure will solve the
problem. As the soil to be improved has a low strength, it is not always possible to place a required embankment
fill in a single stage. Most often, preloading is done by staged construction to avoid the instability at the edge of
embankment. Preloading by embankment fill is one of the oldest technique to improve this kind of soils.
As is mentioned above preloading is given to the ground with final target of increasing effective stress. The
same can be achieved by alternative techniques of decreasing the pore water pressure. The alternatives are the
application of vacuum, dewatering the ground water, electro-osmosis and quick lime piling. When there is a
certain limitation on the space for the required embankment construction or there is no resources for fill materials,
these alternatives are most effective. The further merit of these alternatives are that they will not accompany the
increased shear stress and hence will not create a stability problem.
With increasing thickness of cohesive soil (with increasing drainage path), consolidation time becomes longer
and unacceptable. The idea of accelerating the consolidation by reducing the length of drainage path was born in
1930’s in USA and in Nordic countries. Commonly used artificial drainage are vertical drainage by means of sand
drain or prefabricated drains.
Apart from the application to cohesive soils, vertical drainage is recently employed to dissipates quickly the
excess pore water pressure induced by the earthquakes in order to tame the liquefaction. This special application
will be discussed in a separate session in the present conference.
4. Grouting: By ASCE Committee on Grouting (1995), Grouting is broadly defined as the placement of a
pumpable material which will subsequently set or gel in pre-existing natural or artificial openings (permeation
grouting) or openings created by the grouting process (displacement or replacement grouting). Major purpose of
grouting technology is to provide increased strength and/or to retard water seepage of soil or rock formation.
Grouting is also used to compensate the unfavorable displacement of existing structures.
When grout, a pumpable material is injected into a soil or rock formation, it may permeate into the natural
openings such as void space of the soil and fissures in the rocks, or create a opening by fracturing the soil mass, or
displace the surrounding soil. The final location of the grout and the maximum distance of travel from the
injection point are determined by a number of factors; viscosity of the grout, get time, size of the particles in
relation to the openings, injection pressure, rate of injection, properties of soil and rock to be grouted, and so forth.
As a consequence, the completed grouted zone usually has an irregular shape and imhomogeneity. Hence the
R&D efforts have been paid during past decades to control the shape and extent of the grout, to control the
expected or unexpected displacement caused by grouting process, and to predict the quality of grouted formation.
These research efforts has led to innovative grouting equipment, development of new grout materials, and
application of sophisticated data acquisition and simultaneous computer control of the grouting process.
Jet grouting which was developed in 1970’s in Japan is in between grouting and deep mixing. The jet grouting
is composed of a combined process of cutting soil with high pressure and filling the space created by cutting with
grout. When most of the soil cut by the jet is discharged to the ground surface, it is thought as a family of the
replacement grouting. However during the process mixing of grout and in situ soils is unavoidable and the
strength gain is influenced with the soil type. When one place importance on this aspect of mixing, the technique
may be classified as a part of deep mixing.
Comprehensive information on Grouting of recent decade may be found in ASCE Committee on Grouting
(1995), Borden et al. (1992), Yonekura, et al. (1996) and Grouting Committee of Geo-Institute (1998).

5. Admixture stabilization: Admixture stabilization is a technique of mixing chemical additives with soil to
improve the consistency, strength, deformation characteristics, and permeability of the soil. These improvement
becomes possible by the ion exchange at the surface of clay minerals, bonding of soil particles and/or filling of
void spaces by chemical reaction products. Although a variety of chemical additives has been developed and used,
most frequently used additives nowadays are lime and cement due to its availability and cost.
The mechanism of the lime and cement stabilization has been studied extensively in 1960s by the highway
engineers in relation to the improvement of base and sub-base materials for road construction. The needs of rapid
construction enhanced the application of the technique to the deep in situ soils. The deep mixing method, deep
admixture stabilization was developed in Japan in 1970s. Lime columns developed in Sweden at the same period
is the same technology in principle. The deep mixing utilize the mixing blades or augers to manufacture a treated
soil column of predetermined size and shape in situ. The strength of treated soils is in the order of 100 to 1,000 kPa
in terms of unconfined compressive strength. For the deep mixing, a series of specialty international conferences
are co-organized by ISSMGE TC-17 and member societies. The proceedings of these conferences will provide
useful case histories and recent development (Yonekura, Terashi and Shibazaki, 1996; Bredenberg, Holm and
Broms, 1999; Rathmayer, 2000). The details will be discussed later in the present report.
Since the late 1980s to the middle of 1990s new technologies to improve the fill materials prior to reclamation
work were developed by the port and harbor engineers of Japan. To cope with the lack of good quality material for
reclamation or to save the environment at the borrow area, even inappropriate materials must be used for sea
reclamation. These materials are often come from maintenance dredging of navigation channels or from
construction waste soils from the urban areas. The common practice is to improve these fill materials after the
reclamation to a desired level by the traditional technologies such as densification or drainage (Henmi et al., 2000).
Sometimes, the improvement of such materials prior to reclamation is cost saving. One of the new technique is
pre-mixing of small amount of cement with sandy material to improve liquefaction resistance (Zen et al. 1987). It
is reported that the unconfined compressive strength around 100 kPa is sufficient to prevent liquefaction in most of
the cases (Coastal Development Institute of Technology, 1999a). The other is pre-mixing of cement and air foam
or EPS (expanded polystyrene) beads with clayey soils to manufacture a good quality fill material with
pre-determined density and strength. Improved soil is called “the Light Weight Geo-material” (Tsuchida et al.
1996; Coastal Development Institute of Technology, 1999b).
6. Thermal stabilization (Heating and Freezing): Thermal stabilization is divided into two groups of heating
and freezing.
Even at the ordinary temperature under the sun shine, properties of fine-grained soils are improved by
desiccation. This is often found as a dry crust formed at the surface of reclaimed sludge. When the reclamation
process is very slow, the thickness of desiccated layer becomes several meters (Katagiri et al, 1996). The
artificial heating is naturally much more effective and the applications of heating the soil up to 300 to 1,000
degree Celsius has been reported. As far as the improvement of the mechanical properties are concerned
Mitchell (1981) is still a good source of information with comprehensive references.
Recently heating finds its application in the remediation of contaminated soils. Heating the soil at moderate
temperature assist the vapor extraction of volatile organic compounds. Soil vapor extraction performance can
be enhanced or improved by injecting heated air or steam into the contaminated soil through the injection wells.
Heating the soil to extremely high temperature is the in situ vitrification by which electrical current is used to
heat and melt the soil in place. The technique is effective for the soils contaminated with organic, inorganic and
radioactive compounds. The details of the soil remediation techniques will be discussed in a different sessions
in the present conference.
The first reported use of ground freezing was in South Wales in 1862 in conjunction with mine shaft
excavation (ASCE, 1997). The strength of frozen soil is in the order of 1 to 10 MPa although it depends on a
variety of factors such as soil type, water content, rate of freezing, temperature of frozen soil. Frozen soil
becomes nearly impermeable materials. The technique is currently used for the temporal increase of strength
and temporal shut off of water seepage around open cut, shaft excavation and tunneling. The specialty
symposia on ground freezing may be a good source of recent applications. Those are International symposium
of ground freezing, Nancy,1994, International symposium on ground freezing (ISGF ’82), Hanover, 1982, and
International symposium on ground freezing and frost action in soils, Lule, 1997.
7. Reinforcement Ground reinforcement consists of creating in-situ a composite reinforced soil system by
inserting inclusions in predetermined directions to improve the shear strength characteristics and bearing capacity
of the existing ground. Ground reinforcement technologies include a constantly increasing diversity of installation
techniques and reinforcing materials which, pending upon the target engineering applications, are designed to
withstand the required resisting forces (e.g. tension, compression, bending moments or their combinations) over
the expected life service of the structure. The recent ASCE State of Practice Review on Ground Improvement,
Reinforcement and Treatment (ASCE, 1977) covers 12 techniques including: bio-technical inclusions, deep
mixing, fiber reinforcement, geosynthetics, ground anchors, jet grouting, lime-cement columns, mechanically
stabilized earth structures, micropiles, soil nailing, vibro concrete columns and vibro-replacement (vibro-stone
columns). Engineering applications of these technologies cover an increasing range of geotechnical construction
projects including soil foundations with vertical micropiles, or jet grouted, vibro-stone or lime-cement columns, as
well as slope stabilization and retaining structures with subhorizontal passive soil nails, horizontal metallic or
geosynthetic reinforcement, or prestressed ground anchors. In the field of ground reinforcement practice has
traditionally preceded research and theoretical developments. Technological innovations put forward by the
construction industry and new construction materials made available by the manufacturing industry have
continuously challenged the infrastructure agencies and the engineering community with the critical needs for
establishing relevant engineering guidelines and quality control standards.
While a wide variety of questions are yet unresolved, it is believed that the soil reinforcement techniques
currently used have matured and became reliable and cost effective engineering tools for geotechnical
construction in difficult site and soil conditions. The increasing amount of full-scale experiments and
performance monitoring of actual structures has provided a significant database for technology assessment and
engineering guidelines. However among the major issues to be addressed (Juran, 1997) one can identify (i)
quality control of innovative materials such as 3-D fiber reinforcement, (ii) long term performance assessment
of various reinforcement types and particularly geosynthetics (i.e. chemical and mechanical durability and in
soil confined creep), (iii) seismic performance of reinforced soil systems, (iv) reinforcement of fine grain soils,
and (v) innovative construction technologies, specifically for in-situ ground reinforcement. These technologies
as well as the current research and State of Practice will be discussed by Professor F. Schlosser in a separate
session of this conference.
8. Miscellaneous ( Combination of above, Use of Heavy weight material, etc)
Engineers facing real life problems must find a solution for their problems at hand. In difficult situations,
combined use of a variety of technologies are common practices. In the construction of embankment on soft
compressible soils, the major portion of the ground underneath the embankment is improve by accelerating the
consolidation with the aid of vertical drainage. At the embankment toe, sand compaction pile method may be
employed to improve stability at the same time. In the braced excavations, excavated bottom may be improved by
deep mixing method to shut off the seepage and prevent the bottom heave. As ordinary deep mixing cannot
improve the vicinity of sheet piles, jet grouting may be employed to fill the gaps between deep mixing and sheet
piles.
When the ground improvement cannot solve the entire problem, the combination of ground improvement and
elaborate foundation systems are combined together. Combination of preloading or lime columns followed by
installation of pile foundations for the housing buildings are common practice in Nordic countries. In the huge
man-made island of the Kansai International Airport, soft alluvial clay layer was improved by vertical drainage,
fill materials above the clay is improved by a variety of densification techniques. To minimize the uneven
settlement of the terminal building which rests on the fill material with changing contact pressures, huge amount
of heavy iron ore was used to bring about the uniform contact pressure. On top of this, the pillars of the building
were equipped with hydraulic jacks to compensate the uneven settlement and save the structure from the
anticipated future settlement.
Retrofit and rehabilitation of the existing structures are increasing demand in the developed countries. To
reduce the external load acting on the structure or to the ground, replacement of a soil in the vicinity of existing
structures with engineered soil is often conducted in Japan. The light-weight geo-materials found the place of
application in such a situation.
The technologies classified above are shown again in Table 1 (Terashi and Miki, 1999). These technologies are
recently employed not only to improve mechanical properties of soils but for the remediation of contaminated
soils. This possibility is well summarized by Mitchell and Van Court (1992).

Table 1 : Classification of Ground Improvement Technologies (Terashi and Miki, 1999)

Category Technology Major improvement effect Appropriate soil type
Surface soil improvement often used in combination with deep in situ ground improvement
Surface Drainage ditch Ensure trafficablity by drainage clay, organic soils
Improvement Sand mat placement Ensure trafficability, Function as clay, organic soils
horizontal surface drainage
Geosynthetic reinforcement Ensure stability of sand mat clay, organic soils
Shallow admixture Increase strength and ensure any soils
stabilization trafficability
Deep in situ ground improvement
Replacement Excavation and replacement Increase shear strength and ensure clay, organic soils
stability, Reduction of settlement
Explosive or displacement Same above clay, organic soils
replacement
Accelerating Preloading by surcharge fill Increase shear strength and ensure compressible soils with
Consolidation by stability, low permeability
Preloading with and Reduction of residual settlement exhibiting large primary
without artificial Preloading by ground water Same above consolidation;
drainage lowering clay, organic soils
Preloading by Vacuum Same above
Drainage Gravel drain Increase liquefaction resistance by Loose liquefiable sand
reducing excess pore water pressure
Densification Compaction piles Increase uniformity of the ground to Variety of granular soils,
Sand compaction pile method reduce uneven settlement, Increase Municipal waste
Vibrating Rod Compaction shear strength, Increase liquefaction materials,
Vibro-flotation resistance with relatively high
Heavy Tamping permeability
Grouting Permeation grouting Shut off or reduce water seepage, uniform permeable soils
Consolidation grouting Increase shear strength of the soil, loose sand
Fracture grouting Compensate the deformation of silty to clayey soils
Jet grouting structures any soils
Admixture Deep mixing method Increase shear strength, Reduction of any soils.
stabilization including Swedish lime settlement, Increase liquefaction mostly for clay and
columns resistance by increasing cohesion or by organic soils, applicable
give constraint to the loose soil mass. also to loose sand
Pre Mixing Increase cohesion to increase loose liquefiable soils
liquefaction resistance of fill materials
prior to reclamation
Light-weight geo-materials Adjust unit weight of soil, increase any soils
shear strength and reduce settlement
Thermal Heating by combustion Improve strength and deformation loess
stabilization characteristics
Ground freezing Temporary purposes/ Increase shear Any wet soils
strength , suppress deformation, and
reduce water seepage
Vitrification/ Injection of Remediation of contaminated soils Contaminated soils
heated air or water into the soil
Earth reinforcement Geosynthetics, Root piles, soil Increase strength and deformation
nails, ground anchors, etc. characteristics of improved soil system
To be covered by different key note
Miscellaneous Use of combined method To cope with variety of problems
associated with difficult soil
conditions
Use of light weight material To reduce the driving force by
such as EPS blocks reducing the weight of soil mass
Use of heavy materials Increase the weight of soils to balance
the pressure acting to the sub soil
3.0 FACTORS TO BE CONSIDERED IN THE SELECTION OF APPROPRIAE TECHNOLOGY

Under the difficult ground or soil conditions, engineers will face a variety of problems during construction or
after the construction until the structures cease their life of function. The problems are caused by inadequate
strength and deformation properties of the soil, by inadequate permeability of soil or the ground formation, or by
inadequate local topography of the construction site. It is not determined by the ground condition alone whether a
particular site is adequate or not. This is to be determined in the relation between the ground and requirements
posed to the ground from the expected function of the structures to be built. The procedure of selecting the
appropriate solution and the process to successful completion of the work will be described in the followings.

Table 2 : Determination of appropriate solution and recommended procedure (Terashi and Tanaka, 1998)

Soil profile and engineering properties Characteristics & requirements of

of soils, Ground water condition, superstructure & its foundation type,
Environmental concern Cost and time constraint.

Preliminary design of expected structures based on the

original ground condition to find out problems

Variety of factors
to be considered

Selection of countermeasures
Alternate foundation type and/or Ground Improvement

Additional soil survey & testing

if necessary for the alternate solutions

Design for each GI technique & Examine problems associated

cost comparison with construction (Site specific)

Determination of Final Solution

Field Trial

Detailed design and QC and QA program

Actual construction with monitoring

Completion of work and monitoring for maintenance

Most of the geotechnical engineers may have an experience of being asked by a client what is the best
technology to accomplish work, for example, on a compressible clay soil with STP N value is nearly zero. It is
impossible question to answer because there is a tremendous lack of information. Our way of thinking may be
summarized as shown in Table 2 (Terashi and Tanaka, 1998).
 Starting point may be understanding the characteristics and function of structures to be built on a specific site.
Figure 1 shows the impression of Japanese practicing engineers on the changing engineering issues, in other words,
requirements raised by the structures in the past 50 years (Nozu, 2000). Stability and settlement are old and
continuing issues on soft ground. In 1950’s and in 60’s when we do not have so much options of ground
improvement, safe construction without excessive settlement was the major requirements posed on the ground. In
1970’s the rapid economic growth urged a use of marginal land and it drove the development of modern
technologies. When a variety of technologies became available the level of requirements increased. Recently we
cannot develop the infrastructure without paying attention to the environment. If the expected structure were the
park we may not worry about substantial total and uneven settlement. Whereas, if the structure were a factory for
precise manufacturing, the foundation for the machinery does not allow even one mm differential settlement.
Embankments for highways and rail roads look similar in shape but it is relatively easier for road embankment for
periodical maintenance. Thereby the requirements for the ground differs even for these similar structures. These
requirements may be given by each authority for buildings, highways, rail roads, ports and harbors, and so forth in
such a way as required factor of safety against stability, allowable residual deformation (settlement and horizontal
movement) under sustained load, allowable deformation during seismic conditions, allowable quantity of water
inflow into the specified area, allowable level of vibration induced by machinery or nearby traffic, and sometimes
by future plan of the use of the site. The cost and time allowed for the construction are also important. If the
construction time is shorter than say one year, we are not allowed to employ staged surcharge preloading.

Figure 1 : Engineering Issues Influencing ground improvement (Nozu, 2000)

Considering the requirements of the structure, preliminary design is undertaken based on the available
geotechnical data on the original (unimproved) ground. If there is no problems anticipated at all, we need not pay
for any countermeasures at the site. Through this examination, the problems of the construction becomes clear.
The problem may be the insufficient bearing capacity, excessive settlement during or after the construction,
uneven settlement leading to the tilt or collapse of the structure, problems associated with excavation or
something else. The investigation also tells us which soil layer is causing these problems. However, quite often,
geotechnical design or structural design of the expected structure must be undertaken with insufficient information
of the ground. For the stability or bearing capacity of small housing we only need information of a surface soil
layers. However, for the settlement prediction of a large structure, we need the information of deep seated
compressible layer, inclination of bed rocks and so forth. When the consolidation parameters are obtained on
disturbed samples, both coefficient of compression and consolidation yield stress are unreliable and the magnitude
of settlement becomes erroneous. This is not rare to see the case histories of successful (?) ground improvement
where there was actually no need of improvement at all. In case where we feel the lack of information, we have to
advise our clients to supplement the geotechnical information. Profile of STP N value alone is almost meaningless,
although better than nothing.
When the problem becomes evident and the depth, thickness and extent of the difficult soil becomes apparent,
the several candidates of countermeasures may be selected. The countermeasures are not always a ground
improvement. The change of foundation type from shallow foundation to deep foundation may be the solution to
escape from the soft surface layers. Simply the elongation of construction time if it is permitted may reduce the
residual settlement. If such a simple solutions are not acceptable, we need the knowledge of ground improvement
as well. The knowledge on major improvement effect and appropriate soil type for each technology is important.
Brief information such as listed in Table 1 may help a preliminary selection. Some technology is appropriate for
clay soil and the others are for granular soils, Fig. 2 provided by Mitchell (1981) may also be helpful. Similarly
important is a soil stratification. Preloading by embankment load is good for clay soil in general. However, when
there is a permeable layer exist within the thick clay soils, vacuum preloading may loose its effect below the
permeable layer. Some are quite powerful but only for the temporary work. In most of the cases, the
countermeasures are combinations of superstructure, foundation type and ground improvement. It may be a
combination of different ground improvement technologies.

Figure 2 : Ground Improvement Method and best suited soils (Mitchell, 1981)

When a ground improvement is proposed for a part of the countermeasures, we carry out comparative
geotechnical design to find out the best solution which meets the requirements within the allowable time and
budget. In the cost comparison, we should not compare the direct cost for the ground improvement alone but we
should compare the total cost of the construction including superstructure, foundations and furthermore the cost
for the maintenance should be taken into account.
In parallel with the design work, various site specific problems associated with the construction must be
studied. Use of certain technologies may be restricted by uneven working surface, limited head room, obstructions
above the compressible layer, unstable working surface, extreme depth of soft layer, extremely soft layer for
anchoring the prefabricated drains, poor site accessibility, overhead or subsurface utility interference. Especially
in the urban area, environmental impacts anticipated by the application of ground improvement may prohibit the
use of certain technologies due possibly by anticipated noise, vibration, deformation of the ground during
improvement, possible ground water contamination, etc. Local availability of materials and equipment,
workmanship and etc. will give rise to cost increase.
When the best solution is reached, quality control and assurance program should be established in parallel with
the detailed design. Field trial of ground improvement is sometimes necessary for assuring the design value by
pilot test before the productive ones. Degree of densification of granular materials by a variety of technology,
average shear strength and uniformity of improved soil by admixture stabilization for example are dependent on
many factors and they should be confirmed through full scale field trial. QC & QA procedures are determined
during these field trial at the actual construction site.
During the construction, monitoring of the improved ground are often undertaken to adjust the inaccuracy in
the design stage and successive construction stage. Prediction never agrees with the actual performance because of
many factors; insufficient accuracy of original soil data, insufficient precision of design tools, inappropriate
materials, inappropriate workmanship, unexpected change in the construction schedule, accidental weather, etc.
Monitored performance of the ground is back analyzed to change the initial design or to change the construction
schedule. Monitoring may be continued even after the completion of the work to obtain the necessary information
for the maintenance work for the future.
The reader may think that the description in this paragraph seems to be rather a text book nature. The authors
placed an importance here to review the current practice in general in order to emphasize that the successful
application of ground improvement depends not only on the development of ground improvement technology.
Often the state of the art of ground improvement is discussed focusing on innovative ideas, materials and
equipment. The following paragraphs concentrate on these issues as usual because of the page limitation.
However, one should not forget that the development of soil investigation and testing to characterize soils, the
development of analytical tools to predict the behavior of ground, the development of instrumentation, capability
of recent computers are all contributing to the state of the art of ground improvement.

3.1 Selection of the Best Solution – An Example (Terashi and Tanaka, 1997)
Even under the similar soil condition and for similar structure, the best solution becomes different. The
example cited here is the expansion of the airport on man-made island at offshore Tokyo. The expansion of the
airport was necessitated to cope with the increased air traffic in the recent years in Japan. The project was
undertaken adjacent to the previous airport in three stages as shown in the Figure 3. While the previous airport was
still in use, the stage 1 (New A R/W) and stage 2 (West Terminal Building) constructions were completed. The
airport function was then moved to the new A runway and a new terminal in the figure. The stage 3 construction of
the new C-runway was started while the facilities completed in the earlier phases are in full operation.

Figure 3 : Construction stages of Haneda Off-shore Expansion Project (Nakata et al. 1997)

Soil profile at the site is schematically shown in the Figure 4. From the surface, there are a construction waste
soil layer (Bs) imported from down town Tokyo, a dredged clay layer (Ac1) created by sludge dredged from
Tokyo Port, a thin layer of alluvial sand (As1), an alluvial clay layer (Ac2), and a relatively stiff clay layers (Dc1)
and stiff sand/clay layer (Dc2). Underneath them is a sand and gravel layer (Dg). There is a general tendency of
increasing thickness of soft Ac2 layer towards the sea (towards the stage 3 area). Another difference at a glance is
that the As1 layer does not exist in the stage 3 area. Besides these difference, the soil profile looks quite similar in
all the three areas.
A variety of structures of airport facilities were constructed in the project site with innovative solutions, the
outline of which has been reported by Nakata et al. (1997). In this article similar structures in the different stages
are focused upon: new A R/W in the stage 1 area, Apron in stage 2 area, and new C R/W in stage 3 area. These are
all similarly vast and flat pavement structures. Although the structures are similar and soil profiles are similar, the
solution became different as shown in Table 3. The specifications of both structures and ground improvement are
caused by different requirements and difference in the details of the soil condition.
 Figure 4 : Soil Profile at Haneda Offshore Expansion Project Site (Nakata, et al. 1997)

Table 3 : Specifications of selected structure and ground improvement (Terashi and Tanaka 1997)
Structure and Area Structure Ground Improvement Employed Improved
Soil Layer
New A R/W Asphalt Concrete Pavement No surcharge load, Ac1
in Stage 1 area Prefabricated vertical drains
Apron Prestressed Concrete Pavement Surcharge Preloading Ac1
in Stage 2 area PC-Lift Up Systems in the with
pavement structure small diameter Packed Sand Drain
Surcharge Preloading with Ac1 and
New C-R/W Asphalt Concrete Pavement Sand drains + upper Ac2
in Stage 3 area supplementary Prefabricated drains

3.1.1 Different requirements of R/W and Apron

Run way is a structure which requires flatness along the longitudinal direction which gives a requirement to the
ground that the gradient should be less than 1.0 % (by the Japanese standard). The traffic load is that of airplanes
when they are passing through. Whereas, the apron is a place for the airplanes to stay sometimes overnight. It
means that the weight of the airplanes becomes sustained load on the apron. The maximum gradient of the apron is
the same with the runway and should be less than 1.0 %. For the apron, however, there exists another restriction of
the minimum gradient of 0.5 % which is necessary to enhance the discharge of rain fall.
Asphalt concrete pavement is preferred for runways in Japan due to its relative ease in the future maintenance
work and due to its flexibility to follow the possible future uneven settlement. Whereas the structure for the apron
is ordinarily cement concrete. This is because the asphalt concrete pavement will susceptible to the creep
deformation under the sustained loading. However, the maintenance of cement concrete pavement to cope with
uneven settlement is much more difficult in comparison with that of asphalt concrete. Even though the structure
looks similar, apron requires higher performance of the foundation ground.

3.1.2 Soil Condition in Stage 1 areas and ground improvement selected for run way
The history of the ground formation at the Stage 1 area is summarized in the Figure 5. Before 1965, the area
was shallow water depth with thick As1 layer deposited on top. The sand of As1 layer was dredged and used for
the reclamation of the previous airport in the period of 1965 to 1970. Due to this dredging water depth became 6 to
15 meters. For 10 years since 1971, the place was used as a disposal pond for the dredged marine clay, which is the
Ac1 layer. Later on, construction surplus soils from down town Tokyo was brought here to cover this marine clay
sludge. Due to the difference in the unit weight of formerly existed sand and imported clay sludge, due to the aging
effect of Ac2, the Ac2 layer beneath these reclaimed layers became an overconsolidated layer in Stage 1 area.
Although the stresses become slightly higher than the consolidation yield stress in the Ac2 layer after the
construction of new A R/W, the settlement of the Ac2 layer is estimated small enough for the requirements given
by asphalt concrete pavement. The purpose of the ground improvement here was decided to complete the primary
consolidation of the Ac1 layer alone to minimize the residual settlement. As the depth of the improvement was less
than 15 m, prefabricated drain was selected as vertical drainage. No surcharge load was provided.

Figure 5 : History of the ground formation at Haneda (Japanese Geotechnical Society, 1999)

3.1.3 Soil Condition in Stage 2 area and countermeasures selected for apron
The history of the ground formation is the same as the Stage 1 area. As the restriction on uneven settlement is
much more strict for the apron, the countermeasure taken here for the apron is a combination of structural counter
measure and the ground improvement. The purpose of the ground improvement here is to reduce the secondary
consolidation of Ac1 layer in addition to complete the primary consolidation before the construction of apron.
Surcharge preloading with vertical drainage was selected. As the depth of improvement here is larger than that in
Stage 1 area, 12 cm diameter sand drain packed in the synthetic net was chosen.
For the structural countermeasure for the future maintenance work, a new idea of PC-Lift-Up system was
chosen for the apron as shown in Figure 6 (Japanese Geotechnical Society, 1999). The pavement structure is
prestressed concrete slab in which cylindrical fittings are installed at regular pattern. When uneven settlement
exceeds the limit value, hydraulic jacks are equipped to the fittings and a block of slab will be jacked up to
maintain flat surface. The opening then will be grouted to compensate for uneven settlement.
 Figure 6 : PC-Lift-Up Method : a structural countermeasure (Japanese Geotechnical Society, 1999)

3.1.4 Soil Condition in Stage 3 areas and ground improvement selected for run way
The history of Stage 3 area was shown in Figure 5. The area is further offshore from the previous airport and
there was no As1 layer. There was no unloading history. The sea reclamation by dredged clay sludge was directly
done on the normally consolidated Ac2 layer. Therefore, all the thickness down to the bottom of the Ac2 layer and
upper part of Dc1 layer became the compressible layer under the weight of runway, Bs layer and Ac1 layer. The
best solution here is no doubt to install the vertical drain all through the compressible layer down to around 45 m
below ground surface to accelerate the consolidation. Unfortunately, however, the site is in the vicinity of new
A-R/W already in full operation. The safety of airplanes limit the height of the working machinery. The maximum
allowable depth of improvement was up to 35 m below the ground surface leaving 10 m of unimproved thickness.
Further difficulty here was that the clay sludge in the Ac1 layer was still almost liquid due to the constraint of
construction time schedule.
A variety of ground improvement technology was employed to cope with these difficulties. They are the
shallow soil mixing to provide the working base for spreading Bs1 layer on top of Ac1 layer. Then the preliminary
soil improvement of Ac1 layer by prefabricated drains to ensure the trafficability of heavy equipment for sand
drains. Finally the installation of ordinary diameter sand drains reinforced its top with fabric reinforcement.
Surcharge load was provided to enhance the consolidation as far as possible during the allowed time.
Even with such efforts, the residual settlement at Stage 3 area was estimated as large as 1.5 m in 10 years after
completion. In the process of designing computer codes was developed to analyze the consolidation process with
complicated configuration of vertical drainage as shown in the Figure 7. Probability based prediction of the
uneven settlement was also conducted. The continuos monitoring of the settlement of the airport was continued
after the completion of the work to program the appropriate maintenance.

Figure 7 : Installation of Composite Vertical Drainage (Japanese Geotechnical Society, 1999)

4.0 STATE OF PRACTICE FOR SELECTED TOPICS

Even the traditional ground improvement has a number of varieties, it is not possible to cover all the
technologies and describe them into details. And it is not necessary to do so because of the availability of
numerous conference proceedings and text books. In the following paragraphs, the advance of a couple of
technologies which authors think important will be described into some details.

4.1 Deep mixing Method

Deep mixing method (DMM) is a kind of deep in-situ admixture stabilization using such hardening agents as
lime, cement, or a combined materials thereof. The concept of lime stabilization has a long history. The approach
of in-situ mixing also may have a variety of roots in the past. However, the research and development of DMM as
it is today was started in the late 1960's using lime as a hardening agent. DMM was put into practice in Japan and
Nordic countries in the middle of 1970’s, then spread into the other part of the world. Lime is replaced mostly with
cement in Japan and replaced with lime-cement mixture in Nordic countries. More recently, the combined
materials of lime or cement with gypsum, fly ash and slags have appeared and are employed for particular
purposes. Two decades of practice have seen the equipment improved, hardening agents changed, research efforts
paid, and experience accumulated. However, the applications are still being diversified to cope with difficult
constructions.
Table 4 is a quick review of a variety of deep mixing method which may be roughly divided into three
categories. The dry method includes Deep Lime Mixing (DLM), Dry Jet Mixing (DJM) and Swedish Lime
Columns that utilize granular or powdered hardening agents. The wet method includes a variety of technique but
are called in general as Cement Deep Mixing (CDM) in Japan which utilize cement slurry or cement mortar. The
wet and dry method of deep mixing all employ some sort of rotating blades for achieving the mixing. The last one,
Jet Grouting use cement slurry at high pressure without the help of any mixing tools. Jet grouting was developed in
Japan to overcome the drawbacks of other grouting technologies such as irregular completed shape and
heterogeneity. Jet grouting in its initial development stage, the diameter of the completed grouted soils still
dependent on the pressure and soil type, although the homogeneity was highly improved. Recently a technique
using cross jet (or collision jet) overcomes the final difficulty of uncertain completed diameter (Shibazaki,1996).
When we ignore the higher cost, Jet Grouting is now equivalent technology with the other deep mixing with
utilizing mixing blades. The equipment of Jet grouting is much smaller in size, it may replace deep mixing when
the working space is limited. The application of Jet grouting and Deep Mixing are now becoming closer. As the
technique is reviewed by an issue lecture, there is no further descriptions here.

Table 4 : Development of deep mixing method

Category Name / hardening agent Application

Dry DLM / granular quick lime ‘74-‘78
method DJM / powdered cement ‘80 to present
Lime Columns / middle of 70’s to
powdered lime/ lime-cement present
Wet CDM / cement slurry or ‘75 to present
method CMC / cement mortar, slurry ‘74 to present
Jet Cement slurry with the aid of ‘68 to present
Grouting high water and air pressures

4.1.1 Historical Review of R & D of DMM

The development of deep mixing method in Japan was initiated by the Port and Harbour Research Institute
(PHRI). The original idea may date back to a PHRI publication of 1968. The idea was then realized by Okumura,
Terashi and their colleagues in the early 1970's. Purposes of the research in the early days were two fold: one was
to investigate the lime reactivity of marine clays, and the other was to develop an equipment which enables
constant supply of hardening agent and uniform mixing at depth. Most of the marine clays tested easily gained a
strength of the order of 100 kPa to 1 MPa in terms of unconfined compressive strength. The equipment (Mark I to
Mark IV) was developed at PHRI. The first trial on the sea was done near Haneda Airport with Mark III which was
capable of improving only 10 meters from the water surface. Basic mechanism of the equipment has been
established by the trial. Construction control procedure was established with Mark IV by a test near-shore of
Nishinomiya. These initial development was continuously publicized through PHRI reports since 1972 and
Okumura and Terashi (1975) first introduced the technology to the international community. As the lime was used
in the initial stage, the method was named as "Deep Lime Mixing (DLM)". The first application was the
improvement of a reclaimed soft clay at Chiba prefecture in June 1974 in which Mark IV machine was used by
Fudo Construction Co. Ltd. In five years until 1978, DLM was practiced at 21 construction sites including two
marine works. In the search of improving the uniformity of improved soil, cement mortar and cement slurry took
over the granular quick lime soon. The method using hardening agents in the slurry form is now called CDM
method and referred to as the “wet method” of deep mixing.
Japanese marine clays are highly plastic and its natural water content is nearly equal to or exceeding the liquid
limit. In parallel with the development of equipment, Terashi et al. (1977, 1979, 1983) investigated factors
affecting the lime- and cement-treatment and studied the engineering properties of improved soils. They also
proposed a laboratory mold test procedure which was later standardized by the Japanese Geotechnical Society in
1990. Kawasaki et al (1981) and Saitoh et al (1985) investigated the improved soils extensively. Tatsuoka and
Kobayashi (1983) studied a influence of confining pressure and drainage condition on the strength and
emphasized the existence of residual strength .
As there is an extraordinary difference in the stiffness of treated and untreated soils, the design should base on
the understandings of the overall behavior of the improved ground as a system. Especially in the group column
type of application, horizontal external force in excess may lead to a bending failure of the treated soil columns in
a progressive way which actually happened in a couple of unreported cases. Based on these lessons, mass
stabilization of a block type, wall type and grid type have been employed in the port facilities which are normally
subjected to large horizontal forces. Group column type best fit for the settlement reduction. The investigation of
the improved ground was started with 1 g model tests, then with centrifuge tests and a full scale test (Terashi and
Tanaka, 1981, 1983; Terashi, Tanaka and Kitazume, 1983; Terashi, Kitazume and Yajima, 1985). In these days,
local Design Offices of the Ministry of Transport collaborated PHRI in developing design procedure. PHRI also
conducted a joint research program in the same line with the Takenaka Doboku. These works were synthesized
into the CDM manual by Japanese contractors group specialized in CDM method. In 1999, Coastal Development
Institute of Technology published a comprehensive technical manual focusing on marine application of deep
mixing method (Coastal Development Institute of Technology, 1999c). The technical manual is now being
translated into English and will be published towards the end of 2000.
Figure 8 shows the working vessels improving the sea bottom soft clay sediment by the wet method in order to
create an artificial island in the Trans-Tokyo Bay Bridge project (Uchida et al., 1996).

Figure 8 : Marine operation of Deep Mixing Method in Tokyo Bay

A research group at the Public Works Research Institute of the Ministry of Construction started the
development study of the similar technique in the late 70’s to the early 80's inviting the staffs of PHRI as advisory
committee members. The technique developed is called Dry Jet Mixing Method in which the dry powdered
cement or lime is used successfully as hardening agent instead of the slurry. This is the so-called "dry method of
deep mixing" . Although the general concept and applications of the dry method is quite similar to the wet method,
most of the application of the technique is limited on land work. The experience and recent research results are
compiled into another technical manual focusing on land work by Public Works Research Center (1999).
Figure 9 shows the DJM machines working on land. Due to the necessity of repairing Yodo River Dikes
collapsed during huge earthquake, deep mixing method was employed here to increase the liquefaction resistance
of the loose sand beneath the dike and improve its stability as well.

Figure 9 : On land use of Deep Mixing to improve loose sand at Yodo River

The development of deep mixing method in Nordic countries is reported in details by Rathmayer (1997). In
1967 a new method for stabilizing soft clay by unslaked lime was developed on the initiative of Kjeld Paus. The
method was named as the Swedish Lime Column method. Linden-Alimak AB developed a light, wheel mounted
mixing equipment in cooperation with the Swedish Geotechnical Institute, BPA Byggproduktion AB and Euroc
AB. A careful description of the Swedish method was first prepared by Assarson et al. (1974). Broms and Boman
(1975) reported this new technique to the international geotechnical community at the 5th Asian Regional
Conference which was the same conference Okumura and Terashi reported the Japanese equivalent. The first
design handbook was issued by Broms and Boman (1977). Recently Soil Reinforcement Committee of the
Swedish Geotechnical Society compiled the technical manual for the Swedish Lime and Lime Cement Columns
(SGI, 1997) although its coverage is the improvement below 150 kPa in terms of unconfined compressive strength.
Figure 10 shows the recent machine for dry method in Sweden.

Figure 10 : Swedish Lime columns with mixing tool (egg-beater at the tip)
A tremendous energy has been paid to improve the Deep Mixing Method in the past three decades almost
independently by two separate groups; one in Japan and the other in Nordic Countries, although there had been a
couple of personal visit to exchange information. By the support of ISSMGE TC-17, the world wide forum for
communication and exchange of ideas between both groups and other countries was opened in 1996 in Tokyo by
the 2nd International Conference on Ground Improvement Geosystems (Yonekura, Terashi and Shibazaki, 1996).
This was followed by two conferences in Stockholm and Helsinki (Rathmayer, 1999; Bredenberg, Holm and
Broms, 2000). Probably due to these exchange of information, wet method was quite recently introduced in
Nordic countries as well. As mentioned before, local standardization of the technology has completed to a certain
level of satisfaction. Next step may be an establishment of a regional or international codes in order to further
develop this technology and to transfer it to the other part of the world. Currently the WG 10 of the Technical
Committee CEN/TC 288 “Execution of Special Geotechnical Works” is drafting in order to standardize the
execution procedures of both dry and wet method of deep mixing. The working group comprises delegates from
nine European countries only. The authors believe that the work may be highly appreciated and recognized if the
draft incorporates the long experience in the wet method of Japan and recent American experience especially
gained at the Boston C/A Tunnel Project.

The first application of Deep Mixing in the North America may be in 1986 when Japanese based specialized
contractor. The growing demands of urban infrastructure development and rehabilitation have created a very
active and rapidly expanding market demand in the United States especially since the early 1990s. One of the
biggest application of Deep Mixing in America is that at Central Artery Project in Boston (Lambrechts and Roy,
1997; Druss, 2000). In parallel with the increase of applications, technology assessment has started in USA since
the middle of 1990s. Federal Highway Administration funded extensive review of world wide applications and
current state of the technology (Bruce, Bruce and DiMillio, 1998, 2000). Figure 11 shows the improved soil
exposed during excavation at the Boston project (Druss, 2000).

Figure 11 : A Shear key cut into the improved soil at the invert of Ramp D Tunnel, Boston (Dave Druss)

4.1.2 Recent Applications

Once a variety of equipment has been established and design procedures proposed, the practical applications
have exploded. Fig. 12 shows typical examples of diversified applications of deep mixing method in Japan. The
application shown at the upper left is the group columns installed beneath low embankments mostly for the
settlement reduction and improvement of stability in some cases. The upper right is an example of adjusting the
uneven settlement between piled abutment and embankment fill. The recent urban redevelopment projects often
require the excavation near the existing structures. The examples in the middle row show these applications.
Lower right and lower middle examples are for the gravity structure for port and harbor facilities which are
subjected to the huge horizontal forces due to wave or earth pressures. These are the application by the wet
method in which continuous improved soil mass are created by overlapping a number of treated soil columns
in-situ. The lower left is an example to tame the liquefaction by grid type, by which the lateral resistance of the
piles in the seismic conditions are improved. In Nordic countries, major purpose of improvement by lime and
lime/cement columns is to reduce the settlement at road embankments, bridge abutment, dwelling foundations and
so on. Also the technique is applied to increase the stability of embankments, excavated slopes and so on.

Figure 12 : Application of deep mixing in Japan

Fig. 13 is the statistics on the volume of soil improved by deep mixing in Japan. As shown in the figure, the wet
method is preferred in marine works, whereas both the dry and wet methods are employed for the land works. In
1993 for example, the total volume of treatment in Japan exceeds 4 million cubic meters. Fig. 14 is the recent
statistics both in Finland and in Sweden. In 1992 the total linear meters of installation exceeds 1 million meters
both in Finland and in Sweden.

Figure 13 : Annual volume of improvement in Japan

 Figure 14 : Annual volume in linear meter of improvement in Finland (left bar) and in Sweden (right bar)

Industries are still looking for new applications. Among them, a development of low strength deep mixing is
exemplified here. When a low strength of the order of 100 kPa is required by some reason, it is necessary to reduce
the amount of cement drastically in the case of the ordinary wet method, which ends up with low strength but at
the same time with non-uniform soil. The development is to apply new stabilizer which is composed of a low
amount of cement and a large amount of fly ash. The needs of this low strength deep mixing is explained in Fig.
11. Many projects of renovating populated sites locate on thick deposit of soft soils. In order to carry out braced
excavation between existing structures, costly work of driving sheet pile walls to a fairy deep reliable stratum is
necessary. The idea is to create an artificial bearing stratum floating in the thick deposit of soft soil as shown in the
figure. The requirement for an artificial bearing stratum is strong enough to support the sheet pile walls and at the
same time soft enough to allow the easy driving of sheet piles into the artificial layer. To realize this, the Electric
Power Development Company (EPDC) has developed a new hardening agent by recycling fly ash which is a
by-product of the company (Asano et al. 1996) and now carrying out a large scale experiment for confirming the
drivability of sheet piles as well as the behavior of sheet pile walls during excavation. Nikken Sekkei Nakase
Geotechnical Institute by contract with EPDC is running a series of centrifuge tests and analyses to develop a
design procedure (Ohishi et al., 1997).

Figure 15 : A new application of low-strength DMM

4.1.3 Prediction, Design & Construction

Details of recent applications and related problems are found in the proceedings of the specialty conferences on
deep mixing in Tokyo, Stockholm and in Helsinki. Looking the case histories in detail, we may arrive at the
following statement; “Deep mixing method both in Nordic countries and in Japan is the same in principle, but they
are quite local in its applications.” This means: when we look at the strength of improved soil, the wet method
creates the soil with strength exceeding 1 MPa in terms of unconfined compression strength. The dry method in
Japan mostly employed as group column type creates a soil with 500kPa. The lime columns are used ordinarily at
the strength less than 150 kPa. The difference in the strength naturally causes the difference of relative stiffness of
treated and untreated soils in the improve ground system, which will strongly influence the overall behavior of the
improved ground as a system. The further difference is that the treated soils in Nordic application are considered
as vertical drainage, whereas the ones in Japan are practically impermeable materials. The differences in the
properties of improved soils may be caused by the different soils’ property and conditions, different
manufacturing process, different climate conditions and different purpose of improvement. For each application,
design should be based on the rigorous understanding of the interaction of stiff treated and soft untreated soils
which leads to the rigorous understandings of the mode of deformation and mode of failure. Details of current
design procedures are described by Kitazume et al (1997) and by Swedish Geotechnical Society (1997). Although
the design concepts varies with the different applications, the flow of prediction, design and construction is nearly
the same as shown in Fig. 16.

Standardized Laboratory Test Data Base on

for representative soils by expected field strength and its uniformity
different binders relation of qu,f vs. qu,l

Estimate target field strength and

design strength

Assuming pattern of application

& dimension of improvement

Design of the Improved Ground

to satisfy the requirements

Field trial to confirm field strength Study site specific problems

& uniformity e.g. influence to nearby structures

Final Mix Design

Construction with QC/QA, Verification

Figure 16 : Flow of prediction, design and construction

When the requirements of the superstructure are given, the flow normally starts with the determination of field
strength often based on the accumulated experience on the similar soils and similar improvement cases. If there is
no experience on a particular soil, standardized laboratory mold test is conducted changing the type and content of
stabilizer. Then again based on the accumulated data base on the relationship of field strength, qu,f and laboratory
strength, qu,l, field strength will be determined. As the field strength scatters, design strength is determined based
on the accumulated data on the uniformity of field strengths.
Taking the requirements posed by the expected structure into account, the initial assumption is made on the
pattern of application (group columns or massive improvement) and dimension of improvement area. The mode of
deformation and/or mode of failure are dependent on the pattern of application and external load conditions. By
these we calculate the stability and settlement either by traditional way or by finite element by trial and error until
the satisfactory design is obtained. The solution is not unique. We may reach a couple of alternatives combined
with dimension and expected field strength.
As these design is started based on the estimated strength, field trial to confirm the actual strength and
uniformity at the real construction condition is carried out. Sometimes, laboratory mold tests are conducted in this
stage for the determination of the field trail test condition. The result of the field trial become a feed-back data for
the design. And we reach a final design of the improved ground as a system.
Finally we decide the mix design as well as QC/QA procedures for actual construction. And construction starts.
These are the current practice of deep mixing
As the basic principle of the admixture stabilization is the chemical reactions in the mixture, factors which are
not considered in the ordinary soil mechanics affect the field strength. Table 5 is a list of influential factors. The
first is the characteristics of hardening agent. The second is characteristics of the soil. Especially the type of clay
minerals , the type and contents of organic matter and water content are important. The third is a manufacturing
condition, especially the degree of mixing which changes from one equipment to another is important. The fourth
is the curing conditions. Temperature is related to the size of improved soil as well as to the amount of hardening
agent. In the laboratory mold test, there is no way to simulate III and IV except for the amount of hardening agent
and curing time. Thus, the laboratory mold test procedure is standardized by fixing the factors in these categories.
Therefore it should be accepted that the strength obtained by the laboratory tests is not a precise prediction but it
gives us an index for the actual strength. Therefore, in a design stage, the database plays an important role. Field
strength must be estimated through the empirical relations between lab. strength and field strength based on the
accumulated experiences.

Table 5 : Factors Affecting the Strength Increase (Terashi, 1997)

I Characteristics of hardening agent

1. type of hardening agent
2. quality
3. mixing water and additives
II Characteristics and conditions of soil
1. physical, chemical and mineralogical properties of soil
2. organic content, 3. pH of pore water
4. water content
III Mixing conditions
1. degree of mixing, 2. timing of mixing / re-mixing
3. quantity of hardening agent
IV Curing conditions
1. temperature, 2. curing time, 3. humidity
4. wetting and drying / freezing and thawing , etc.

4.1.4 Tasks for the coming decade

As is described in the previous paragraphs, the design of the deep mixing should base on somewhat
uncertain prediction of field strength. Once the improved soil were created and if the soil did not satisfy the design
condition, there would be no measure for further improving the already improved soils. Therefore, the current
practices ordinarily involve a large margin of safety. To improve the cost efficiency and to design more precisely,
the following tasks should be focused on in the coming decade. Some of them have already been answered in the
last specialized conferences in Tokyo, Stockholm and Helsinki, and the rest should be tackled in the coming
decade.
1 Understand properties of in situ improved soil & develop an appropriate failure criteria
The key issues on this subject may be; unconfined compression to triaxial compression, increase the precision
of testing, rigorous determination of design strength taking variability of in situ soils into account.
2 Understand clearly the mode of deformation/failure
Mode of deformation leading to failure is governed by such factors as geometry of improvement, relative
stiffness of improved and unimproved soils, loading condition typical for a specific application, interface
properties between structure and improved soil/ between improved and unimproved soils. These should be
examined through the full scale test or centrifuge model test.
3 Understand the influence of manufacturing process
 These may be investigated on each different manufacturing process and the issues may be property of lap-joint
face, and influence of disturbance on hardening process.
4 Influence to nearby ground and to existing structures
5 Establish a design procedure for new application
In order to save the cost of construction, application of slender soil wall is increasing. In such a case bending
failure should be taken into account as one of the failure mode. Similar is said to the group column type
application under large horizontal external force.
6 Develop new methods for the determination of the properties of in situ improved soils.
Application of geophysical method may be promising for the mass property
7 Develop new equipment appropriate for new applications
This has been done by the efforts of specialized contractors for example the combination of deep mixing and Jet
Grouting by one process.
8 Develop new stabilizer for changing needs
- control of strength and rate of strength increase - use of industrial by-products: fly ash/gypsum/slag/etc.

4.2 Consolidation by Preloading – Capability of two dimensional analyses of the improved ground
As the preloading with vertical drainage already has a long history, there are a number of text book and
technical manuals are available (for example, Hansbo, 1993). In most of the places, including Japan, the time for
consolidation is calculated based on the Barron’s equation taking into the smear effect (Hansbo, Onoue), well
resistance (Yoshikuni), and in some situation stress concentration on the drain (Yoshikuni). The authors do not
want to duplicate these in the current state of the art paper, but concentrate on the following two topics. One is the
recent capability of two- or three- dimensional FEM analyses of the improved ground and the other is the recent
development of Vacuum Consolidation.
Preloading with vertical drainage experienced nearly 50 years history. During the history of applications, a
number of negative case records were published in which the prediction and performance (predicted settlement
and actual settlement) were not in good accordance. Vertical drain shorten the drainage path, thereby accelerates
the pore pressure dissipation and accelerates the settlement as well if employed in the huge area (in one
dimensional situation). However, under the embankment load (in two dimensional situation), accelerated
dissipation of the pore pressure does not always accompany the accelerated settlement. A number of similar case
records are reported in Japan.
In the test section at the highway construction at Ebetsu, Japan, settlement and pore pressures were measured to
investigate the effect of vertical drain (Kurihara, 1982). The test site is divided into two. One is improved with
vertical drainage and the other unimproved. Figure 17 shows the time settlement relation. SD stands for the
measurement on the section improved by sand drain and N stands for the one on the unimproved ground.
Obviously no improvement effect was detected. Sand drain seems to be ineffective as far as the settlement is
concerned.

Figure 17 : Time – settlement relation measured at Ebetsu Test Site

 Figure 18 : Excess pore pressure dissipation measured at Ebetsu Test Site

Figure 18 shows the excess pore water pressure distribution at different time stage measured at the same test
section. the upper figure shows the excess pore water pressure of the improved section and the lower corresponds
to unimproved section. Although there was actually no difference in time settlement relation, sand drain is
effective in the dissipation of excess pore pressure.
Kobayashi (1990) back analyzed the case with his two dimensional computer code. Actual complicated soil
stratification was simplified and Modified Sekiguchi-Ohta model was used for clay. Figure 19 shows calculated
time – settlement relation. Two dimensional FEM could successfully explained the measurements. The computed
pore pressure dissipation also showed the similar tendency with the measurement.

Figure 19 : Calculated time – settlement relation for Ebetsu Test Site.

The routine design procedure of the vertical drain is in most case carried out by the Barron’s equation modified
to include such effects as smear, well resistance, sand mat resistance, difference of the permeability in horizontal
and vertical directions. However, when the width of the embankment is relatively small in comparison with the
thickness of the compressible layer, designers are advised to employ recent computer codes as well at least to
confirm his design.

4.3 Vacuum Induced Consolidation Process (VCP)

4.3.1 Introduction
Vacuum Consolidation (Figure 20) is an effective means for accelerating the improvement of saturated soft
soils while preventing pre-loading indeed formation failures. The soil site is covered with an airtight membrane
and a vacuum is created underneath it by using a dual Venturi and vacuum pump. The technology can provide an
equivalent pre-loading of about 4.5 m high as composed with a conventional surcharging fill.
Instead of increasing the effective stress in the soil mass by increasing the total stress conventional mechanical
surcharging, vacuum-assisted consolidation preloads the soil by reducing the pore pressure while maintaining a
constant total stress.

Figure 20 : Menard Vacuum Consolidation

The effectiveness can be increased when applied with combination of a surcharge fill. Field experience
indicates a substantial cost and time savings by this technology compared to conventional surcharging. This paper
presents the principle behind vacuum consolidation, the technological know-how and several case histories from
France for example where this technologies has been successfully implemented. The performance of the vacuum
consolidation system is evaluated based on settlement as well as pore pressure measurements at these case sites.
Vacuum consolidation was first proposed in the early 1950s by Kjellman (1952), the developer of the
prefabricated vertical “wick” drain. In the 1960s, The Corps of Engineers investigated the feasibility of vacuum
consolidation of hydraulic fill. Isolated studies of vacuum induced or assisted consolidation continued for the next
two decades (Haloton et al. 1965; Holtz 1975). However, except for specialized applications like landslide
stabilization, vacuum consolidation was not seriously investigated as an alternative to conventional surcharging
until recently due to the low cost of placing and removing surcharge fills and the difficulties involved in applying
and maintaining the vacuum. The steadily increasing direct and indirect costs of placing and removing surcharge
fill and the advent of technology for sealing landfills with impervious membranes for landfill gas extraction
systems have now made vacuum-consolidation an economically viable method as a replacement for or
supplement to surcharge fill. The current main application of vacuum assisted consolidation include:
• Replacement of standard pre-loading techniques, eliminating the risk of pre-loading induced foundation
failures.
• Combining VCP with water pre-loading in scare fill areas. The method has been used to build large
development projects on thick compressible soil.
• Combining VCP with surcharge pre-loading to increase foundation stability and thereby optimize
pre-loading stage sequence and reduce project time.
Field trials conducted over the past two decades in China (Choa, 1989; Liu, undated), France (Cognon, 1991;
Cognon et al., 1996), USA (Jacob et al., 1996; TETC, 1990), Japan (Shinsha et al., 1991), Bangkok (Woo et al.,
1989), Sweden (Tortenssen, 1984; Holm, 1996) and elsewhere have verified the effectiveness of vacuum-assisted
consolidation in conjunction with vertical drains for site improvement. Cost estimates based on these projects
indicate a significant potential for cost savings over conventional surcharge fill pre-loading for an equivalent
surcharge of 4.5m height.

4.3.2 Equipment and Construction Process

The Vacuum Consolidation construction process involves (Cognon et al, 1996):
1. Placing a free drainage sand blanket (60 – 80 cm thickness) above the saturated ground in order to provide
for a working platform.
2. Installation of vertical drains, generally of 5 cm in equivalent diameter, as well as relief wells from the
sand blanket.
3. Installation of closely spaced horizontal drains at the base of the sand blanket using a special laser
technique to maintain them horizontal.
4. The horizontal drains in the longitudinal and transverse directions are linked through connections.
5. Excavation of trenches around the perimeter of the preload area to a depth of about 50 cm below the
groundwater level and filled with an impervious Bentonite Polyacrolyte slurry for subsequent sealing of
the impermeable membrane along the perimeter.
6. The transverse connectors are linked to the edge of the peripheral trench. They are then connected to a
prefabricated module designed to withstand future pressure due to the vacuum.
7. Installation of the impermeable membrane on the ground surface and sealing it along the peripheral
trenches. The membrane is delivered to the site folded and rolled in elements of approximately 1000m2.
The membrane elements are welded together and laid in the peripheral trench where they are sealed with
the Bentonite Polyacrolyte slurry. The trenches are backfilled and filled with water to improve the tight
sealing between the membrane and the Bentonite Aquakeep slurry.
8. Vacuum pumps are connected to the prefabricated discharge module extending from the trenches. The
vacuum station consists of specifically designed high-efficiency vacuum pumps acting solely on the gas
phase in conjunction with conventional vacuum pumps allowing liquid and gas suction.
The process combines dewatering and vacuum action to maintain the water table at the base of the granular
platform during the entire application of the consolidation process. Eventually an additional drainage system is
installed at a required depth to allow for a conventional de-watering under the membrane. Indeed, the fill will
maintain a non-submerged action even when it has settled below the original ground water level. Therefore, with
this technology, unlike the case of a surcharge preloading, the load intensity will not decrease during the vacuum
application. The discharge drains are manufactured by extrusion of cylindrical and perforated PVC .Use of a
suitable filter cloth with proper filtering properties to cover the perforated PVC avoids infiltration of sand and
fines during vacuum application. The discharge drains are brought to the surface at every 150 meters spacing
where they are connected by transverse drains to the vacuum station as shown in Figure 21.

4.3.3 Conceptual Design

Vacuum-assisted consolidation provides an effective alternative to surcharging for pre-loading soils. Instead of
increasing the effective stress in the soil mass by increasing the total stress, using a conventional mechanical
surcharging, vacuum-assisted consolidation preloads the soil by reducing the pore pressure while maintaining a
constant total stress. Figure 21 presents a typical layout of a vacuum-assisted prefabricated vertical drain
consolidation scheme.
Figure 22 graphically portrays the initial total stress in the ground and pore pressure induced in due to: (a)
conventional surcharge, and (b) vacuum loading applied at the ground surface. Figure 23 shows a typical pore
pressure at the end of vacuum consolidation measured in China (Choa, 1989) during vacuum consolidation. The
Tianjin Harbor site consisted of about 4m thick hydraulic fill (thinly laminated silty clay/clayey silt) underlain by
15 m of silty clay, 3m of clayey loam, and another 3m of sandy loam, respectively. The site overlies a fine sand
deposit. The groundwater level was very close to the ground surface at this site. Vertical band drains
(100mm*4mm) were installed at a spacing of 1.3 m up to a depth of about 20-m into the clayey loam above the
sandy loam.
 Figure 21 : Schematic layout-Vacuum Consolidation

Figure 22 : Theoretical Pore Pressure and Vertical Stress Distribution during Surcharge and Vacuum
Consolidation (assuming 100% efficiency) by Cognon et al. (1996)

Figure 23 portrays two profiles of initial hydrostatic pressures and final pore pressures measured during
vacuum consolidation after 110 days and 180 days of vacuum application at the site. The straight lines indicate
theoretical pore pressures under various vacuum pressures.
An atmospheric pressure corresponds to about 100 kPa. For a site where the water level is at the ground surface,
cavitation of water at negative 1 atmosphere (gage pressure) theoretically limits on-land vacuum-consolidation to
an effective surcharge pressure of about 100 kPa, equivalent to approximately 6 m of surcharge. Practical
problems in maintaining the efficiency of a vacuum system may reduce its effectiveness in the field. A system
with an efficiency of 75 percent results in only 4.5 m of equivalent surcharge height.
 For the case presented in Figure 23, about 70 to 80 per cent efficiency is evident near the ground surface. At
large depths, however, the efficiency was only about 50 percent or less. The loss of efficiency with depth appears
to be related to the presence of relatively permeable layers at large depths. Vacuum consolidation in underwater
site conditions (off-shore land reclamation) can yield much higher equivalent preloads (Thevanayagam 1996).
In essence, geotechnical design analyses used to evaluate wick drain spacing, and strength gain for preload fills
are equally applicable to the engineering design of vacuum consolidation system. There are many technical and
operational factors, which play important roles in vacuum consolidation. Primary considerations governing the
effectiveness and economics of a selected VCP scheme include: (1) integrity of the membrane at the ground
surface, (2) seal between the edges of the membrane and the ground, (3) soil stratification including permeable and
seams within the clay deposit, and (4) depth to groundwater. Breaks in the membrane, a poor seal between the
membrane and the ground, and wick drains extending into layers of high hydraulic conductivity all tend to reduce
vacuum efficiency, reducing equivalent surcharge height and increasing pumping yields and pumping costs. The
success of a vacuum consolidation system depends upon a combination of technical know-how and careful
implementation of design details.

Figure 23 : Initial and final pore pressure data during Vacuum Pre-loading: Tianjin Harbor (Choa, 1989)

4.3.4 Case Histories

1. French Experience
Several soft ground sites consisting of different soil types including peat have been vacuum consolidated in
France. Cognon (1991) and Cognon et al (1996) presented a detailed summary of French experience. The sites
include highways, airports, oil tank farms etc. The depth of improvement at these sites ranged up to 12 m. Ground
settlements observed due to vacuum consolidation were about 10-20 per cent of the thickness of the soft layers, a
typical range encountered due to conventional surcharging. Ground improvement work at some of the sites in
France included placement of additional surcharge on top the airtight membrane to obtain a larger equivalent
preload. These field trials have proven the feasibility of vacuum consolidation combined with conventional
surcharge.
The advantages of vacuum assisted consolidation is well demonstrated by the pilot testing conducted in
Ambes, France, for the construction of a highway embankment on a very compressible saturated clayey soil
(Cognon et al, 1996). As shown in Figure 24a, the soil profile at this site indicates the presence of about 1.7 m
thick peat layer with a moisture content ranging from 400% to 900 % underlain by about 2.0 m thick highly
organic, compressible clay layer with moisture content ranging from 140% to 210%.
 The project plan called for the construction of a highway embankment, 2.15 m high, across the site in order to
protect the highway from floods. Alternative solutions, including soil replacement, raft foundation supported by
piles and conventional surcharge pre-loading were considered and rejected for economic reasons. For example,
due to stability concerns, the conventional pre-loading system required very gently sloping embankment with a
base width of 65 m, which was economically prohibitive. As vacuum consolidation has not been previously
applied under such soil and moisture conditions, a pilot field-testing program was implemented. The combined
application of surcharge (1.3 m) and vacuum provided an equivalent pre-load of about 150 kPa after only two
months of vacuum application, where as under conventional surcharge, slope failure would have occurred at this
site under a pre-loading stress of 35 kPa.

Figure 24 : (a) Typical Soil Profile at Ambes Site, France, and (b) Measured
Settlement vs. Time Record during Vacuum Consolidation (Cognon et al. 1996).

Figure 24 b depicts the monitored settlement record during the consolidation process. It indicates that the
recorded settlement with vacuum consolidation is approximately equivalent to that induced by a 4.5m high
surcharge embankment. Analysis of the settlement records indicated that at the end of the consolidation process
the settlement of the peat layer reached about 80% of the reference settlement that would have been induced by a
4.5 m high surcharge embankment, while the settlement of the underlying highly organic clay reached about 50%
of that reference settlement. At the end of the vacuum application the ground surface rebounded by about 3 cm
during 48-hour period and then stabilized. During the vacuum application the groundwater level rose by about 40
cm up to the level of the horizontal drains while the upper part of the granular fill remained dry.
Following the pilot testing the vacuum consolidation was selected as the best available solution for the
consolidation of about 17,550 m2 along this highway construction site.
The A837 Highway in the western part of France, crosses three (3) swamps. The site is divided into three parts
each distant of 15 km. The first two swamp crossings are 500 meters long (1640') and the third one 70 meters
(230') long. The thickness of the highly compressible material reaches 12 meters ( 40'). The final embankment
height varies between 4 and 9 meters (13' - 30') above grade. An initial study of the potential settlement had
indicated that approximately 2 meters (6') were to be expected for a 4 meter (13') embankment height over a period
of more than fifty years. Thus, preloading became necessary to accelerate the consolidation and to reduce the
post-construction settlement to a manageable magnitude. In view of the project size and the highly developed
agriculture tenure surrounding the proposed embankment, the transportation of additional fill for surcharge from a
long distance (50 miles) became economically prohibitive. The Vacuum Consolidation Process was thus selected
as the only viable alternative to Conventional Surcharge.
Figure 25 shows the settlement recorded during construction, first due to the vertical drains only, and then the
substantial increase as vacuum is applied. V.C.P. has permitted the construction of up to a 6-meter (20') high
embankment within a 3-month period. Total settlement during this period was 1.7 meters (5.5') and the residual
deformation anticipated for the next thirty years is estimated at 10 cm (4") only.

Figure 25 : A837 Saintes Rochefort Highway, France, Measured Settlement (Cognon, 1991)

2. Japanese Experience – Consolidation of a very soft reclaimed land

Recent field trials in Japan have verified the effectiveness of vacuum assisted consolidation to reduce the water
content of very soft ground sites reclaimed by dredged fill material or waste materials such as soda ash (Shinsha et
al., undated; Shinsha et al., 1991). The primary purpose of consolidation in these field trials was to assess the
potential for using vacuum consolidation to reduce the water content of slurry-like material disposed at temporary
storage areas and subsequent transport of the material for off-site disposal. The vacuum-consolidation
implementation schemes in these cases are significantly different from the above typical on-land applications.

Figure 26 : Schematic horizontal wick drain installation ( Shinsha et al 1991)

 Figure 27 : Water Contents Profile at Soda-Ash Site (Shinsha et al 1991)

Wick drains (l00mm x 3mm or 150mm x-l~ mm) were placed horizontally (or vertically) at a spacing of
0.7-1.5m within the dredged fill from a lightweight barge after fill placement (e.g. Figure 26 for horizontal
placement). Each of the wick drains is connected to a hose from the vacuum pump at the ground level. Unlike the
cases of on-land applications discussed earlier, the airtight membrane is not used in this system. The soft soil
above the top wick drain acts as an air-tight seal. In the case of vertical drains each of the wick drains is terminated
about 1 m below the top of the soft ground; a hose is attached to the wick drain at that point and connected to the
vacuum pump. Both methods of wick installation provided similar ground improvements.
Figure 27 shows the water content distribution prior to and: after application of the vacuum consolidation at the
soda-ash site. A significant reduction in the water content from about 300-400 percent to less than 200-300 percent
was achieved. In general, ground settlements up to about 40-50 percent of the thickness of the soft deposits could
be achieved by this method and therefore it can be used to significantly increase the storage capacity of disposal
sites.

4.3.5 Technology Assessment

The efficiency of this technology has been demonstrated under different site conditions where it has
successfully provided cost effective solutions to substantially accelerate the consolidation process while leading to
significant savings in project costs. Unlike the case of a conventional surcharge, VCP does not raise any stability
concerns, while resolving the environmental problems associated with the conventional method of surcharge
preloading. The vacuum consolidation technique is often combined with surcharge preloading either by placing an
additional backfilling surcharge or by using water placed at the top of the impervious membrane. The major
practical advantage of the vacuum consolidation is that it generates in the granular layer an apparent cohesion due
to the increase of the effective stress and the granular layer provides a useful working platform to accelerate the
surcharge backfilling process. Experience indicates that within days after vacuum pump is turned on, construction
vehicles can maneuver on the top of the membrane.
Vacuum consolidation is an effective means for improvement of highly compressible soft soils. In essence,
vacuum consolidation can yield an effective equivalent preload of about 4 to 5 m of conventional surcharge fill. A
combination of conventional surcharge with vacuum application can yield much higher equivalent preload.
Experience from US and China, and the case histories from France describe herein indicate that this technology
can be applied cost effectively under various challenging site conditions. In certain difficult site conditions where
the stability under the conventional surcharge is of concern, VCP allows to cost-effectively accelerate the
consolidation process as compared to conventional stage loading.
 In Europe, the engineering use of vacuum consolidation is currently rapidly expanding and it is of interest to
note that this technology has been used to cost effectively replace conventional surcharge preloading for the
development of about 57,000 m2 of industrial on land applications at the Channel Euro Tunnel Terminal.

Figure 28 : Vacuum Consolidation with Vertical and Horizontal Drains (Thevanayagam et al., 1996)

On-land applications are most suitable for soft soil sites with shallow ground water level. Presence of stratified
soils can render vacuum consolidation ineffective unless deeper vertical cut-off-systems are installed. Recent field
trials also indicate that on-land vacuum consolidation combined with dewatering can be an effective solution to
further accelerate the consolidation process.
Experience from on-land field applications of this technology indicates a high potential for use of vacuum
technology for improvement of existing hydraulic fills, strengthening weak sediments in the sea floor adjacent to
or beneath waterfront retaining facilities, and consolidation of fine-grained hydraulic fills during construction. The
technology holds technical merit, and can be cost-effectively compared to other conventional techniques. As
indicated by Thevanayagam et al. (1996) in the new land reclamation works the benefits of vacuum consolidation
can be realized by inclusion of prefabricated horizontal drains and selective placement of dredge materials.
Underwater applications or on-land applications with dewatering (Figure 28) appear to be most beneficial in such
cases. Lack of performance data on prefabricated drains as well as of field trials directly applicable for such cases
appears to limit its potential uses for further major land reclamation projects at present time.

4.3.6 Current Research and Future Developments

Current research is primarily focused on expanding the fields of VCP applications in particular for coastal and
land reclamation projects and addressing the current technical limitations. These limitations lie mainly in:

The practical needs for water proof membrane to be sealed to the soft soils thereby allowing for in-depth
soil consolidation, which is particularly useful for offshore application.

The need for exceeding the ultimate suction pressure limit of 1 ATM (0.1Mpa) and thereby efficiently
increase the VCP cost-effective range of applications.
In order to address these limitations a significant research has been conducted over the past decade by the
French Geotechnical firm Euro-Geomat (Robinet et al, 1997). The research involved the development, laboratory
testing and field demonstration of an innovative Hydraulic Exchange induced Suction Soil Consolidation System,
called the HELLOS TM consolidation Process (HCP).
The HELLOS TM Hydraulic Exchange induced Suction Soil Consolidation Process (HCP).
The HCP system, illustrated in Figure 29, allows, in any soil type, an application of suction pressures exceeding
100 MPa while eliminating the need for a water proof membrane to be sealed to the soft soil. Further, the method
consists of de-saturating the soil and thereby eliminating secondary compression and may therefore significantly
reduce post construction maintenance costs.
The high suction pressures are generated using a network of specially designed drains (Figure 30) in which air
flow has a pre-established relative humidity lower than 100%. The Kelvin's law yields the relation between the
suction and the relative humidity.

S= [(-ρwRT)/ Mw ]*ln (Hr/100) (1)

Where: Hr = Pv/Pvs is the relative humidity

Pv is the vapor pressure (Pa)
Pvs is the saturated vapor pressure (Pa)
S = Pa – Pw is the suction pressure (Pa)
Pa is the air pressure (Pa)
P is the water pressure (Pa)
Pw is the volumetric unit weight of water (Kg/m3)
R is the perfect gas constant
T is the absolute temperature (K)
Mw is the water molar mass (18.103 Kg/mol)
Pvs = I 10-3 [T -273] 4

1-Vertical drains;7-Van;8-relative humidity regulator;9-ventilated air

Figure 29 : Schematic view of the hydraulic consolidation device (Robinet, 1997)

1-Drain;2-Plastic barrier;3-Connector;4-Geotextile membrane;5-Water flow out

Figure 30 : Schema of a separated drain (Robinet, 1997)

Thus an increase of the relative humidity from 11% to 98% will lead to a decrease of the suction pressure from
360MPa to 2.7Mpa. The hydraulic consolidation is generated due to the induced suction field in the soil which
allows for water drainage and moisture exchange through the network of the perforated drains. The HELIOSTM
system is designed to maintain the air in the network of drains with the specified relative humidity using an Hr
conditioner (a moisture absorption unit) to dry the moisture accumulated in the circulating air due to the hydraulic
exchange with the soil by condensation of the water vapor.
Unlike Mechanical Consolidation Process (MCP) and Vacuum Consolidation Process (VCP), the HELIOSTM
Consolidation Process results in de-saturating the soil to a degree of saturation which is controlled by the relative
humidity of the air circulating in the drain. It therefore results in the maximum settlement that can be achieved
prior to the compression of the dry soil skeleton. Typical HCP hydro-mechanical consolidation/ de-saturation
curve of index (e) vs. induced suction pressure (P), illustrated in Figure 31, presents two phases of:
I) Consolidation of the quasi-saturated soil with volumetric change due to the hydraulic exchange generated by the
suction field.
II) The de-saturation due to a continuous hydraulic exchange with practically no volumetric change as the elastic
compression of the dried soil skeleton is negligible.

Figure 31 : Variation of the void ratio versus the HCP induced suction pressure for smectite consolidated at 5Mpa
(Robinet et al, 1997)

The de-saturation suction pressure (or the air entry pressure / bubbling pressure). Pb is a function of the
physio-chemical properties of the soil-water system and its capillary capacity.
The de-saturation phase will continue until equilibrium is reached while the gradient of the relative humidity
dissipates as the relative humidity in the soil pores becomes equal to the relative humidity of the air circulating in
the drains. Therefore depending upon the specified relative humidity this de-saturation process may involve both
the free interstitial water and the bonded water layer and could be most effectively used for decontamination of
polluted soils and dredge materials.
The System components
The HELIOSTM System, displayed in Figure 29, is designed to allow for the circulation of the air through the
network of drains with the minimum head loss of the vapor pressure for an effective control of the relative
humidity. It consists of:
• A network of perforated drains (Figure 30) surrounded by a geotextile filter liner. Drains are separated in two
parts to create an in-flow and out-flow double air circulation.
• The upper part of each drain is equipped with connectors, which allow to the air inflow through one part of the
drain and the air outflow through the other part.
• The vertical drain connectors of the same row are connected to a horizontal tube, which is connected to two
pipes on both sides of the drain at the soil surface
• The pipes at the soil surface allow the air to circulate inside the drains with one pipe being used for the inflow
of the air with a prefixed relative humidity while the other pipe being used for the outflow of the air, which has
increased its relative humidity due to the dehydration of the consolidated soil.
• . A fan provides a turbulent air flow (Reynolds number greater than 2500) to accelerate the air flow-
• A Hr conditioner (a moisture absorption unit) allows drying the moisture accumulated in the circulating air
due to the hydraulic exchange with the soil by condensation of the water vapor (Dehumidification). The van
pumps in the moist air and forces its flow through the Hr conditioner, where the water vapor is condensed The
liquid water is then collected for monitoring the hydraulic consolidation process
Hydro-mechanical tests
The testing protocol developed by Euro-Geomat consists of subjecting the soil specimen to a boundary suction
field using a pre-calibrated saturated salt solution and measuring the response of the soil specimen in terms of:
i) The oedometric uni-dimensional settlement - time relationship,
ii) The relative humidity (Hr) isochrones and temperature (t0C) in the soil using the Euro-Geomat Hr probe,
or psychrometers.(Olson et al, 1981)
iii) The relative humidity (Hr) and temperature of the air circulating at the contact interface with the soil
specimen
iv) The weight of water desorbed through the Hr conditioner (moisture absorption unit) using the
pre-calibrated saturated salt solution.
The tests are performed using the Euro-Geomat suction controlled high-pressure consolidometer (Robinet, et
al, 1997). The suction field is generated due to the controlled relative humidity of the air circulating in the system
in contact with the soil specimen at the top and at the bottom of the cylindrical sample. Due to the hydrau1ic
exchange at the interfaces the relative humidity of the air circulating in contact with the soil increases while the
relative humidity of soil porous decreases. The Hr conditioner (moisture absorption unit) allows to re-establish
and control the relative humidity of the in-flow air using a pre-calibrated saturated salt solution. The mass transfer
due to the moisture absorption in the Hr conditioner is monitored with an analytical balance and the relative
humidity of the air is monitored in the cell using the Hr probe.
Typical test results are illustrated in Figure 32 and 33. Figure 32 shows for Kalonite consolidated at 2.5 MPa,
the variation of the void ratio versus the suction pressure calculated from equation (1) for the relative humidity
monitored with the Hr probe at the contact with the soil specimen. The test involved four stages of loading
followed by a fifth stage of unloading. Figure 33 shows the variation of the relative humidity monitored
respectively with the Hr probe and in the soil specimen and the corresponding variation of the void ratio for the
fourth loading stage and the unloading phase. The results clearly demonstrate the two phases of behavior produced
by the HCP system. The first phase involves the hydro-mechanical consolidation under a relative humidity of 35
% at the contact of soil specimen with a volume change of the pores which at the center of the specimen are still
saturated (Hr soil center = 100%). The second phase involves a de-saturation due to a continuos hydraulic
exchange with practically no volumetric change, as the elastic compression of the dried soil skeleton is negligible.
Results of the unloading phase conducted under a relative humidity of 100 % at the contact of soil specimen
illustrate that the relative humidity in the soil specimen increased but remained lower than Hr = 100%.
Consistently as the center of the soil specimen did not reach saturation no rebound was observed during the
unloading phase.

Figure 32 : Kaloinite consolidated at 2.5 MPa; Variation of the void ratio versus the
suction pressure on drying-wetting path (Robinet et al, 1997)
 Figure 33 : Kaloinite consolidated at 2.5 MPa; Variation of the void ratio and the relative humidity
monitored with Hr probe and in the soil specimen during drying and wetting path (Robinet et al, 1997)

The HCP system invented by Robinet et al., (1997) appears to provide technological break through for
substantially increasing the effectiveness of vacuum consolidation process. However, full scale experiment of this
system are now required in order to assess the energy efficiency, advantages, and limitations of the HCP system
under actual field conditions.

4.4 Dynamic Compaction, Consolidation & Replacement

4.4.1 Introduction
Dynamic compaction is one of the oldest forms of ground improvements in existence. The Romans reportedly
utilized a variation of this technique and it was used in the United States as early as 1871. Although the dropping
of weight on the soil had probably been used for centuries, in 1970 Louis Menard patented the technique in France
and reintroduced to the profession under its present form, into sporadic use in the United States in the early 1970's.
The use of dynamic methods for the densification of granular fill is well documented in the literature, particularly
the technique of dynamic compaction. Useful information on techniques and equipment employed and ground
response to dynamic compaction may be found in Mayne, et al. (1984), Varaksin (1981), Liausu (1984) and
Findlay and Sherwood (1986).
As the availability of suitable construction sites decreases due to developments of the urban areas, the need to
utilize the sites with poor bearing and settlement characteristics for foundation support increase. Dynamic
compaction has proven to be an economical alternative to other available methods such as excavation and
replacement, surcharging, compaction grouting and other soil improvement techniques. One of the most common
and effective uses of dynamic compaction is to compact man made deposits of waste and rubble fills which are
frequently placed in the old quarries or clay pits, mine spoil, and landfills for both old and recent sites.
Dynamic compaction was developed and successfully used for densification of loose, saturated, cohesionless
soils and has proven to be particularly effective for liquefaction potential reduction. The densification process is
similar to that of vibro-compaction. Although used also in fine cohesive soils, its success in these soils is uncertain
and requires special attention to the generation and dissipation of pore pressures. On occasion, other ground
improvement techniques such as stone columns are used in conjunction with dynamic compaction (Bayuk and
Walker, 1994).
 Although developed for densification for loose natural soils, the majority of the dynamic compaction work in
the US has recently been performed at sites of solid waste, questionable or uncontrolled old fills and mine spoils.
Another common application in recent years has been the stabilization of collapsible soils which are stiff and dry
in their natural state, but lose strength and experience significant settlement when they become wet (Rollins and
Kim, 1994).

Figure 34 : Typical Dynamic Compaction Equipment (ASCE, 1997)

Table 6 illustrates soil classification based on the mechanism of compaction.

The method of dynamic compaction and replacement has been reported by Liausu (1984), and Findlay and
Sherwood, (1986) leading to the following definitions:
a) Dynamic Compaction: the compaction by heavy tamping of unsaturated or highly permeable saturated
granular materials. The response to tamping is immediate.
b) Dynamic Consolidation: the improvement by heavy tamping of saturated materials in which the response to
tamping is largely time dependent. Excess pore water pressures are established as a result of tamping and
dissipate over several hours or days after each tamping pass.
c) Dynamic Replacement: the formation by heavy tamping of large pillars of imported granular soil within the
body of the soft saturated soil to be improved. The original soil is highly compressed and consolidated
between the pillars and the excess pore pressure generated requires several hours to dissipate. The pillars are
used both for soil replacement and drainage.

Table 6: Classification Based on Mechanism of Compaction

Soil type Technology Ground Applications Comments
Improvement Target/
mechanism Improved properties
Cohesionless Dynamic Densification Relative density Reduce liquefaction
Loose sand compaction Internal Friction Potential
Cohesive clayey Dynamic Remolding & Shear Strength Thixotropic
soils consolidation strength gain properties
Cohesive soft Dynamic Replacement and Enhance settlement Improve soft soils
clayey soils Replacement drainage characteristics and between pillars
bearing capacity
Collapsible soils Dynamic Densification Relative density Break soil structure,
compaction minimize voids ratio
Waste Dynamic Densification Compression and Reduce the voids
Mine tails compaction oxidation reduction ratio, minimize
decaying problems
 Figure 35 illustrates grouping of soils by sieve analysis for dynamic compaction (Lukas, 1992). It is worth
noting that the method is suitable for pervious and semi-pervious soils with fine contents less than 20% (Zone I,
and Zone II). Zone III is relatively impervious soils such as clays and organic deposits. When these deposits are
saturated, excess pore pressures develop quickly but because of the low permeability, long periods of time are
required for dissipation, which make dynamic compaction impractical. Dynamic compaction has been successful
in zone II deposits, but the construction procedure has to be carefully planned so as to allow excess pore pressures
to dissipate between impacts. As a guide to categorize the soil deposits into one of the three categories described
above, either field permeability tests or laboratory grain size gradation tests could be undertaken.

Figure 35 : Grouping of soil by sieve analysis for dynamic compaction (Lukas, 1992)

4.5 Deep Dynamic Compaction

4.5.1 Basic procedure

The American Society of Civil Engineers (ASCE) ten years updates on deep dynamic compaction (ASCE,
1997) provides a review of the state of practice of dynamic compaction and its engineering applications.
Dynamic compaction is applied in a systematically controlled pattern of drops on a coordinate grid layout The
initial impacts are spaced at a distance dictated by the depth of the compressible layer, depth to groundwater, and
grain size distribution. Initial grid spacing generally approximates the thickness of the compressible layer.
Typically, 5 to 15 blows per grid point are applied.
Often, the proximity of groundwater or excessive crater depth limits the number of blows applied to each grid
to avoid getting the tamper stuck, or to allow for pore water pressure dissipation. Standard practice is to curtail
energy application when crater depth exceeds one and a half to two times the height of the tamper, or when the
groundwater surface rises into the crater. When this occurs, additional passes after ground leveling, or backfilling
the crater are required to complete the required number of drops.
This first phase of treatment is designed to improve the deeper layers. Incorrect spacing and energy level at this
stage could create a dense upper layer making it difficult or impossible to treat loose material below. The initial
phase is also called the "high energy phase" because the compaction energy is concentrated on a wider grid.
Completion of the high energy phase is usually followed by a low energy phase, called "ironing," to densify the
surficial layers in the upper 1.5 m (5 ft). Here, the tamper is only raised from 5 to 6 m (15 to 20 ft), and is dropped
on an overlapping grid.
 After each pass, the imprints are either backfilled with the surrounding materials or with off-site material. In a
situation where groundwater is at shallow depth, the craters should be backfilled with imported materials to insure
staying above the water table. At least 1.5 m (5 ft) is generally required between the tamping surface and
groundwater.
In saturated fine-grained soils, the process is complicated by the creation of excess porewater pressures during
compaction, a phenomenon which reduces the effectiveness of the subsequent compaction passes unless the pore
pressure is adequately dissipated. For clayey soils, dynamic compaction is generally not recommended unless the
craters are backfilled with crushed stone and repounded, creating large diameter columns of compacted stone
(dynamic replacement).

4.5.2 Type of soil improved

The single most determinative factor in the suitability of a soil type to be improved by dynamic compaction is
its ability to dissipate the excess pore pressure generated by the DC process. During dynamic compaction, soil
particles are displaced into a tighter configuration or a tighter state of packing. If water is present in the soil voids,
an instant rise in pore water pressure occurs. It is necessary for this pressure to dissipate before additional
densification can occur under repeated high energy drops. If this isn't allowed to happen, then repeated drops from
the tamper only cause displacement of the ground, and not densification.
As with the increase in applications of dynamic compaction over the last decade, the types of materials
treated by dynamic compaction have also increased dramatically. Originally, the predominant soil types
considered for dynamic compaction included only granular natural or fill soils. But because of the inherent
economic advantages involved with the use of dynamic compaction, a multitude of materials have been improved.
They include

Uncontrolled fills: Soil types within old fills can include the entire spectrum of natural soils, manmade debris,
byproducts, and any combination of the three. Dynamic compaction works best, however, on dry granular fills,
including sand, gravel, ash, brickbats, rock, shot rock, and steel slag.
Dynamic compaction in granular fills is similar to a Proctor compaction test, in that there is a physical
displacement of particles into a denser configuration. Dynamic compaction produces a low frequency vibration, in
the range of four to ten cycles per second, and it is this low frequency excitation along with this input of impact
energy that reduces void ratio and increases relative density resulting in improved bearing capacity and enhanced
settlement characteristics
For deposits below the water table, the vibrations cause an increase in pore pressure, and after a sufficient
number of surface impacts, cause a sufficient rise in pore pressure as to induce liquefaction, very similar to the
process occurring during earthquakes. Once this occurs, additional energy application is ineffective until the pore
pressure dissipates. Additional pounding following pore pressure dissipation produces more low frequency
vibrations that reorganize the particles into a denser configuration.
Dynamic compaction has been used more recently to improve fine-grained fills as well. These fills are much
more difficult to improve, and require much tighter field control and experience. Clays and silts tend to "heave"
after repeated pounding, and if additional pounding continues, can have a detrimental effect on compaction. If
heaving occurs, pounding at that point should stop, and the number of passes should be increased with either a
reduced drop height or fewer drops per point
A more common technique that has been increasingly employed in the US over the past decade to improve
fine-grained sites is dynamic "replacement" technique. This technique consists of producing a crater by
conventional heavy tamping, and filling the craters with a "boney" or granular backfill material to create in-situ
highly compacted large diameters granular pillars, which is either floating or driven to a firm strata.
This boney material can be gravel, shot rock, brick bats, reprocessed concrete, or anything that will lock
together under additional heavy tamping. Because of the higher permeability of this backfill, pore water pressure
from the underlying and adjacent fine-grained soils will dissipate more quickly. This process is repeated until a
noticeable decrease in crater formation occurs. This technique essentially results in large diameter columns of
compacted stone underlying a site or individual column locations
Dynamic compaction is often used in conjunction with other ground improvement techniques. A retail site in
New Jersey was constructed over an old fill which was underlain by organic soils is an example. Here, a vibroflot
was used to install stone columns at each interior footing location, and then the surface deposits at each of these
column locations was dynamically compacted (Bayuk and Walker 1 994)
 There have been several old steel mill sites that have been underlain by steel slag (Troy, NY; Youngstown, OH;
Trenton, NJ, St. Louis, East Chicago, IN) Steel slag is generally quite granular, and responds very well to dynamic
compaction
Municipal Solid Waste (MSW): Post-construction settlements of sanitary and rubble landfills under
embankments are difficult to predict. Without site improvement, settlements can sometimes range from 1.5- 4.6 m
(5- 15 ft). The main causes of settlement in landfill deposits are due to:

• Mechanical compression due to distortion, reorientation of the materials under self-weight,

• Biological decomposition of organic wastes,
• Physio-chemical change such as oxidation, corrosion, and combustion,
• Ravelling of fines into larger voids

Dynamic compaction has been used extensively on MSW to remediate the above causes and for a multitude of
reasons. Again, experience is essential in improving MSW, in that grid spacing, weight contact pressure, and
number of passes are crucial in achieving the desired results. Highway embankments, roadways, parking lots, and
even retail structures have increasingly been constructed on dynamically compacted MSW.
In sanitary landfills, settlements are caused either by compression of the voids or decaying of the trash material
over time. Dynamic compaction is effective in reducing the void ratio, and therefore reducing the amount of
immediate and long-term settlements after construction. It is also effective in reducing the decaying problem,
since collapse of voids means less available oxygen for decaying process. Future settlements, however, can still be
expected due to a secondary consolidation process, and future decaying of the trash material.
A distinction must be made between older landfills and more recent landfills when considering the long-term
settlement of the landfill after improvement with dynamic compaction. Organic decomposition has generally
already taken place in older landfills, and the landfill usually consists of a dark-colored soil matrix containing
varying amounts of bottles, metal fragments, wood: and debris. Decomposition generally takes more than 25 to 30
years to occur.
For deposits where biological decomposition is complete, dynamic compaction has its greatest benefit.
Densification results in higher unit weight and reduction incompressibility under load with little long-term
subsidence under load.
For recent landfills where organic decomposition is still underway, dynamic compaction increases the unit
weight of the soil mass by collapsing voids and decreasing the voids ratio. It will not however, stop the biological
decomposition, which may result in a loosening of the soil structure followed by long term settlement.
Coal Mine Spoil: Drumheller and Shaffer (1997) discussed 19 coal spoil sites in the US that have been improved
by dynamic compaction. Dynamic compaction methodology in coal spoils varies with the consistency of coal
spoil. Some spoil are predominately shotrock with minor amount of cohesive material, whereas some spoils have
much higher concentration of fines.
Contact pressure of tamper, size of weight, and grid spacing are generally considered important factors in coal
spoils. Depending upon the nature of the proposed structure, dynamic "replacement" is sometimes used following
the DC area pass at column locations to further reduce the risk of intolerable settlement.
Collapsible Soils: Rollins and Kim (1994), Drumheller and Shaffer (1996), and Davis (1996) discussed eleven
sites in Western States where dynamic compaction was used to improve collapsible soils. Settlement associated
with collapsible soils can lead to expensive repairs, either in highway or structure construction.
In 1982, FHWA conducted an extensive field test program of various ground improvement techniques to
improve collapsible soils in New Mexico. The various techniques included vibroflotation, deep mixing,
pre-wetting, and dynamic compaction. Dynamic compaction was found to be the most cost effective, and was
selected to improve three separate sections of I-25 and I-40 around Albuquerque.
Liquefiable Soils: dynamic compaction is a useful ground improvement tool to reduce liquefaction potential as it
increases the relative density as well as lateral earth pressure. Dise et al., (1994) with the US Bureau of
Reclamation, discussed three large projects where dynamic compaction was used to remedial liquefiable soils and
improve the seismic stability of several embankment dams.
20 to 30 tone tampers were employed at all of these locations. Wick drains and surface drainage were installed
in conjunction with the dynamic compaction at the sites.

4.5.3 Dynamic Consolidation

For soft cohesive soils, the densification of soil following heavy tamping is attributed to
 (a) Compressibility of saturated soil due to the presence of micro-bubbles;
(b) The gradual transition to liquefaction under repeated impacts;
(c) The rapid dissipation of pore pressures due to high permeability after soil fissuring;
(d) Thixotropic recovery.
With successive tamping, energy is imparted to the soil, a certain amount of immediate volumetric strain is
mobilized, and excess pore pressure is generated. The level of energy input into the system is called the 'saturation
energy' when the pore pressures equal 100% liquefaction pressure. No further volume change can be achieved by
imparting additional energy to the soil. Dissipation of pore pressures with time leads to consolidation and gain in
strength of the soil. The process of densification under a number of passes with time delays between each pass can
be visualized from Fig. 3. The background of the analysis of the heavy tamping mechanism has been described in
detail by Van Impe (1992).
For low-velocity impacts on soft cohesive soils the impact energy is used efficiently to improve the soil only in
a thin layer. If the impact energy is very high, as in the case of common dynamic consolidation of normally
consolidated soils, the depth of influence and the final compacted density are greater, although the energy partly
dissipates due to radiated longitudinal stress waves.

Figure 36 : Heavy tamping principle for cohesive soils (Menard, 1974)

Beneficial effects such as inhibiting heave and greatly increasing the impact efficiency have been obtained
recently in Belgium by the Soils company patented impact block, capable of extending the duration of the pulse on
the soil being treated, towards a more 'plastic collision', and allowing implementation of variable block stiffness by
prestressing the anchors (Van Impe, 1992). The extent to which heavy tamping improves the in situ soil is one of
the primary parameters studied.

4.5.4 Design and Analysis Considerations:

The design of a dynamic compaction project involves determination of tamper weights, grid pattern, drop heights,
and depth of influence. The following section briefly discuss theses design parameters (Lukas, 1986; Menard and
Boris, 1975; Van Impe et al, 1997; Mayne et al, 1984).

Depth of Improvement, D, Prediction of the depth of influence and the level of improvements are the primary
concern when using the dynamic compaction method. These, however , depend on several other factors which
include: the soil conditions, energy per drop, the contact pressure of the tamper, grid spacing, number of passes
and the time lag between each pass.
Impact Energy, E, The energy induced by the dropping of the tamper is simply the weight of the tamper times the
height of the drop. These represent the main design parameters in determining the depth of improvement when
using dynamic compaction. Menard and Broise, (1975) proposed that depth of influence was simply proportional
to the square root of the energy per blow, the equations was modified later by Lukas, 1986.

D= n (WH)0.5 (1)

Where,
D = Depth of Influence (meters)
W = Weight of Tamper (tonnes)
H = Height of Drop ( meters )
n = empirical coefficient that depends on the type of soil (0.3 to 0.6)
This equation is based on the free falling of the weights.
The factor n, is to account for the applied energy, tamper contact pressure, influence of cable drag, presence of
energy absorbing layers and ground water table. Table 7 lists the proposed values of n, for applied energy with the
range of 34 to 100 ton.ft/ft2. Figure 37 graphically shows the range for various case histories (Leonard et al, 1980).
The grid spacing is related to the impact energy by the following equation,

E = (NWHP)/S2 (2)

Where, E is the average applied energy over the treated area, N, is the number of drops, P is the number of
passes, and S is the grid spacing. Lukas (1986) ves typical impact energy values per unit volume of treated soils.
These values can be multiplied by the thickness of the treated soil to estimate the required applied energy at the
surface. The estimated energy is used in the above equation to determine either the number of drops for a specific
spacing or the minimum spacing for a particular number of drops. The grid spacing usually used is about 1.5 to 2.5
times the dimensions of the tamper (Munfakh, 1997)

Figure 37 : Range of n from various case histories

Table 7 : Value of n for deriving the depth of influence D of heavy tamping: D = n (Wh)1/2
 Reference n
Menard and Broise (1975) 1.0
Leonard et al. (1980) 0.5
Bjolgerud and Han (1963) 1.0 (rockfill)
Smoltcyk (1983) 0.5 (soils with unstable structure)
0.67 (silts and sands)
1.0 (purely frictional sand)
Lukas (1980) 0.65 – 0.8
Mayne et al. (1984) 0.3 – 0.8
Gambin (1984) 0.5 – 1.0
Qian (1985) 0.65 (fine sand)
0.66 (soft clay)
0.55 (loess)
Van Impe (1989) 0.65 (silty sand)
0.5 (clayey sand)

Influence of Cable Drag, since dynamic compaction is a repetitive process, substantial amount of time is required
to manually rehook the weight after each drop. As a result, drops are conducted with the cable attached to the
tamper This however poses another problem due to the influence of cable drag which is due to friction of the cable
unwinding over the spool drum and reduction in tamper velocity due to air resistance. Lukas (1992) bases his
observations on five separate dynamic compaction projects indicating that whenever tampers are reused and
dropped with a single cable with a free spool, the measured velocity was found to range from 0.88 to 0.93 of the
theoretical velocity. Results were, however, encouraging since the influence of the cable drag on the energy
applied was relatively constant and, hence, does not have to be measured for each equipment or tamper weight.
Equipment Limitations, The type of equipment used will also have an effect on depth of influence. Conventiona1
crawler cranes with a rated capacity of136 tons are commonly used for dynamic compaction for drop heights up to
24 meters. The usual mass of tamper used however is in the order of 10 to 20 tons with drop heights usually
ranging from 10 to 20 meters. Higher drop energies have been achieved with tamper masses of up to I50 tons and
drop heights as high as 40 meters with special cranes or tripods.
Influence of Tamper Size, Tamper size is instrumental in controlling the contact pressure at impact. Contact
pressure which is defined as the weight of tamper divided by the contact area is commonly used in the range from
30 to 75 kN/m2. Low contact pressure could develop a crust of soil and prohibit any soil improvements below this
crust Conversely contact pressure higher than those indicated above could result in the temper punching into the
ground upon impact, which reduces energy efficiency.
Grid Spacing, The print spacing (the spacing between the compaction points) used in dynamic compaction has a
significant effect on the soil improvement within the grid (Chow, et al, 1994). The first pass is designed to improve
the deeper level, and is dependent on the thickness of the compressible layer, grain size distribution and depth of
the groundwater. Initial grid spacing is usually at least equal to the thickness of the compressible layer. Other
passes that follow are aimed at densifying shallower level, which may also require lesser applied energy. Finally,
an "ironing" pass to densify the top layer of the ground is conducted by dropping of a square or rectangular tamper
over the entire surface area with relatively low drop energy.
Time Delay between Passes, Where pore-water pressure can develop, the timing between each pass must be such
that it will allow for the pore-water pressure to dissipate Piezometers can be installed to monitor the dissipation of
pore-water pressure following each pass
Soil Conditions, As described before, dynamic compaction is best suited for densification of deposits grouped as
pervious and semi-pervious (Lukas 1986). In addition, the position of the water table and the amounts of fine
contents generally influence the effectiveness of dynamic compaction. Presence of clay content greater than I5%
fines by weight, generally renders this method less effective (Luongo, 1992)
Degree of Improvement, Main factor controlling the degree of improvement is the applied energy. Menard and
Broise (1975) stated that the applied energy used for Dynamic Compaction, should produce a minimum static load
of 2 - to 3 t/m2 at the depth corresponding to the water table level. Other suggestions were also made for magnitude
of the applied energy. Lukas (1992) based his recommendations on different types of soil conditions as shown in
table 7.
 Although increasing the number of drops in each pass and the number of passes can be used to achieve a greater
depth of influence, there is a point for which the further application of energy produces only minimal gains. This
threshold is called the "saturation energy". Lukas, (1992) presented a typical graph of depth of improvement
versus the number of drops. Depending on the soil type, increase in the number of drops will have very little gain
in improvement of the lower levels.
For cohesive soils, Charles et al. (1981) proposed an influence depth which takes into consideration other
parameters such as soil type, surface area and shape of the pounder.

D = 0.4 (EdB/Apcu)0.5 (3)

where B is the width or diameter of the pounder, Ed/Ap is the total impact energy applied per unit area of the
pounder and Cu is the undrained shear strength of the soil.
An exhaustive compilation of data from over 120 sites was presented by Mayne et al. (1984). Moreover, useful
correlation for normalized crater depth, D/(WH)0.5, overall subsidence of the ground, peak particle velocity, and
maximum depth of influence, D, have been presented. The normalized crater depth increases with the number of
passes (Fig. 5a), the trend showing a limit for this parameter. The overall ground subsidence increases with applied
energy (Fig. 5b), while peak particle velocity decreases (Fig. 5c) with scaled distance, d. The maximum depth of
influence is proportional to the energy per blow.

(a)
 (b)

(c)
Figure 38: a) normalized crater measurements (Mayne et al, 1984); b)observed magnitude of ground
subsidence with level of applied energy; c)attenuation of ground vibration (Van Impe et al, 1997)
Site Preparations, The site to be consolidated must first be prepared to support the weight of the tamping machine
(60- 200t). Occasionally 1 to 2 meters of granular materials are applied to the ground surface, particularly in
landfills and other soft ground conditions, to provide bearing surface for the machine. It must also be safeguarded
against bad weather. If sensitive to rain water (alluvia and clays ), and removal of water rising to the surface during
the consolidation process must be facilitated by means of peripheral trenches, drains, and so on.

4.5.5 Environmental Considerations

One major concern with heavy tamping is the high impact energy that generates damaging ground vibrations
The use of such methods of ground improvement is therefore much more limited in the urban surroundings.
Ground vibrations caused by dynamic compaction not only can be damaging to the nearby structures, under
ground utilities and electrical or mechanical equipment, and are also disturbing to people.

Figure 39 : Variation of Peak Particle Velocity with Tamping Energy

Vibrations are normally quantified in terms of the Peak Particle Velocity (PPV). Peak particle velocity is
generally used to define the damage criteria for buildings and the annoyance tolerance levels to people. Figure 39
shows the relationship between particle velocity is and scaled energy factor. It illustrates that well-constructed
buildings can tolerate a PPV of 50 mm/sec, however, a limit of12.5 mm/sec is often used as a maximum value for
safety margin. To facilitate comparison between various projects, the peak particle velocity has been plotted
against the inverse scaled distance as shown in Figure 39. The inverse scale distance is the square root of
compaction energy, (WH)0.5, divided by distance from the impact point.

4.5.6 Other Design Considerations

Many dynamic compaction sites have irregular subsurface conditions particularly boulder and rubble deposit
where it is difficult to interpret verification test results and assess soil properties after dynamic compaction
improvement. Based on extensive laboratory model study of dynamic compaction of dry sand with measurements
included tamper acceleration and soil pressure during impact, a procedure was presented by Poran et al (1992) to
use Dynamic Settlement Modulus (DSM) to determine the degree of improvement during construction. DSM was
defined as the slope of the tangent of the loading portion of impact stress-relative settlement curve where relative
settlement, εj, is defined as the tamper settlement dt, divided by its diameter , D.
 DSM = ∆pt / ∆(dt/D) (4)

where, pt is the impact stress.

Also, DSM values is a function of tamper drops, As it can be observed, for most of the tests, the rate of DSM
increase was significantly reduced from the 12th drop on This relative change in the DSM values was found to be
proportional to the rate of densification (density increase as a function of Number of drops).
Although not widely used, other design methods have been developed in order to make a better prediction of
the effect of factors affecting dynamic compaction and the required or the desired level of ground improvement -
Lo, et al (1990) introduced a DC design method based on their findings that a relationship exists between the
saturation energy and enforced settlement. The authors presented a plot (Figure 40) of enforced settlement versus
total applied energy intensity, pointing out that for a given initial soil consistency and energy per blow of pounder
, a hyperbolic curve may reasonably be fitted to the field results Thus, a saturation energy can be defined beyond
which further enforcement of settlements would be relatively insignificant. Furthermore, saturation energy
intensities were plotted against the ratio of energy per blow of tamper EB (energy per blow) to the initial
pressuremeter limit pressure PL of each site. As shown in Figure 41, a unique relationship may reasonably be
constructed between parameters Is (saturation energy) and EB/ PL.
The authors also emphasized that in view of scarcity of well-documented tamping projects to draw upon, the
assumed trends for Is in Fig 8 have been inferred on the basis of rather limited data. Additional research would no
doubt refine the results, establishing more reliable design curves.
Further, since Is is uniquely determined by EB/PL it should in principle, be reasonable to expect that a collective
term of these parameters relating to the operational requirements of each tamping project would similarly
determine the maximum degree of ground improvement uniquely. This collective term or operational factor
maybe expressed as

.Ω = Is (EB/PL) (5)

where.

Is = Saturation Energy (t-m / m2)

EB = Energy per Blow (t-m)
PL = Pressure meter Limit Pressure (t/m2)
The maximum degree of ground improvement maybe expressed in terms of enforced settlement per unit
thickness of treated soil deposits as follows:

η = SE/Ht (6)

where, Ht = total initial thickness of soil deposits requiring treatment According to Figure 41, the relationship
would then be applicable.

η = Ω/(30+3Ω) (7)

Thus, given the initial ground conditions, it may in principle be possible to specify a priori, with the characteristics
of Figure 41, energy per below of pounder print spacing, and number of below per print to achieve the required
degree of improvement, and thereby rationalize perforn1ance design Furthermore, due to the subsequent
thixotropic recovery and dissipation of pore-water pressure, long terms effects of heavy tamping might result in
greater ground improvement than inferred herein.
 Figure 40 : Enforced settlement due to heavy tamping (Lo et al, 1990)

Figure 41 : Saturation energy, Is, versus scaled tamper energy per blow

4.5.7 Dynamic Replacement

Dynamic replacement is based on the same mechanism of densification as dynamic compaction and utilizes
similar plant, essentially a tall rig with a drop weight. It further incorporates ground replacement techniques by
progressively filling the crater, or "prints" formed by the drop weight with rock fill to create columns of strong,
heavily compacted material. The columns greatly improve the average stiffness of near surface zone, as well as
better transmitting the impact forces from the drop weight to greater depths within the fill. The basic processes
involved in dynamic replacement are indicated in Figure 42
 Figure 42: Typical Dynamic Replacement process to create stone pillars (Liausu, 1984)

The main design parameters for this technique could be summarized as (Barksdale and Bachus, 1983), i) shear
strength of the composite mass of vertical compacted granular material within the existing soft soils; ii) stress ratio
defined as loads transferred to columns to that transferred to adjacent soils, which will be a function of the soft
material stiffness (cohesion), and the volume of the replaced material per unit volume of the soft soils, iii) vertical
capacity of the columns, governed by the diameter and allowable confining pressures provided by the surrounding
soils; iv) drainage properties of the replaced soils and their influence on the consolidation and strength gain of the
soft soils.
The main draw back of dynamic replacement is that the improved soil may experience 40 % of the estimated
settlement without improvement due to the fact that soils between columns, while improved through the
compaction induced lateral compression are not treated in the same manner.
As reported by Bevan, 1997 a major dynamic replacement project for a multi-story building was carried out in
Iran for ground improvement to enhance settlement and strength characteristics of soft sandy clayey soils.
Boreholes revealed either very silty collapsing sandy soils above ground water or soft compressible sandy clay
below ground water table overlying a marl bedrock at a depth between zero to 12 meters below finished facility
ground level. Low standard penetration test (SPT) blow counts are observed just above ground table. Ground table
varied between about 3 to 4 meters below finished floor level. There appeared generally to be a stiff surface crust
across the whole site with loose to very loose horizons just above and below the ground water table. The initial
proposal was to improve the alluvial deposits by Dynamic Compaction. However, due to the high content of fines,
mainly the clayey fraction, the process becomes dynamic replacement. In this particular site, plugs of stone,
consisting of imported backfill, which is repeatedly filled into the craters by the pounder, were driven into the soil.
These plugs were reinforcing the existing soil to a certain extent as load supporting elements. They reduced the
settlement, increased the bearing capacity and accelerate the consolidation process.
To test the effectiveness of the spread treatment from the tamping points, tests were carried out with various
number of drops and CPT's made at the center of the print and at 1,2,3,4, and 5 meter from the center of the print.
Figure 43 illustrates that the effect is concentrated in the layer above the water table and at the center of the
tamping print area in the plan. Disruption of the original surface crust between 0 and 2 meter can be observed,
which would need careful restoration in ironing pass.
 Figure 43: CPT results before and after dynamic replacement; a) results at and around the pillars

Field Monitoring and Quality Control

As noted before, due to various factors affecting the design and the process to achieve the desired results, tests
must be conducted during the dynamic compaction program to verify the selected operational and design
parameters. Usually a minimum degree of improvement is established for a dynamic compaction project based on
some type of a test which should be followed when testing the dynamically compacted soils.
Generally site investigation and verification testing may include one or more of the following tests: Standard
Penetration Test (SPT), Cone Penetration Test (CPT), pressuremeter, dilameter and plate load tests, and,
occasionally, down hole and cross hole shear wave velocity measurements and spectral analysis of surface waves
(SASW). Figure 44 illustrates a typical compaction control of municipal waste using SASW (Haegeman and Van
Impe, 1995). The dispersion curves measured before and after dynamic compaction are presented and compared
with the theoretical predictions (crossed). The general trend is that the shear wave velocities after compaction
significantly increased at depth with respect to initial values However, in many such sites where soil conditions are
irregular and hard to sample (such as in boulder and rubble deposits), site testing is difficult and the results are
often confusing and misleading
 Figure 44 : Rayleigh wave dispertion curves before and after compaction (Van Impe et al, 1997)

The control during the operation is carried out with

a) penetrometer and pressuremeter for measurement of the shear strength and compressibility. In fine-grained
soils, these tests are greatly influenced by the delay at which they are carried out after each pass; for final
control, a minimum delay of several weeks should be observed,
b) numerous piezometers placed at different elevations to determine the minimum delay between each pass,
c) gamma densitometer and water content measurement of samples to check on the variation of dry density layer
by layer,
d) topographical measurements of the ground surface for determination of overall variation of dry density.
Other factors which may affect testing results following dynamic compaction is the phenomenon called
thixotropy which refers to the increase in the shear strength of fine soils after the pore-water pressure, caused by
freeing the absorbed water, is dissipated over time.
Figure 45 illustrates a schematic of QC system. The system is designed (Poran et al, 1992) to have an
interactive, instantaneous data processing capability. The system includes rugged accelerometers connected to
central unit (CU) also mounted on the tamper. The wiring are highly protected in special impact resistance tubes.
The CU consists of several components including short-range FM transmitters connected to remote monitoring
station. A computerized data acquisition system (DAS) consisting of multi channels is used to display and reduce
data. A surveying unit is included for settlement measurements and other topographies. However, the main
concern of such systems is durability, it has to withstand repeated high acceleration shocks and harsh environment
of DC field operation.

4.5.8 Research and Development for DC

Dynamic compaction method makes use of the free fall energy of the heavy ram. On the other hand Kanatani et
al, 1997 proposed a new dynamic compaction technique making use of not only the free fall energy but also of the
rotational energy of the ram for the purpose of the advancement of the densification of the granular soil ground
(Yoshida et al, 1995). The authors conducted a reduced scale experiments to simulate the usual dynamic
compaction and the proposed alternative with a rotational ram. The effectiveness of densification of the sandy
ground achieved with both methods was compared from test results. Details of the experiment are illustrated in
Figure 46. Very loose artificial sandy ground with relative density of 10% was made in a rectangular container
whose depth and width were 110cm and 150cm respectively.

Figure 45: Schematic layout of QC system for dynamic compaction

Figure 46 shows the mechanism of hanging and rotation of the ram. Cone penetration tests were conducted
before and after tamping at several points. In addition to cone penetration test, the settlement of the ram, the
vertical displacement of the ground surface were measured. Figure 47 shows typical test results of cone resistance
before and after tamping and at different locations.
Based on this study, the main conclusions were: (i) increments of cone penetration resistance of ram dynamic
compaction were generally larger than conventional DC; (ii) settlements of the ram for RDCM were about as
twice as large as those in conventional DCM.
Figure 46 : Reduced scale model for Rotational dynamic compaction (Kanatani et al, 1997)

Figure 47: Comparison of normalized increment of CPT

4.5.9 Summary and Recommendations
1. Dynamic compaction today is a well-established ground improvement technique, which offers a versatile
engineering tool for a variety of geo-construction applications. In particular the technique has become a cost
effective solution for turning municipal solid waste disposal into valuable construction sites.
2. The application of dynamic compaction especially in fine grain soils requires field monitoring for process
optimization. Current trends in both research and deployment indicate, the potential benefits of using SASW
techniques for interactive user friendly quality control system which may significantly increase the efficiency of
DC during construction.
3. Current research strongly suggest a combination of dynamic compaction and rotational / vibratory process
presents a strong potential for substantially improving the efficiency of dynamic compaction. Such combination
can offer a flexible solution which may be adopted to specific site conditions and better address environmental
concerns in major urban centers. However, field testing are necessary in order to assess and demonstrate the
efficiency of such a hybrid system under field conditions.

5.0 CONCLUDING REMARKS – PERSPECTIVE TO THE FUTURE

In order to have a perspective to the future, it is often useful to looking back the past. By the Figure 1 in the
earlier part of the paper, the authors showed the practitioners impression on the engineering issues related to the
ground improvement. We would like to add some words on the figure. The time in the followings are only
approximate ones.
In the middle of 1940s to the middle of 1950s, the geotechnical engineers in Japan paid a lot of energy on
infrastructure re-development in order to repair the country from the ruins of the last world war. Sand drain
method was imported from the USA and employed to accelerate the consolidation. Geotechnical community at
that time had not much option other than sand drain and replacement. The major issues then were stability and
excessive settlement. Still the number of the unexpected failures have been reported.
In the late1950s to 1960s, along with the economic growth, water front development in the coastal area
necessitated the advanced use of sand drains. In 1962, in the construction of high breakwater in Nagoya, 50,000
sand drain piles are installed to the sea bottom sediment in this specific project site. Continuous measurement of
settlement together with the pore pressure measurement were systematically done for the first time. The subject
issue in the marine construction was strength increase and the acceleration of settlement and these were
successfully achieved. At the same time on land, the highway construction by embankment was increased and the
sand drains were also employed. The major issue here was accelerating the consolidation settlement during the
embankment construction and thereby reducing the residual settlement. Unexpectedly there found a number of
cases where the settlement – time relations does not differ too much between improved and unimproved section.
This triggered a discussion on the effectiveness of vertical drains. A number of factors influencing the
consolidation process became and still now the agenda of discussion since then. The difference is recently
analyzed with the aid of sophisticated two dimensional finite element code. The Niigata earthquake on June 16,
1964 struck a wide area centering Niigata and caused damage to a variety of structures. The reason was the
liquefaction of loose sand (Tsuchida, 1998). This triggered the development of deep densification process..
Since the end of 1960s to the 1970s, further economic growth urged the development of new ground
improvement technologies which enables rapid construction even with increased cost. Sand Compaction Pile
Method, deep mixing, jet grouting were born in these time. The geotechnical community has had a variety of
options of the ground improvement. When the needs exists, the new technologies emerges by the efforts of the
practitioners. As Schlosser and Juran (1979) stated, experience always preceded the development of theory.
In the 1980s to present, the renovation of the already-developed urban area and rehabilitation of old
infrastructures have necessitated the construction in the close vicinity of the existing structures. The engineering
issues now involve the precise prediction of and control of deformation of the ground during and after the
construction. At the same time, excessive noise and vibration are prohibitive for the ground improvement
technologies used in the urban area. The level of requirement has thus increased. The needs to remedy the
contaminate ground are expanding the role of ground improvement and modification (Mitchell and Van Court,
1992). The renovation of urban area accompanies the excavation of the ground, annual amount of the
construction surplus soils in 1995 exceeded four hundred cubic meters in Japan alone. If we can recycle these
construction waste effectively in the other construction sites, we can reserve a good quality materials for the next
generation and we can preserve the environment of the borrow area. These urged the development of pre-mixing
method or the light-weight geo-materials, a family of admixture stabilization. But at the same time we need an
efficient market to develop. We should also carefully examine the possibility of the contamination of these waste
soil by the preceding land use.
As we have seen in the past, the ground improvement technique may be further developed in the future along
with the emerging new needs from the construction site and more generally from the public. The development
may be helped by the innovative materials and advanced equipment. With increasing needs of accurate prediction
and precise control of the ground deformation, the necessary development is not only within the specialized
ground improvement technology. We need to grasp the accurate ground condition before the ground
improvement, we need to employ more sophisticated numerical prediction to supplement the design based on the
traditional design codes, we need to carry out monitoring during and after the construction with the aid of new
instrumentation and computer. You may find even today a computer and displays in the air-conditioned cockpit of
recent deep mixing machine or sand compaction pile machine, some of which still simply tells the operator what is
going on in the soil beneath the ground but some of which now controls automatically the ground improvement
work as prescribed in the design. Perhaps in the near future, these equipment judges the difference in the actual
ground condition from those information used in the design stage and propose us to change the pre-determined
specification.
Due to the increased level of requirements, we can no more rely always on single ground improvement
technology. The combined use of different ground improvement may become a common practice and further
more the combined use of ground improvement with the innovative idea of structure’s side.
The key words for the further development of the ground improvement may be;
- Increased accuracy of prediction and back-analyses during the ground improvement work,
- Quality Control and Quality assurance aided by means of computer and GPS,
- Synergy of ground improvement and innovative structure,
- Environmental concern,

What else? We have to face and overcome the different needs of the future.

6.0 ACKNOWLEDGEMENTS

Mr. Sherif Hanna, Research Fellow at Polytechnic University and Mr. Masato Nishimura, Research Engineer
at Nikken Sekkei Nakase Geotechnical Institute contributed in the preparation of this report. Their assistance is
acknowledged with appreciation.

7.0 REFERENCES

Aboshi, H. Mizuno, Y. and Kuwabara, M. (1990) “Present State of Sand Compaction in Japan”. ASTM STP 1089,
pp. 32-46.
Asano, J., Ban, K., Azuma, K. and Takahashi, K. (1996) “ Deep mixing method of soil stabilization using coal
ash”, Grouting and Deep Mixing, Balkema, Vol. 1, 393-398
ASCE (1977) “Soil Improvement History, Capabilities, and Outlook”. ASCE Special Technical Publication
ASCE Committee on Grouting (1995) “Verification of geotechnical grouting”. ASCE Geotechnical Special
Publication No. 57, 177 p.
ASCE (1997) “Ground Improvement , Ground Reinforcement, Ground Treatment – Developments1987-1997”.
Proceedings edited by Schaefer, V.R., ASCE Geotechnical Publication No. 69
Assarson, K.G. et al. (1974). “Deep stabilization of soft cohesive soils”, Linden-Alimak. Skellefteå
ASTM Committee on Soil and Rock (1990) “Deep foundation improvements: Design, Construction and Testing”.
ASTM STP 1089, 337 p.
Barksdale, R.D., and Bachus, R.C., (1983), “Design and Construction of Stone Columns, Vol. 1”, Report No.
FHWA-RD-83-026, Federal Highway Administration, Washington, D.C.
Bayuk, A.A., and Walker, A.D. (1994). "Dynamic Compaction: Two Case Histories Utilizing Innovative
Techniques." In-Situ Deep Soil Improvement, Geotechnical Special Publication No. 45, ASCE, pp 55-67
Bell, A.L. Editor (1992) “Grouting in the ground”, Proceedings of the Conference organized by ICE, London, 658
p. ICE
Beven, D.J. (1997), “The Use of Dynamic Compaction to Remove / Reduce Risks due to Collapse Compression
and Seismicity in Silty Fine Sands”, Proceeding of the 3rd Int. Ground Improvement Geosystems, London,
June
Borden, R.H., Holtz, R.D. and Juran, I. Editors (1992) “Grouting, Soil Improvement and Geosynthetics”. ASCE
Geotechnical Special Publication No. 30. 2 volumes.
Brauns J., and Kast, K. (1981). "A new technique for preparation of dynamically compacted samples of rockfills."
Proceedings from Tenth International Conference On Soil Mechanics And Foundation Engineering. Vol. 3,
Stockholm, Sweden, pp. 669-671.
Bredenberg, H., Holm, G. and Broms, B.B. Editors (1999) “Dry Mix Methods for Deep Soil Stabilization”,
Proceedings of the International Conference on Dry Methods for Deep Soil Stabilization, Stockholm, 358 p.
Balkema
Broms, B.B. and Boman, P. (1975).”Lime stabilized column”. Proc. 5th Asian Regional Conf. on SMFE, Vol. 1,
pp. 227-234
Broms, B. B. and Boman, P. (1977). “Stabilization of soil with lime columns”, Royal Institute of Technology,
Stockholm
Bruce, D.A., Bruce, M.E. and DiMillio, A.F. (1998) “Deep Mixing Method: A Global Perspective”. Soil
Improvement for Big Digs, ASCE Geotechnical Special Publication No. 81, pp. 1-26
Bruce, D.A., Bruce, M.E. and DiMillio, A.F. (2000) “Deep Mixing: QA/QC and Verification Methods”. Grouting,
Soil Improvement including Reinforcement”. Proceeding for the 4th Int. Conf. on Ground Improvement
Geosystems, Helsinki, pp11-22, Finnish Geotechnical Society
Charles, J.A., Burford, D., Watts, K.S. (1981). "Field observations of the effectiveness of dynamic consolidation
to improve ground bearing capacity." Proceedings from Tenth International Conference On Soil Mechanics
And Foundation Engineering. Vol. 3, Stockholm, Sweden, pp. 617-622.
Choa, V. (1989) " Drains and Vacuum preloading pilot test, " Proc. XII Intl. Conf. On Soil Mech. And Found.
Eng., Rio de Janeiro, Brazil, 1347 - 1350
Chow, Y.K., Yong, D.M., Young, K.Y. , and Lee, S.L. (1994). "Dynamic Compaction Of Loose Granular Soils:
Effect Of Print Spacing." Journal Of The Geotechnical Engineering Division, ASCE, Vol. 120, No. 7, July
1994, pp. 1115-1 133.
Coastal Development Institute of Technology (1999a) “Technical Manual for Pre-mixing method”. 152 p. (in
Japanese)
Coastal Development Institute of Technology (1999b) “Technical Manual for Light-weight Geo-material”. (in
Japanese)
Coastal Development Institute of Technology (1999c) “Technical Manual for Deep Mixing with reference to
marine works”. 226 p. (in Japanese)
Cognon, J.M. (1991). " Vacuum Consolidation”, Rev. French Geotechnic,Vol.57 (October), 37-47.
Cognon, J.M., Juran, I, and Thevanayagam, S.,(1996)“Vacuum Consolidation Technology-Principles and Field
Experience” proceedings of International Workshop on Technology Transfer for Vacuum-Induced
Consolidation: Engineering and Practice”. The National Science Foundation, Los Angels(CA), November 22
and 23 (1996).
Dise, K., Stevens, M.G.,and Von Thun, J.L. , (1994) ‘’ Dynamic Compaction to Remediate Liquefiable
Embankment Foundation Soils’’, In-Situ Deep Soil Improvement, ASCE Geotechnical Special Publication
No. 45 pp 1-25
Drumheller J.C., and Shaffer, R. V. (1996) Densification, Inc. Report to GME Engineering, Dynamic Compaction
at the Quarry Retail Center, Minneapolis, MN
Druss, D. L.(2000) “Boston’s Central Artery Project – A Showcase for Ground Treatment Technology”, Grouting,
Soil Improvement including Reinforcement”. Proceeding for the 4th Int. Conf on Ground Improvement
Geosystems, Helsinki, pp23-30, Finnish Geotechnical Society
Findlay, J.D., and Sherwood, D.E. (1986).”Improvement of a hydraulic fill site in Bahrain using modified heavy
tamping methods” Building on Marginal & Derelict Land., May 7-9.
Grouting Committee of Geo-Institute (1998) “Grouts and Grouting”. ASCE Geotechnical Special Publication No.
80, 195 p.
Haegeman, W., and Van Impe W.F. (1995),”Evaluation of Heavy Tamping on a Waste Disposal by the SASW
Method”, Proceeding of the XI ECSMFE, Copenhagen
Haloton, G.R., Loughney, R.W., and Winter, E. (1965) "Vacuum Stabilization of subsoil between runway
extension at Philadelphia International Airport," Proc. VI Intl. Conf. On Soil Mech. And Found. Eng.,
Montreal, Canada, 62-65.
Hansbo, S. (1993) “Band drains”. Ground Improvement edited by M.P. Moseley, pp. 40-64, Blackie Academic &
Professional – An Imprint of Chapman & Hall
Henmi, K., Katagiri, M., Terashi, M. and Fukuda, K. (2000) “Case History of the Reclamation at Island City in
Fukuoka, Proceedings of IS-Yokohama
Holm, G (1996) “Consolidation and Stabilization of Soft Clays” Proceedings of ”International Workshop on
Technology Transfer for Vacuum-Induced Consolidation: Engineering and Practice”. The National Science
Foundation, Los Angels(CA), November 22 and 23 (1996).
Holtz, R.D. (1975). " Preloading by vacuum: current prospects, "Transportation Research Record No. 548,
Washington, D.C., 26-29.
International symposium of ground freezing (1994), Nancy
International symposium on ground freezing (ISGF ’82) (1982), Hanover
International symposium on ground freezing and frost action in soils (1997), Lule
International Workshop on Technology Transfer for Vacuum-Induced Consolidation: Engineering and Practice".
Los Angeles, California, November 22 and 23, 1996.
Jacob, A., Thevanayagam, S., and Kavazanjian, E. (1994) "Vacuum-Assisted Consolidation of a hydraulic
landfill", Proc. ASCE Spec. Conf., Settlement 94,Texas A&M Univ., Texas. proceedings of International
Workshop on Technology Transfer for Vacuum-Induced Consolidation: Engineering and Practice”. The
National Science Foundation, Los Angels( CA), November 22 and 23 (1996).
Japanese Geotechnical Society. (1990). “JGS T 821-1990 : Practice for making and curing non-compacted
stabilized soil specimens.”
Japanese Geotechnical Society (1999) “Geotechnical Engineering of Tokyo International Airport Offshore
Expansion Project – Slide Library”
Juran, I. (1997) “Panel Discussion on R&D Challenges in Soil Reinforcement”, Proc. 14th ICSMFE, Hamburg,
Vol. 4, pp. 2467-2471
Katagiri, M., Terashi, M., Kaneko, A. and Uezono, A. (1996) “Consolidation Settlement of Pump-Dredged Clay
Suspension – Analysis of a Case Record –“. International Workshop on Technology Transfer for
Vacuum-Induced Consolidation: Engineering and Practice, Los Angeles. pp. 51-65, Pre-workshop Copy
Kawasaki T. et al. (1981). “Deep Mixing Method Using Cement Hardening Agent”, Proc. 10th ICSMFE, Vol. 3,
pp. 721-724
Kanatani, M., Yoshida, Y., Kokusho, T., and Ikeda, M., (1997) “Model Tests on Densification of Sandy Ground
by Dynamic Consolidation method with Rotational Ram” Proceeding of the 3rd Int. Ground Improvement
Geosystems, London, June
Kitazume, M. et al. (1977). “JGS TC Report: Japanese design procedures and recent activities of DMM”,
Grouting and Deep Mixing, Vol. 2, 925-930
Kjellman, W. (1952). "Consolidation of clay soil by means of atmospheric Pressure. " Proc. Conf. On Soil
stabilization, Massachusetts Institute of Technology, Cambridge, Massachusetts, 258-263
Kobayashi, M. (1990) “Application of Finite Element Analysis in Stability and Deformation Problems”, Doctoral
Thesis, Tokyo Institute of Technology, 71 p. (in Japanese)
Kurihara, N. (1982) “The Problems in Vertical Drain”, Tsuchi to Kiso, Journal of Japanese Geotechnical Society
Vol. 30, No. 11 (in Japanese)
Lambrechts, R.L. and Roy, P.A. (1997) “Deep Soil-Cement Mixing for Tunnel Support at Boston’s I-93NB/I-90
Interchange”, Paper presented at ASCE Geo-Logan Conference
Leonards, G.A., Bennett, W.T., and Holtz, R.D., (1980), “Dynamic Compaction of Granular Soils”, Journal of the
Geotechnical Engineering Division, ASCE, Vol. 106, GT10.
Liausu, P., (1984) Renforcement de Couches de Sol Compressibles par Substitution Dynamique, In-Situ Soil and
Rock Reinforcement Conference, Paris.
Lo. K. et al. (1990) Unified Approach to Ground Improvement by Heavy Tamping, Journal of Geotechnical
Engineering, ASCE, Vol. 116, No. 3, pp514-527
Lukas, R.G. (1986). "Dynamic Compaction For Highway Construction." Volume 1, Report
No.FHWA/RD-86-133, Federal Highway Administration, Washington, D.C.
Lukas, R.G. (1992). "Dynamic Compaction Engineering Considerations" Grouting / Soil Improvement and
Geosynthetics, Vol. 2, pp. 940-953.
Luongo, V (1992). "Dynamic Compaction Predicting Depth-Of Improvement." Grouting/Soil Improvement And
Geosynthetics, Vol. 2, pp. 927-939.
Mayne P.W., Jones, J.S., and Dumas, J.C., (1984).”Ground Response to Dynamic Compaction”, Journal of
Ground Engineering Div., ASCE, 109, GE10
Menard, L., (1974)”Le Pilonnage Intensif ‘’ Annales de I’ITBTP, No 320, Sept., Paris
Menard, L, and Broise, Y., (1975) "Theoretical And Practical Aspects Of Dynamic Consolidation." Geotechnique,
London, England, Vol. 25, No 1, pp3-18.
Menard, L (1975) “Theoretical and Practical Aspects o Dynamic Consolidation”. Geotechnique, Vol. 25, No. 1,
pp. 3-17
Mitchell, J.K.(1981) “State of the Art – Soil Improvement”. Proceedings of the 10th ICSMFE. Stockholm, Vol. 4,
pp. 509-565.
Mitchell, J.K. and Zoltan, V.S. (1984) “Time dependent Strength Gain in Freshly Deposited or Densified Soil”.
Journal of Geotechnical Engineering, ASCE Vol. 110, No. 11, pp. 1559-1576
Mitchell, J.K. and Welsh, J.P. (1989) “ Soil Improvement by combining methods. Proc. 12th ICSMFE, Rio de
Janeiro, pp. 1393-1396
Mitchell, J.K. and Van Court, W.A. (1992) “The role of soil modification in environmental engineering
applications”. Grouting, Soil Improvement and Geosynthetics. ASCE Geotechnical Special Publication No. 30,
Vol. 1, pp. 110-143.
Munfakh, G.A., (1997), “Ground Improvement Engineering – the state of the US practice: Part 1. Methods”,
Journal of Ground Improvement, Vol. 1, No. 4.
Nakata H. et al. (1997) “Haneda Airport offshore expansion project”. Civil Engineering International, Proc.
Institution of Civil Engineers, 120, pp. 36-48
Nozu, (2000) personal communication, source unknown
Ohishi, K. et al. (1997). “Application of coal fly ash for improving soils prior to excavation”, Proc. International
Conf. on Engineering Materials, Vol. 2, 531-541
Okumura, T. and Terashi, M. (1975). “Deep lime mixing method of stabilization for marine clays”, Proc. 5th
Asian Regional Conf. on SMFE, Vol. 1, 69-75
Olson, R.E. and Daniel, D.E. (1981) “Measurement of the Hydraulic Conductivity of Fine Grained Soils”
Permeability and Ground Contaminant Transport. 1981 pp18-64
Oshima A. and Takada, N. (1997) “Relation between compacted area and ram momentum by heavy tamping”.
Proc of 14th IICSMFE, Vol. 3, pp. 1641-1644.
Oshima, A. and Takada, N. (1998) “Evaluation of compacted area of heavy tamping by cone point resistance”.
Proc. of the International Conference Centrifuge 98, Tokyo, Vol. 1, pp. 813-818.
Poran, C.J., Heh, K.S., and Rodriguez, J.A (1992). "A New Technique For Quality Control Of Dynamic
Compaction." Grouting / Soil Improvement and Geosynthetics, Vol. 2, pp. 915-926.
Poran, C.J , and Rodriguez, J.A. (1992) "Design Of Dynamic Compaction." Proceedings of the 45th Canadian
Geotechnical Conference Innovation Conservation And Rehabilitation, Toronto, Ont. Canadian Geotechnical
Journal V 01.29, pp. 796-802.
Public Works Research Center (1999) “Technical Manual of Deep Mixing with special reference to on land
applications”. 326 p. (in Japanese)
Rathmayer H.G. and Saari, K.H.O. Editors (1984) “Improvement of Ground”. Proceedings of the 8th ECSMFE,
Helsinki, 3 volumes, Balkema
Rathmayer, H. (1997).” Deep mixing method for soft subsoil improvement in the Nordic countries”, Grouting and
Deep Mixing, Vol. 2, 869-877, Balkema.
Rathmayer, H. Editor (2000) “Grouting, Soil Improvement including Reinforcement”. Proceeding for the 4th
International Conference on Ground Improvement Geosystems, Helsinki, 570 p. Finnish Geotechnical Society
Rathmayer, H. Editor (2000) “Grouting, Soil Improvement including Reinforcement”. Proceeding for the 4th
International Conference on Ground Improvement Geosystems, Helsinki, 570 p. Finnish Geotechnical Society
Robinet, (1997) HELIOS “Hydrical Exchanges and Consolidation System” Brevet d’invention francais numero
9614363.
Robinet J.C, Pasquiou A., Palut J.M.(1997) “ Phenomenology of Hydrical Exchanges between the Underground
atmosphere and Storage host-rock”, Final report of the European Commission Nuclear Science and
Technology.
Rollins, K.M. and Rogers, G.W. (1994) “Mitigation Measures for Small Structures on Collapsible Alluvial Soils”.
Journal of Geotechnical Engineering, ASCE, Vol. 120, No. 9, pp1535-1553
Rollins, K.M., Kim, J.H. (1994). "U.S. Experience With Dynamic Compaction Of Collapsible Soils." In-Situ
Deep Soil Improvement, Geotechnical Special Publication No. 45, ASCE, pp. 26-43.
Saitoh, S. et al. (1985). “Hardening of soil improved by deep mixing method”, Proc. 11th ICSMFE, Vol. 3,
1745-1748
Schlosser, J.F. (1997) “Theme Lecture – Soil Improvement and reinforcement“. Proceedings of 14th ICSMFE,
Vol. 4, pp. 2445-2466. Balkema
Schlosser, J.F. and Juran, I. (1979) “Design parameters for artificially improved soils”, General Report, Proc. 7th
European Conf. on SMFE, Brighton
Scott, R.A., and Pearce, R.W., "Soil Compaction By Impact." Geotechnique, London, England, Vol. 25, No. 1,
1975, pp. 19-27.
Shibazaki, M. (1996) “State of the Art Grouting in Japan”. Grouting and Deep Mixing, Vol. 2, pp. 851-867
Shinsha, H., Watari, Y., and Kurumada, Y., (1991) ”Improvement of very soft ground by vacuum consolidation
using horizontal drains” Conf. On Geotech. Eng for Coastal Development (Geo-Coast 91), Port and Harbor
Research Institute (Editor), Vol. I, Yokohama, Japan , 387-392.
Swedish Geotechnical Society. (1997). “Lime and Lime Cement Column - Guide for project planning,
construction and inspections – “, SGF Report 4:95E, 111p
Tatsuoka, F. and Kobayashi, A. (1983). “Triaxial strength characteristics of cement-treated soft clay”, Proc. 8th
European Conf. on SMFE, Vol. 1, 421-426
TETC(1990) “Geotechnical Investigation for the Pier 300 42-acre Landfill Ground Modification Project, vol. I
and II” Report prepared by Earth Technology Corporation. Los Angeles(CA)
Terashi, M. et al. (1977). “Fundamental properties of lime-treated soils (1st Report)”, Rept . of PHRI, Vol. 16-1,
3-28 (in Japanese)
Terashi, M., et al. (1979). “Engineering properties of lime-treated marine soils and DM method”, Proc. 6th Asian
Regional Conf. on SMFE, Vol. 1, 191-194
Terashi, M. and Tanaka, H. (1981). “Ground improved by Deep Mixing Method”, Proc. 10th ICSMFE, vol. 3, pp.
777-780
Terashi, M. and Tanaka, H. (1983). “Settlement analysis for Deep Mixing Method”, Proc. 8th European Conf. on
SMFE, Vol. 2, pp. 955-960
Terashi, M., Tanaka, H. and Kitazume, M. (1983). “Extrusion failure of ground improved by the Deep Mixing
Method”, Proc. 7th Asian Regional Conf. on SMFE, Vol. 1, pp. 313-318
Terashi, M. et al. (1983). “Fundamental properties of lime and cement-treated soils (3rd Report)”, Rept of PHRI,
Vol. 22-1, 69-96 (in Japanese)
Terashi, M., Kitazume, M. and Yajima, M. (1985). “Interaction of soil and rigid buried structure”, Proc. 11th
ICSMFE, Vol. 3, 1757-1760
Terashi, M. (1997) “Theme Lecture: Deep Mixing Method – Brief State of the Art”, Proc. 14th ICSMFE, Vol. 4,
pp.2475-2478
Terashi, M. and Tanaka, H. (1998) “Selection of appropriate countermeasures for the soft ground”. Tsuchi to Kiso
(Journal of Japanese Geotechnical Society), January 1998, pp. 55-60 (in Japanese)
Terashi, M. and Miki, H. (1999) “Importance of prediction in ground improvement”, Prediction and Performance
of Ground Improvement, Practitioners Series No. 11, Japanese Geotechnical Society, pp. 1-10, (in Japanese)
Thevanayagam, S., Kavazanjian, E., Jacob, A. and Juran, I. (1996) “Prospects of vacuum-assisted Consolidation
For Ground Improvement Of Coastal and Offshore Fills”. Proceedings of International Workshop on
Technology Transfer for Vacuum-Induced Consolidation: Engineering and Practice”. The National Science
Foundation, Los Angels(CA), November 22 and 23 (1996).
Torstensson (1984).”Konsolidering av lera Medelst Konstgjord Grundvattensankning och/eller Vakuummedoten
I Kombination med Vertikaldranering” Swedish Council of Building Research 75p
Tsuchida, H. (1998) “Liquefaction and countermeasures”, NNGI Report No. 5, Nikken Sekkei Nakase
Geotechnical Institute, pp. 35-56.
Tsuchida, T. et al. (1996) Field test of light-weight geo-materials for harbor structures, Technical Note of the Port
and Harbour Research Institute, No. 833, 30 p. (in Japanese)
Uchida, K., Imai, K., Tatsuoka, F. and Kohata, Y. (1996) “Ground Improvement by Cement –Treatment in
Trans-Tokyo Bay Highway Project”. Grouting and Deep Mixing, Proc. of the 2nd Int. Conf. on Ground
Improvement Geosystems, Tokyo, Vol. 1, pp. 669-674. Balkema
Van Impe. W.F., Cock, F., Van Der Cruyssen, and Maertens, J., (1997).”Soil Improvement experience in
Belgium: Part I. Overview and Dynamic Compaction”, Journal of Ground Improvement, Vol. 1, No3.
Van Impe. W.F, (1992).”Dynamic Soil Improvement methods”, Proceeding of the seminar on soil dynamics and
geotechnical earthquake engineering workshop on seismic zoning methodologies for geotechnical hazards,
Lisbon.
Varaksin, S., (1981).”Recent Developments in Soil Improvement Techniques and their Practical Applications”,
Sols Soils No 38/39
Woo, S.M., Moh, Z.C , Van Weele, A.F., Chotivittayathanin, R. and Trangkarahart, T.(1989) “Pre-consolidation
of soft Bangkok clay by vacuum loading combined with non-displacement sand drains”, Proc.12th ICSMFE,
Rio de Janeiro ,1431-1434.
Yoshida, Y., Kanatani, M., Kokusho, T., and Ikeda, M., (1995) “Model Tests on Rotary Dynamic Compaction
with Rotation of Hammer”, 30th Annual meeting of Japan Geotechnical Society, pp. 2047-2048
Yonekura, R., Terashi, M. and Shibazaki, M. Editors (1996) “Grouting and Deep Mixing”, Proceedings of the
2nd International Conference on Ground Improvement Geosystems, Tokyo, 2 volumes. Balkema
Zen, K. et al. (1987) “Study on a reclamation method with cement-mixed sandy soils – Fundamental
characteristics of treated soils and model tests on the mixing and reclamation”, Technical Note of the Port and
Harbour Research Institute, No. 579, 41 p. (in Japanese)