Smart Sensing in Geotechnical Engineering

I. INTRODUCTION
The advent of low-power wireless chip transceivers has enabled new possibilities for widespread
instrumentation of large civil structures and distributed geotechnical systems for which large-scale
testing has been limited to unique cases due to the economic burden associated with cable-based
hardware. However, while the number of unique wireless sensor platforms and proposed applications
has continued to rapidly expand, there has been limited success in replicating previous cable-based test
programs in regards to the number of deployed sensors and data rates. Consequently, a degree of
skepticism and disillusion has developed over the past several years of wireless sensor network
development in the context of geo-structural monitoring. Instruments, such as pore pressure
transducers, geophones, accelerometers, settlement monitors, and inclinometers, are commonly used
in geotechnical fieldwork. Traditionally, these instruments have been used with cable-based data
acquisition (DAQ) systems that restrict their placement in urban areas and construction sites.

Wireless sensing systems are compatible with many geotechnical instruments and represent an
improvement over tethered DAQ by eliminating concerns about maintaining of a wired connection
between instruments and a global DAQ. Additionally, wireless sensors are endowed with both memory
capacity and embedded data processing resources. For example, wireless sensor networks can maintain
recorded data at the node so that the data can be locally processed or held until transmission is
possible. . Due to their durability, economical pricing, and the previously mentioned advantages,
wireless sensors have potential for greatly expanded use in geotechnical engineering. With a little more
research and experience, the health monitoring of buried infrastructure could be as mature an area as
structural health monitoring is now. Potential applications where wireless sensors could advance the
current state of geotechnical engineering practice include monitoring strain in buried pipelines and
tunnels, recording vibrations near construction or blasting sites, measuring foundation settlement, and
providing early warning of slope instability.

I. Traditional Approach

Today there are a large number of wireless applications in many engineering areas, such as
communications and informatics. Wireless technology (or communication) dates back to the nineteen
centuries and since then they have developed and successfully been applied to a wide range of devices
and equipment used in daily life. For this reason, this technology should be considered in other areas,
since many times the use of cables for testing or monitoring can bring various issues.The main
disadvantages of using a cable-based system are.

 not all the structures can accommodate cables,

 limited physical space causes limitations on the amount of data transmitted,

 the cables may be damaged during the assembly, tests and/or transportation, and when damaged
they are much more difficult to inspect to detect the problem, and possibly needing full replacement,
resulting in delays. Therefore, a cable-based system requires a higher maintenance and human
intervention. On the other hand, the wireless systems:

 allow an optimal placement of the sensors that normally have a small size, and diversity in usage,
 when using wireless systems, we have fewer components to relocate as the construction proceeds,

 wireless technology need low power consumption and has a limited need for surge protection

 plus, digital data (via wireless) has a higher reliability, and a lower likelihood of data loss when
compared to analog transmissions (via cable).

Based on all this, it is clear that only benefits can come from the use of wireless technology. The use of
wireless can become an important tool of current use in many areas, including civil engineering and
geotechnical engineering. Therefore, they should become a routine tool in structures and infrastructures
health monitoring, between others. Nowadays there is a fair number of wireless applications to
geotechnics. However, such applications refer to surface transmission of geodata, such as monitoring
and acquisition systems and not to subsurface transmission of signal collected from sensor in
underground testing.

II. WIRELSS DEVELOPMENTS

Wireless sensor networks are a sub-class of wireless networks, which were developed to overcome
specific problems related with wireless sensing. Wireless sensor networks may be used for many
applications, but most of them share the same main requirement: to reliably measure the required
variables, without interfering with the sensing environment, and requiring as low maintenance as
possible. With this in mind, many solutions have been proposed along the way, and others are still
under development, without no consensus about which solution fits best for all sensing applications.
One of the most well-known technologies is Zigbee, which was specifically developed to satisfy the
particular requirements of wireless sensor networks. It uses an electromagnetic carrier operating in the
2.4 GHz Industrial, Scientific and Medical (ISM) bands, where the wireless communication channel must
be shared with Wi-Fi and Bluetooth, also sometimes used for data communication. Such wireless sensor
networks may be of great help for geotechnical applications, when the data is gathered at the surface,
or even in underground, when the transmission is done in free space. However, a few considerations
must be made when attempting to use such technology for underground data collection. The
development of wireless sensors technology presents a great opportunity for developing low-cost
monitoring or data acquisition systems for civil structures and infrastructures.

3.1. Applications in Geotechnical Engineering, Monitoring and Field

There are different types of sensors available for monitoring tests and for data acquisition intended for
almost any kind of structure or infrastructure, onshore or offshore, mainly in urban excavations , slopes
and compaction of highways, railways or earth dams and bridges . Many of the monitoring systems
involve measurements of different types of sensors such as: pore water pressure cells, inclinometers,
earth pressure cells and settlement meters. For example, the work of presents commercially available
soil moisture probes and soil tilt sensors combined with lowpower, wireless data transmitters to form a
self-configuring network of soil monitoring sensors (solar-powered). This research has shown that
commercially available wireless instrumentation can be modified for use in geotechnical applications.
This type of monitoring is specially installed when instability can have lost of human lives or if the
construction site has high seismic hazard. Wireless sensors have brought many advantages for
monitoring, such as a perfect cut-off of water, no weak point avoiding data loss due to a cable damage,
influence of lightning, lower costs with placement and no cables. Furthermore, wireless ethernet data
acquisition systems were also implemented in physical tests, like centrifuge tests. This type of tests
requires transmitting data collected in a moving, increased-acceleration environment, and the
traditional methods (slip rings) have limited accuracy as well as limited number of instruments that can
be used to data acquisition. This study obtained good results with the wireless technology although it
concluded that it depends on the g-level, transmission direction, and type of wireless card used.
Nevertheless, wireless technology is still scarce in nearly all kinds of “in situ” tests, like cone
penetrometers, pressure meters, vane tests, seismic waves-based tests or dynamic load plate tests,
among others, use the transmission of signals by a cable with wires from the sensors installed on the
probe. This implies, in the case of transmission of signals from down-hole to surface via cables, that
before the test all the necessary rods and cables are gathered in the surface, assembled and the cables
inserted inside the rods. This may cause some inconvenient during the test, but still these are the
current tools available and the only that engineers use for soil testing. The use of a wireless system for
the soil test itself is something that some have mentioned but few have tried

V. CONCLUSION

WSNs with dense wireless sensors provide a potential solution for long-term, scalable SHM of bridges by
providing easier installation and efficient data management at a lower cost than that of traditional
tethered monitoring systems. The researchers mentioned previously have dedicated their efforts to
promote the WSNs-based bridge health monitoring system with the aim of replacing wired structural
monitoring systems. Many available wireless sensor systems are already quite capable and can be
expected to replace the traditional wired sensor systems for bridge monitoring. However, wireless
sensing technology is still in its infancy; much work remains for bringing this promising technology to
fulfill the requirement of complex bridges monitoring and evaluation

Trenchless Technology’ ‐ Techniques and Examples of Successful Practice

Most construction projects require new supply and waste disposal pipes to be laid or to replace the old
ones present therefore trenchless technology has been developed to offer the ability to achieve this
without disturbing the layers of ground underneath the topsoil which could be inhibited or used by
traffic.

Trenchless Technology can be defined as ‘the technology for placing new pipe, cable, or conduit in the
ground between two defined points without continuous, open cut excavation between them, or for
renovating, replacing, and rehabilitating. Trenchless technology emerged in such countries as Japan and
Australia where modern sewer networks were needed as the population of the countries were
increasing. The primitive technologies used in these countries were then adapted by UK pioneers in the
1970s and early 1980s.

The advantages of using trenchless technology as opposed to open cut excavation methods are
numerous such as minimising the environmental effects which open cut excavation methods cause due
to disturbing the soil, organisms and water bodies. Minimal ground disturbance results in discovering
fewer unknowns from the ground. Trenchless technology also saves the project time and excessive costs
due to increased engineering from surveying and design calculations. It also minimises the installation
time in comparison to open cut excavation methods as well as increasing the safety of the construction
related to steep excavations. There are various methods of trenchless technologies that may be used
and the type of method chosen for a certain project is dependent upon the pipe size that needs to
installed, the depth it needs to be installed to, the soil conditions of the ground and the overall cost of
the method. The different methods have been outlined below.

Impact Moling: “a technique in which a percussive mole (soil displacement hammer) is launched from
an excavation to displace the soil and form a bore. The new conduit is normally drawn in behind the
mole or pulled back into the bore using the hammers reverse action.

Thrust Boring: “a solution for installing new pipes in virgin ground where accuracy is critical. Working
from a compact launch pit a rod is thrust into the ground which pulls through a pipe into its proposed
position

As can be seen from the table, the two main methods of trenchless technologies are pipe jacking and
tunnelling/microtunnelling due to their wide range of use within soils. Whatever trenchless method is
chosen though, the soil of the ground on the site has to have an adequate strength to remain
withstanding within the shafts. The finer the grains and particles of the soil are the more effective these
methods are, as using trenchless methods can prove difficult in hard rocks due to their displacement.

Pipe jacking

Pipe jacking is used to describe the technique of installing man-entry pipes by adding sections of pipe at
the drive pit and jacking the line forward to form the tunnel lining behind the cutting shield.’ This
method allows for man-entry-size pipes to be installed using hydraulic methods. Pipe jacking has
developed widely throughout the years as the first use of this method was recorded in 1892 in USA for
Northern Pacific Railroad Company. This method of trenchless technology is preferred over other
methods due to its simplicity of installation and benefits such as avoided settlement of ground

Method:

1. The tunnelling machine has a thicker wall version with a female end fitted with a collar rimmed with a
thin ring of fibre board and a male end fitted with a rubber seal. This connection forms a watertight seal
between the two ends minimising friction.

2. Jacks are extended and released to push the tunnelling machine into the earth and the speed of
movement is synchronised with the force of the jacks.

3. Lubricant is required between the annulus and surrounding earth to minimise friction acting at longer
distances.

4. To complete, the manholes are built and the shafts are backfilled.

As well as having many advantages, pipe jacking can acquire many disadvantages such as its cost. Due to
the technology used its cost may be quite high however it is able to save on other engineering costs. The
pipes used for jacking need to possess the correct strength otherwise the method might fail. If a fail
does occur, there is great difficulty in replacing those damaged pipes. To achieve a fully functioning
system, the pipes must align perfectly which can also help with the water tightness of the tunnel.

Pipe jacking Case Study:

The increase in urbanization on Route 63 in Macon, Missouri led to upstream flooding which initiated
the need for two additional culvert pipes. These were to be installed into the ground using the pipe
jacking method. After a thorough site investigation, it was concluded that the soil was mostly lean clays
scattered with sand and gravel, with no boulders. Post construction analysis of the successful installation
indicated the need for more culvert pipes to further decrease the amount of flooding.

Microtunnelling

This method entails installing pipes into the ground through hydraulic methods without the aid of
personnel, as opposed to tunnelling, which are remotely controlled from the outside. It requires the use
of a microtunnelling boring machines and Kramer defines it as ‘those methods that install pipes with a
diameter of less than 36 inches (900mm) to a predetermined line and level by remotely controlling the
cutting head.

Method:

Microtunnelling uses two shafts to and hydraulic jacks to fit the tunnelling system underground allowing
the utility pipes to be placed inside it permanently. This is done by the following method:

1. Shafts are sunk in at each end of the intended drive usually at man-hole positions, one end drive shaft
and one reception shaft. The shafts must be long enough to accommodate the tunnelling equipment.

2. The microtunnelling machine is then lowered on the guide tracks by a crane allowing it to be thrust
forward by hydraulic jacks. It is guided by a laser which projects onto the target surface.

3. The excavated material at the front is crushed by the head of the machine and any slurry is brought to
the surface by slurry shafts in the tunnelling machine. Any ground movements are eliminated due to the
counterbalanced ground pressures. Its journey is monitored by screens above the ground.

4. After the machine has penetrated the other shaft, excavation is halted and the pipes are disconnected

Microtunnelling Case Study

‘The project mentioned involved the installation of about 2.5 km of reinforced concrete pipes with
diameters ranging from 600 to 1000 mm at an average depth of 5 m below surface.’ (Jebelli et al. 2010)
Sewer lines were installed using micro tunnel boring machines and hydraulic pipe jacking. Problems
encountered during the construction were face instability, shaft failure and groundwater issues. These
problems were faced due to the low stability of small grained soils such as sand, therefore the mistakes
learnt from this case study can be implemented on other similar situations.

Comparison between Microtunnelling and Pipe jacking

Essentially both methods use the same technique however pipe jacking allows for bigger diameters of
pipes to be installed ‘Microtunnelling has more limitations than Pipe jacking in terms of ground
conditions. The smaller diameters and remote aspect of the operation make Microtunnelling more
prone to problems where soil conditions change rapidly or when obstructions are encountered

Case studies: Trenchless technology methods are used mostly for installing utility pipes especially those
running under busy roads or rivers. Treatment works for sewage pipes in Poland’s Vistula River in Poland
have been installed in record time using pipe jacking

Formerly, site investigations and trial pits were used to investigate ground conditions however over the
years this has evolved to virtual visualisation which can be seen in the TUNCONSTRUCT project in the
UK. This project illustrates a successful case study in which state of the art technology was used to
construct underground without the need for digging trenches. The use of software tools, visual
prototyping and robotic devices have allowed trenchless technology to develop from the use of
tunnelling machines and devices therefore reducing the overall time, cost and risk endured in a project.

Another case where trenchless technology was used underground was for the build of Victoria
underground station in London, UK. Jet grouting was the method chosen for this as it creates less
potential for heave compared to other methods such as ground heave. This was important to achieve
due to the excessive use of the ground above the station, and any disturbance in the ground below may
cause high risks.

Critical risks can occur from trenchless technology such as pipes bursting underground as seen in the
Yorkshire, Asquith, Birkett and Reynard case study. Other risks include settlement of the ground which is
caused by the ground movement during the installation of the tunnelling machines. This can be
prevented firstly by maintaining stability of the excavation face and secondly by avoiding inadvertent
loss of soil in the tunnel. Microtunnelling is able to avoid this consequence as the pressure acting on the
pipe is always counterbalanced with the volume of soil being moved. Another problem that the use of
trenchless technology may cross is the flowing of groundwater into the shafts and tunnels formed by the
machines. This can cause flooding of the trenches, and may disturb the project overall therefore
adequate ground investigation and site surveys are required to fully comprehend the water table
beneath the ground of that specific site.

Obstacles such as boulders can be encountered during underground works and this can lead to
excessive excavations which can over cost the project. This affected the progress of a project mentioned
in which was remedied by removing boulders without losing the face, and the upper part of the face had
to be breasted from within the excavation. Some cases of rock encounter can be so severe that the
trenchless method has to be elapsed and a full open cut excavation method used to carry out the
required work.

Discussion and Conclusion:

Trenchless technology has been in development since it was first used; nevertheless, the innovation of
technology itself in the twenty first century has allowed the use of software and databases to predict the
ground beneath the surface which was one of the preventions for these methods before. Gaining access
to the history of the ground usage of a site has proved difficult in the past however with the emergence
of such databases as the Underground Construction information System (UCIS) this aspect of trenchless
methods could be implemented on most projects in the future, therefore benefitting the environment
by minimizing disruption to the ground.

DETERMINING THE OPTIMUM LEVEL OF INVESTIGATION FOR A TRENCHLESS INSTALLATION PROJECT

1. INTRODUCTION

Today’s modern trenchless technology and its associated techniques provide utility networks owners
with a number of tools to undertake their utility upgrade or renewal works with a minimum of
disruption and disturbance to their customers and the general public. However, without sufficient
planning, these advantages can quickly disappear and the very disruptions that the trenchless methods
sought to eliminate or minimize become very real. Part of the project planning includes pre-construction
investigation undertaken by the Client during the design phase. This paper will summarise the optimum
level of investigation for various types of trenchless installation projects, from a cost of risk minimization
perspective. The main key project risks which will be considered are time, cost, liability and image
perception. There are a vast number of items that need to be considered and investigated within the
design phase of any project, however, not every issue impacts on ALL four of the key risks. This paper
will focus on those items that do impact on all four key risks and therefore must be investigated as a
minimum. For each issue, the consequence and mitigation strategy for each trenchless technique from
new installations to pipe rehabilitation, will be reviewed. These include medium and maxi HDD, micro
tunneling, pipe bursting and pipe lining.

CRITICAL ISSUES THAT REQUIRE INVESTIGATION

II.1 GROUND CONDITIONS

Encountering unexpected ground conditions may result in the contractor using the wrong equipment
and not being successful in completing the installation within budget or having to complete open trench
excavation to install the service. Examples of unexpected ground conditions that can significantly affect
a trenchless installation range from encountering hard rock or uncontrolled fill to running cohesionless
sand to very sticky bentonitic clays. If any of these situations, or similar, are encountered, without the
correct forward planning, the likelihood that the installation won’t be able to proceed, is very high. In
the case that these ground conditions are unexpected/unforeseen, then the Client may be liable for any
additional costs incurred by the contractor, in completing the installation, as well as any time extension
appropriate for the additional installation works. Furthermore the Client’s liability to 3rd party costs
and/or damages also increases. The public perception of poorly planned works negatively impacts on
the client’s public image, which can impact on future sales etc.

Geotechnical borehole sampling at approximately 200m intervals will assist in minimizing the likelihood
of unforeseen ground conditions being encountered and assist in ensuring the contractor has the
correct equipment to complete the works, with reduced risks, thus maximizing the trenchless
technology benefits of less disruption and less impact on the environment. Boreholes should be taken at
10m offset from the proposed route, so as to avoid pressurization during the actual installation. Bore
depths should extend another 30% deeper than the proposed design depth, to confirm material
consistency and/or to allow for variations in the final drill alignment. Should the results from the bore
holes show consistent ground geology then samples at 200m will be satisfactory. If, however, the bore
samples show varying ground conditions then more sampling may be required. The borehole data
should also show the level of the groundwater table, as again, different controls need to be
implemented dependant on whether or not the works are above or below the water table. There have
been many specific papers written and presented on geotechnical investigations for a trenchless
installation project. To generally summarise, the typical type of data and information that should be
gathered from the geotechnical bore sample, will include:

 Atterberg limits, clay swelling limits

 Direct or triaxial shear,

 Unconfined compressive strength testing / Rock Mohs strength

 Standard Penetration Testing ( SPT ) / Cone Penetration Test ( CPT )

 Grain size distribution and grain shape angularity

 Groundwater table depth

This information is also invaluable at the planning/design phase as it inputs into the option analysis and
risk assessment that the client/designer will be carrying out to determine the optimum solution.

IMPACT OF UNFORSEEN GROUND CONDITIONS

Horizontal Directional Drilling Installation

These are the most common of trenchless techniques for new utility infrastructure installation projects
currently being completed in New Zealand. Some typical examples of this type of project would be the
replacement of existing local and/or trunk reticulation water, waste and storm pipes, up to 1200mm in
diameter. In some cases, although the product pipe to be installed is of a smaller size, the scope of the
project may encompass many kilometres of installation or it may be a single installation greater than
one kilometre. Unexpected ground conditions being encountered will result in the contractor using the
wrong equipment i.e. drill heads, reamers and incorrect drill fluids/”muds”. This will result in the
contractor not being successful in completing the drill shot and having to complete open trench
excavation to install the pipe. Cobbles and boulders are the curse of the HDD contractor. These are the
most difficult and demanding ground conditions to work in and, in some instances, even with the best
strategies in place, can prevent the pipe installation from being successfully completed. Hence it is
vitally important that the presence of cobbles and/or boulders is clearly identified.

UNDERGROUND SERVICE LOCATIONS

The location of the existing underground service and their impact on the new utility being able to
installed in the designed position, is similar to the ground condition issue, in that inaccurate position
determination may prevent the installation from being completed without some disruption and will
impact on all the four key risk areas. Within this issue there are two components, one that relates to the
nature of the service and the consequences of damage to it and the second issue is that of available
corridor space.

In both cases the first step to be undertaken is a desktop review of the underground service plans from
all of the utility providers. This will identify any high consequence/risk services in the construction
window and give some indication, when plotted, as to the level of underground congestion. In terms of
high consequence/risk services, these could typically be; high pressure gas pipes, trunk sewers including
rising mains, bulk watermains, high voltage power cables, fibre optic and international
telecommunication cables and the like. Clear position determination and separation from these services
must be ensured. Although there is a requirement on the contractor to carry out location and protection
measures of these services, when they are installing the new utility, if the design does not ensure and
allow for this to be reasonably achieved, by the contractor, then the combination of the high
consequence and likelihood of a strike creates an unacceptable risk, that the client can not ignore and
must mitigate. On-site potholing to confirm the correct position of these high risk services is therefore
required to manage and mitigate the risk, for both the client and contractor. On site location
determination, as well as managing the client’s liability and image risks, also will protect against any
potential variation cost and time claims.
The second aspect regarding underground service locations is from the point of view of congestion ie
having sufficient room and clearance space to install the new utility product. Again the first step in
identifying if this is a real risk is the desktop review of the underground service plans from the various
Utilities. These should be plotted onto the base topographic plan and then checked, where possible,
from a site inspection looking for obvious surface boxes to confirm the accuracy of the drawing. If, at
this stage, a lack of space or clear working room is identified, it should then be confirmed by on-site
potholing or a strip trench. The paper plans from the utility owners provide general information but
cannot be relied upon for accurate positioning, either horizontally or vertically. Minimum separation
distances from the existing service to the back reamed dimension of the new product also need to be
checked. As a general “rule of thumb” the separation distances should be 500mm, however these
should be increased in cases where the drilling is adjacent to either a high risk pipeline or a structurally
deteriorated asset.

2.2.1 IMPACT OF LOCATION OF UNDERGROUND SERVICES

2.2.1.1 Horizontal Directional Drilling / Micro Tunnelling Installation

The consequence of inaccurate underground utility location is that the HDD / Micro installation will not
be able to be installed on the design alignment. A change in alignment, in some cases, may not be
acceptable or achievable eg replacement of a drainage gravity pipe between two fixed points within the
reticulation. This will lead to design and/or methodology changes that will impact on cost, time, liability
and image. In other situations, the new asset maybe able to be installed, however the final location will
be less suitable than the original design eg having to install the asset in the carriageway rather than the
grass berm due to insufficient clearance space. This again will impact on costs, time, liability and image.

II.2 WORKING SPACE

II.3 Large scale projects involve larger plant and machinery, as well as more ancillary equipment, the
need for mud ponds and larger product pipe laydown area’s. Not being able to provide the
necessary area/space to the contractor is going to prohibit the works from proceeding on a large
scale. As such, the works would need to be completed either by a series of smaller scale
operations; which will result in impacts to the client of cost, time, liability and image.
2.3.1 ISSUES ASSOCIATED WITH WORKING SPACE
2.3.1.1 Horizontal Directional Drilling Installation
Notwithstanding that each contractor/tenderer will have their own specific methodology for
carrying out the works, there are a number of set common factors, with respect to working
area’s, that the designer must allow for, if large scale maxi rig HDD is the proposed
methodology. These area’s are the drill rig footprint, the mud mixers and recyclers, the mud
settlement ponds and the pipe stock/welding/laydown area. Noise control may also be an issue
of a contractor’s methodology. For a typical maxi rig HDD shot of 1,000m the working space at
the Rig side can be up to 70m wide by 100m long and 30m wide by 50m long at the respective
pipe side. The Rig side of an HDD site includes the area where the drill machine and rods are, the
separation equipment, slurry mixing and pump equipment are, as well as the offices and slurry
ponds. The pipe side includes such aspects as the cuttings settlement pits, the pipes to be
jointed and the drill rods. Similarly, whatever the proposed pipe installation length, that same
length needs to be available, to string the pipe out, on the pipe side. It may be possible to break
the whole full length into sections, but this has other issues associated with it. The pipe laydown
 either needs to be clear of all driveways and road crossings/intersections or appropriate traffic
management needs to be in place. This is generally not possible outside of working hours.
2.3.1.2 Micro Tunnelling
In the case of micro tunneling, the size and location of the launch and exit shafts and their
ancillary equipment is the critical defining factor. The pipe stock will still need to be adjacent on-
site, as well as the slurry mixing and settling tanks. Whereas, in the case of HDD, the positioning
of the drill rig and equipment may be able to have some flexibility, as the drill can be set back
from the actual pipe alignment and the rods extended, as required, this is not the case with
micro-tunnelling. The launch and retrieval shafts can only be on the pipeline alignment.
Therefore, not only does this limit the options, but also increases the criticality that there is
sufficient clear working space for the shafts. The location of underground services, as well as
being critical for the tunnel alignment, also needs to be confirmed for the shaft installations.
Note that, dependant on the size of the tunnel bore to be completed, the launch shaft can range
in size from as small as 3.5m diameter to 10m diameter or more. Quality construction of the
shaft is paramount and critical to the success of the tunneling operation and one of the key
aspects/elements in ensuring the quality construction of the shaft is the impact of any
underground services in the near vicinity of the shaft position.
2.3.1.3 Pipe Bursting and Slip Lining
With respect to pipe bursting and slip lining techniques, similar aspects to both HDD and micro
tunneling apply, in so much as, the pipe laydown area consideration with regard to pipe bursting
and continuous slip lining and the launch pit location for sectional or segmental slip lining is
similar to micro tunneling. Failure to investigate fully these aspects may result in the pipe
installation procedure not being able to be completed as planned resulting all four of the key
risk elements occurring.
2.3.1.4 Pipe Rehabilitation
The consideration of working space, with respect to pipe rehabilitation, is concerned with access
to the manholes, both upstream and downstream and the ability to get the necessary plant and
equipment to the manholes. Insufficient room and/or lack of accessibility may result in either
the works not being able to proceed and/or significant changes to the methodology, both
impacting on the four key risk elements for the client, cost, time, liability and public image
perception.

II.4 CONDITION OF THE EXISTING PIPELINE

The existing pipeline to be replaced either by renewal or rehabilitation will have defects, hence
the need for replacement. The defects can be categorized into two groups, those that need
investigation so as to ensure a successful replacement and those that are classified as latent
defects. A latent defect, dips and deflections, for example, will not prevent the replacement but
will still remain, post a pipe bursting replacement or a lining. The latent defect issue is for the
client to determine, whether it is acceptable to them or not, but it does not impact on the four
key risk elements of a project. However, the other elements of the pipeline condition that will
impact could be a collapsed section, large debris build up, a bend or a change of material. A
current pipeline CCTV inspection is therefore critical to a successful project. It is not uncommon
for there to be a 6 month or longer time period between project initiation and the actual
carrying out of the physical works. If a pipeline is in a degraded state, then the potential that any
 defects may have further deteriorated, is greatly enhanced. A six month old CCTV is not
acceptable to mitigate the risks of accelerated degradation and the impact that may have on the
successful completion of the works
II.5 LATERAL CONNECTIONS
Determination of the location and status ie Live or Dead/Blank of all of the service connections
on the pipeline is critical to the successful completion of the pipeline rehabilitation. Failure to
incorrectly identify the status of a lateral may result in either an overflow, as a result of a live
lateral not being re-connected or unnecessary dead laterals being re-connected undermining
the project objective’s. It should be noted that the location and status of all connections can be
carried out as a subsidiary work package to the main pipeline CCTV investigation ie done at the
same time
II.5.1 ISSUES ASSOCIATED WITH LATERAL CONNECTIONS

The positions of house service lines that connect to the pipeline also need to be determined and
then accurately plotted onto a topographic plan as these will need to be excavated to be re-
connected. These positions and depths should also be shown on the surface of the pipe alignment
by pegging and photographed. Although uncommon, it is not unheard of, that the lateral connection
may be under a structure or even a significant tree that prevents the excavation. In the most
extreme case, a house lateral connection was underneath the concrete slab foundation of the
house, resulting in the owners being re-located for two weeks and the floor slab excavated to re-
connect the private drainage.

II.6 NON CRITICAL INVESTIGATION ISSUES

Aside from the above discussed issues there are still a number of other issues that need to be
considered for a trenchless project; issues that will not impact on all four of the risk elements
and can be managed through alternative methods eg through the contract procedure or
methodology. Every project will have it’s own specific issues and the same issues will not apply
to every project. Management of these types of issues is best covered through either the agreed
methodology to be used and/or contract scheduled items and rates and contingency planning
between the client and contractor, at the start of the project, or even during tender award
discussions with the preferred tenderer. These issues typically are of the type that are “when
they occur, not if” and have smaller, lower risk consequences. As such, contingency can be
included in a contract to cover any cost and time related effects, as well as methodologies
determining a defined course of action, so that “when” the event does occur, everyone knows
what to do. This also leads towards minimizing the impact of the event, from a publicity, liability,
lost time and cost perspective. Aside from the issues relating to the carrying out/completion of
the physical works, there are other issues that need to be considered as part of the design and
whether the methodology proposed is feasible. Some examples of these issues are:
 Bending radii, both of drill rods and pipe
 Volumes and type of By Pass pumping ie constant or variable flows
 Damaged laterals preventing Lateral Junction repairs to be installed

3 CONCLUSIONS
Modern trenchless installation and renewal techniques and technologies provide many
advantages and benefits to network owners. However, insufficient planning and investigation, at
 a project’s initial design phase, can mean that all of the benefits of the trenchless technique are
lost or overwhelmed during the actual construction and the client suffers from the four key risk
elements of a project; cost, time, liability and public image perception. From the perspective of
risk management/mitigation there is a cost effective amount of investigation that needs to be
carried out to ensure a successful project. For a trenchless project, of varying scales, the issues
that need to be investigated are; ground conditions, location of underground services, site
working room, the condition of the existing pipeline and lateral connections. Management of
these issues at the initial design stage will manage the clients risks throughout the project and
help towards achieving a successful project outcome.

Fuzzy Reasoning for the analysis of risks in geotechnical engineering. Application to a French
Case
INTRODUCTION

The analysis of risks induced by unstable cliffs is one of the major and important tasks in geotechnical
engineering. This analysis should take into account the fact that some engineering parameters that are
necessary for any risk analysis around unstable cliffs are difficult to be quantified and/or they are vague
and imprecise. That is why such risk analysis should depend meagerly on the experience of the
geotechnical engineer in charge for measurements. According to the French law, an area that is
subjected to a PPR study, should be classified into four categories expressing the natural risk into High,
Intermediate, Low, or Negligible. This classification is then used by local administrations, insurance
companies, and political issues of urban planning. This paper shows part of the analysis made by the
engineers of the INERIS (Institut National de l'Environnement Industriel et des Risques, in France) for the
evaluation of geotechnical risks around an unstable cliff zone situated in the Parisian basin. This work is
based mainly upon the experience of the geotechnical engineers as well as observation and proposition
of discrete values describing the state of the cliff overlying sometimes abandoned underground
workings. The authors have proposed -with the help and consultation of INERIS- some modifications to
the current methodology in order to account for the vagueness, uncertainties and the imprecision in the
estimation of each observed parameter. This methodology has been optimized and automated using
Visual Basic Programming in order to aid and facilitate the judgment of the experts' work needed for the
analysis. In order to perform such operations, fuzzy numbers have been proposed in place of the
discrete numbers used in the analysis and the fuzzy reasoning is used to evaluate the risk.

GENERAL METHODOLOGY FOR NATURAL RISK ANALYSIS

The methodology for risk analysis proposed by INERIS is based upon the observation of some
parameters at the sites under consideration. The method of analysis is inspired by the classical method
of Rock Mass Classification proposed by Bieniawski (1989). The procedure of this analysis is divided into
4 stages. The first one is the observation stage in which the engineer has to observe and document
some parameters at the field concerning the cliff, and the area around it. This stage is followed by a
stage of calculations in order to find out the sensitivity of the site towards a certain phenomena. The
other two stages consist of table crossing in order to find out the possibility of occurrence of the
phenomena and the risk imposed on the site. Figure 1 shows the different stages of this analysis.
All observed parameters (Intensity, Activity, Sensitivity Parameters) are classified into 4 classes. In order
to calculate the sensitivity of the site towards a certain phenomena, five parameters are measured.
These parameters are: joint spacing (JS), joint roughness (JR), humidity of fissures (HF), direction of
joints with respect to the front (DIR) and the dip angle of joints with respect to the front (DIP). The
activity of the site could be considered as the history of the site and classified into four classes as
sleeping, inactive, fresh, and active. The intensity of the phenomenon is also classified into four classes
determining the volume of the possible blocks fall. Table 1 shows the different classes of the sensitivity
parameters and their classification. The second stage of the analysis consists of the calculation of the
sensitivity from the notes given to the sensitivity parameters mentioned earlier. This calculation of the
sensitivity is adapted and adjusted to the present case according to the available data and is done in a
way similar to the methods used for the RMC of Bieniawski (1989). The calculation of the sensitivity of
the studied cliff follows the next equation:

S=((4 JS)+(2 JR)+(HF)+(DIR)+(2 DIP))x(1/3)

The third and the fourth stages of the analysis consist of crossing rules tables in order to find the
possibility (the activity vs the sensitivity), and another crossing to find the risk (the possibility vs the
intensity). These table crossings define the rules for the fuzzy logic reasoning used in the analysis.
COMPUTER PROGRAMMING

In order to perform the classical analysis as well as the fuzzy analysis, a computer program has been
developed using Visual Basic edition Application (VBA) on EXCEL™. This computer program includes
three modules, the first one is for classical (Discrete) analysis, the second is for statistical simulation
analysis using different types of probability distribution for each parameter, and the third module is
devoted to the analysis using fuzzy sets theory. This program could be adapted in later stages in order to
take into account different parameters and to analyze the risk in other sites encountering the same
types of problems. In the third module of the program, we have proposed two types of fuzzy numbers in
order to describe each observed parameter. The first type is a continuous –Beta shapenumber which
could serve to describe an "around X" number. In this case, the user has to supply only the minimum
and the maximum values of the number. The second type of fuzzy number is a trapezoidal shape
number, which could serve in describing a parameter that has multiple values with multiple possibilities.

FUZZY REASONING

In order to find out the results of the analysis using fuzzy numbers, we have applied the method of a-cut
in order to convert –after normalization- the fuzzy numbers supplied by the user into 50 a-cuts. Then a
fuzzy reasoning is applied to these numbers in accordance with the rules supplied in the tables in order
to find out a fuzzy result for the sensitivity, the possibility, and the risk. This fuzzy reasoning is based on
the Min-Max reasoning method as shown in figure 3. The geotechnical engineers along with the experts
have to define the classes of each parameter interfering in the analysis as well as the classes of the
results for the sensitivity, possibility, and the risk.
RESULTS
a result of the analysis carried on in a part of the cliff. It can be seen that the sensitivity plot which is
situated to the left hand side of figure 4 is located in the second and the third class with a possibility of
82% in the second (Favorable) and an 18% in the third (Unfavorable), and the same possibilities for the
risk plot which is on the right hand side of figure 4. In our analysis, we have used the method of center
of gravity in order to defuzzify the fuzzy results. We can notice that the center of gravity for the
sensitivity is at 51.6 as for the risk it is at 47.55. These values are measured on a pre-defined scale from
0 to 100

CONCLUSION

Since the method of geotechnical risk analysis at the cliff under consideration is based upon the
experience of the geotechnical engineer in charge, a computerized assistance is essential in order to aid
the judgment of the engineer, to normalize the decision, and to automate the procedure of the analysis.
Another need for this type of analysis is to overcome the problems of uncertainties, imprecision, and
vagueness in parameters estimation. The fuzzy algorithms are used here in order to find a feasible
solution although some points have to be re-considered such as the definition of the classes for the
input and the results and also in order to re-define the reasoning rules. This work is critical and requires
the gathering of various experts and specialized engineers in the field of geotechnical engineering. It was
our main concern in this paper to present the methodology of fuzzy reasoning for the quantification of
risk. We have found out that this method is applicable to our case and could be adapted to other cases
and sites