Figure 7.

4 – roughly one third of the projects studied were

delayed by up to a month, and a further third suffered more
serious delays. Where data on the cause are available, about a
third to a half of delays were blamed on troublesome ground
conditions. Thus signifi cant delays due to ground conditions
probably occur some 17–20% of the time. It is notable from
Figure 7.4 that the incidence of delays does not seem to have
improved with time.
Van Staveren and Chapman (2007) examined fi gures on the
fi nancial impact of project failures, which were huge. In the
Netherlands alone, failure fi gures are assessed between 5%
and 13% of the yearly expenditure of the Dutch construction
industry (for all failures, not just in the ground). As Dutch
spending on construction totals some 70 billion euro per year,
between 3.5 and 9 billion euro are spent to respond to failures
(van Staveren, 2006). Chapman and Marcetteau (2004) showed
that in the UK, about a third of construction projects are signifi
cantly delayed and of those, half of the delays are caused
by problems in the ground. Applying this ratio to the overall
Dutch construction failure rate would suggest that an average
of some 2.5 billion euro are due to ground-related causes in
the Netherlands, or probably equal to about 50 billion euro
across the whole of the European Union; (the extrapolation is
based on the size of the Dutch GDP in 2006 at US$613 billion
to the whole of the European Union at US$13 600 billion; CIA
factbook, 2007).
These ground-related problems often originate in an earlier
phase than the phase in which they occur, as highlighted by
Chowdhury and Flentje (2007), based on the extensive work
of Sowers (1993). This latter study showed that in 57% of the
projects studied, the geotechnical problems originated from
fl aws in the design phase and in 38% of the cases from the construction
phase. However, these geotechnical problems actually
materialised 41% of the time in the construction phase and
57% in the operational phase of the project.
An important part of the structural and geotechnical design
processes should be for the whole project team to work together
to reduce risks in the ground. Ground risks are one of the major
causes of project delay and when they occur they are seldom
resolved easily or quickly. As the installation of foundations
■ Civil engineering projects suffer from environmental impact (presumably
mitigation too late in the process), poor contractor capabilities
and poor procurement leading to disputes and claims.
It is worth noting that better geotechnical design is an excellent
way to mitigate the uncertainties that the allowance for
optimism bias is meant to counter. By doing so, the allowance
for optimism bias can be reduced – making the project more
affordable. This service is of signifi cant advantage to the project
promoter.
7.4 Typical frequency and cost of ground-related
problems
When anyone asks me how I can best describe my experience
in nearly forty years at sea, I merely say, uneventful. Of course
there have been winter gales and storms and fog and the like
but, in all my experience, I have never been in an accident …
of any sort worth speaking about. I have seen but one vessel in
distress in all my years at sea. I never saw a wreck and have
never been wrecked nor was I ever in any predicament that
threatened to end in disaster of any sort.
Captain Edward John Smith, 1907
(5 years before he became captain of the RMS
Titanic for her maiden voyage.)
A review of 5 000 industrial buildings by the National
Economic Development Offi ce (NEDO, 1983) found 50%
of the projects overran by at least a month. Study of a representative
group of these projects showed that 37% of the
overruns were due to ground problems. Of 8 000 commercial
buildings, a third overran by more than a month and a further
third by up to a month (NEDO, 1988). Half of a representative
Contributory
factor
Project cost %
contribution
to optimism
bias
Works
duration %
contribution to
optimism bias
Example
cost of risk
management
Poor contractor
capabilities
5 5 ￡0
Design complexity 3 2 ￡140 000
Inadequacy of the
business case
23 22 ￡700 000
Poor project
intelligence
6 5 ￡10 000
Site characteristics,
e.g. ground
investigation *
1 3 ￡40 000
* In the example, it is suggested that a Trust has owned a site for at least 20 years
with a comprehensive site investigation carried out within the last 5 years, therefore
only desk study and limited site investigation are needed.
Table 7.7 Main contributory factors to optimism bias for project cost
and works duration for an example project given by HM Treasury
Reproduced from HM Treasury (2003b) c Crown Copyright
Copyright © ICE Publishing, all rights reserved.
Geotechnical risks and their context for the whole project
ICE Manual of Geotechnical Engineering c 2012 Institution of Civil Engineers www.icemanuals.com 65
■ it has an extreme impact;
■ despite being an outlier, once happened, human nature can concoct
explanations for its occurrence – it is therefore obvious, but
only with the benefi t of hindsight.
Professor Taleb identifi es many examples, ranging from the
9/11 events to the ‘winner takes all’ phenomenon for the wages
of pop stars, footballers and artists. His thesis is that Black
Swan events happen all too frequently and tend to dominate the
success of many endeavours. He distinguishes between ‘mild
randomness’ and ‘wild randomness’. Dealing with ground risks
is potentially a wildly random type of Black Swan because the
consequences can be so severe, and so it is important that the
possibility of unexpected events should not be discounted in
geotechnical assessments.
It is possible by good planning to minimise the likelihood of
a Black Swan. However, by the very nature of a Black Swan,
their incidence can never be entirely eliminated. Those who
manage complex geotechnical projects need to appreciate the
possibility that they may be subject to Black Swans – a sense of
humility is needed, plus an understanding of the range of catastrophic
discoveries that can arise on any apparently innocuous
site. Often Black Swans arise because engineers rely purely
on previous experience and don’t consider what might happen.
This point is considered in greater detail in the example
towards the end of this paper.
7.6 Importance of site investigation
Without site investigation, ground is a hazard.
Site Investigation Steering Group, 2003
and construction of basements are usually on the project’s
critical path, any delay affects all follow-on activities, and can
impact directly on completion date.
7.5 Expect the unexpected
… as we know, there are known knowns; there are things we
know we know. We also know there are known unknowns;
that is to say we know there are some things we do not know.
But there are also unknown unknowns – the ones we don’t
know we don’t know.
Former U.S. Secretary of Defense,
Donald Rumsfeld, explained a state of intelligence at a US
Defense Department Briefi ng on February 12, 2002
To a large extent, this awareness of what might happen underpins
successful geotechnical design. Managing risks for the
‘unknown unknowns’ is diffi cult as it is tricky to identify a
likelihood and severity for unimagined events, which are not
amenable to statistical understanding.
The concept of ‘the ‘Black Swan’ was introduced by
Nassim Nicholas Taleb (2007) – the name based on the unassailable
belief by people in the Old World that all swans were
white, confi rmed by centuries of empirical evidence, until
black swans were encountered in Australia in January 1697
(by Dutch explorer Willem de Vlamingh, http://birding.about.
com/od/birdsswans/a/blackswan.htm). Taleb has created a
whole theory based on the Black Swan, to which he attributes
the following characteristics:
■ as an outlier event, it lies outside the realm of regular expectations
because no prior experience can point to the possibility;
Figure 7.4 Typical UK delays for projects against time
The sources, shown below each project type, are all listed in the reference section of this chapter
%100
50 Proportion
delayed by
> 1 month
Proportion of
these delays due
to ground
problems
Proportion
delayed by
up to 1 month
70
67
17
33
%100 %100
Public projects
%100
Proportion
delivered
late
BAA
estimate of
industry
average
BAA
estimate of
own jobs
40
17.5
Proportion
delayed by
>1 month
Proportion of
>1 month delays
due to ground
problems
19
5000
Industrial
buildings
NEDO, 1983
8000
Commercial
buildings
NEDO, 1988
Public
projects
NAO, 2001
The Economist
“Blue skies thinking”
Anon, 2005
Copyright © ICE Publishing, all rights reserved.
Context
66 www.icemanuals.com ICE Manual of Geotechnical Engineering c 2012 Institution of Civil Engineers
The site investigation should also address constructability –
it is therefore important that likely foundation types and processes
needed on the site should be considered before the site
investigation takes place. The risk of delay will be considerably
increased if the contractors cannot derive useful information
on the effi cacy of installation for different types of foundation
from the site investigation reports.
It is worth remembering that a site investigation is a careful
but selective sampling of the ground, therefore the vast majority
of the ground will not be encountered until main construction
takes place. Using the soil volumes described in Appendix A
of this chapter, a typical ground investigation will only recover
some 0.03% of the soil beneath the site; only a fraction of that
will be examined by a professional geologist or engineer.
The GeoQ process, introduced by van Staveren in 2006, is
a formalisation of normal good geotechnical practice, emphasising
the need for early data gathering. Figure 7.5 shows a
conventional contract, with the tendering stage between design
and construction. A simplifi ed example of applying the GeoQ
steps, during a number of normal project stages, is shown
in Table 7.8.
As explained in Section 4 of this manual, site investigation
consists of two main activities, as explained in BS5930 (BSI,
1999):
■ Desk study, also known as Preliminary Investigation by
Eurocode 7 (BSI, 2004) – the gathering of information about
a site from documentary sources, made easier by free online
aerial photography and other easily available resources, such as
Envirocheck in the UK. A very valuable part of the desk study
process is often to summarise all the ground-related hazards
onto one site plan which can be used throughout the whole
design and construction process. For larger (and more complex)
sites, a geographic information system (GIS) can fulfi l
the same function.
■ Ground investigation – the physical investigation of a site by boreholes
and trial pits, with laboratory testing and reading of piezometers,
to establish the ground and groundwater conditions. The
ground investigation should be driven by the hazards identifi ed
in the desk study, together with an assessment as to how those
hazards may affect the intended development proposals and construction
processes.
In addition to normal geotechnical considerations like stratigraphy
and stratum properties, it is important that the following
aspects of a site are also investigated:
■ potential presence of valuable archaeological remains;
■ potential presence of contaminated ground or groundwater in the
vicinity;
■ potential presence of obstructions or other impediments to
foundation installation, as well as neighbouring foundations,
tunnels, etc.;
■ buried cables and other services.
Often these man-made issues are a bigger risk for the project
than normal geological variability. Desk studies that only consider,
for instance, easily available mapping and not mapping
data that go back to the start of the site’s development, should
be seen as potentially defi cient.
GeoQ Step Normal project stage Typical example
1. Gathering project
information
Desk study Collecting historical and geological maps and other
information
2. Risk identifi cation Discovery of old maps showing a fi lled-in watercourse
crossing the site
3. Risk classifi cation Normally done during the ground investigation stage,
when the magnitude and extent of the hazard can be
quantifi ed
Prove that channel is up to 10 m in depth and infi lled with
soft organic alluvium
4. Risk remediation Normally done during design, when the development can
protect itself against the main risks
Design of different foundations for that part of the site;
inclusion of methane protection measures for basement
5. Risk evaluation Prior to tender; collection of unmitigated risks so they can
be managed and/or transferred to the contractor as part of
the contract documents
Assess remaining extent of issues connected with channel
6. Risk mobilisation – Filling
in the risk register
Highlight to contractor that his plant may get bogged
down in alluvial formation; care of piling rig stability
Table 7.8 A simplifi ed example of GeoQ steps during a few normal project stages
Figure 7.5 The GeoQ risk management model
Feasibility Pre-design Design Contracting Construction Maintenance
Six GeoQ risk management steps:
1. Gathering project information
2. Identifying risks
3. Classifying risks
4. Remediating risks
5. Evaluating remaining risks
6. Filling in risk register
Copyright © ICE Publishing, all rights reserved.
Geotechnical risks and their context for the whole project
ICE Manual of Geotechnical Engineering c 2012 Institution of Civil Engineers www.icemanuals.com 67
■ having to pay any dispute costs, which can be several times more
than the original loss being argued over.
The developer’s art involves predicting, potentially several years
ahead, when there will be most demand for property and initiating
the planning, design and construction process in suffi cient time
to provide property that meets the demand. This is illustrated in
Figure 7.6. Thus, he is already vulnerable to uncertainties in the
state of the property market when his facility is completed, as
well as uncertainties in construction cost and programme.
If the developer skimps on the site investigation, a saving
of half of the investigation cost might translate into ￡22 500
‘saved’ (but often a poorly thought-out site investigation makes
very little saving).
Omitting dispute and occupier costs (as well as contractor
liquidated damages), the delay costs for 1 month might be:
Client – delay interest ￡583 000
Main contractor
Establishment ￡100 000
Extra investigation ￡30 000
Redesign ￡30 000 ￡160 000
￡743 000
There may also be increased construction costs due to a less
effi cient foundation design caused by the inadequate site investigation.
As an example, the London District Surveyors Association
(2000) allows a design basis for piles with a good site investigation
that is 20% more effi cient than one that is defi cient. Thus a
poor site investigation may result in piles that are some 20% less
effi cient, which may cost 10% more to construct. Thus for the
example under consideration, the extra foundation costs may be
￡60 000 more, increasing the ￡743 000 to ￡803 000.
The extra ‘average’ cost for defi cient ground investigation
can be calculated as ￡209 000, spread over all developments.
This fi gure includes an average extra delay cost due to unexpected
ground conditions of ￡149 000 based on 20% of the
calculated typical delay cost of ￡743 000, and an allowance for
less effi cient design of ￡60 000.
7.7 Costs and benefi ts of site investigation
Quality is never an accident; it is always the result of intelligent
effort.
John Ruskin (1819–1900)
Geotechnical risks can be managed and mitigated. Sound geotechnical
practices – for example hazard recognition and exploration
by means of the desk study and ground investigation – can
reduce risks substantially. The value of these investments needs
to be seen in the context of the risks reduced.
For the building considered earlier and described in Appendix
A of this chapter, a site investigation could consist of:
Desk study ￡5 000
Ground investigation
5 boreholes 20–40 m ￡25 000
10 trial pits ￡5 000
Contamination testing ￡5 000
Soils laboratory testing ￡5 000 ￡40 000
￡45 000
This needs to be seen in the context of the ￡24 million building
cost excluding fi t-out costs (or ￡13.2 million structural cost or
￡2.4 million substructure cost). This site investigation is equal
to 0.19% of the building cost, 0.34% of the structure cost,
and 1.9% of the substructure cost. Compare this with Rowe’s
(1972) estimate for site investigations of between 0.05% and
0.22% of building cost.
The relative costs for delay can be compared to the savings
that may have contributed to them. Using the statistics given
earlier, it can be assumed that some 20% of projects are signifi
cantly delayed by ground conditions, by perhaps 1 month
or more. From the perspective of the developer, the cost of
1 month’s delay can be calculated as follows:
Land purchase cost ￡68.8 million
Building cost (with fi t-out) ￡31.2 million
Total Cost ￡100 million
Taking 7% annual return, the annual cost to service the loan
is ￡7 million or ￡583 000 per month. If the delay prevents the
occupier from working in the building then the costs will be
much higher, as they will also have to include the lost productive
effort of the people working there, which is assumed to
exceed the cost of employing them – a further ￡31.25 million
per year (equivalent to ￡2.6 million per month) or nearly fi ve
times more than the building fi nance costs.
Of course the loss of use of the building could be much
higher as these fi gures do not include:
■ missing the top of a property rental cycle, so that the achieved rent
over subsequent years is less;
■ missing an important season, such as Christmas for a shop or summer
for a holiday facility;
State of
economy
Boom
Recession
to complete
a development
at top of cycle...
Time
...need
to start
a process
much
earlier
Figure 7.6 Graph showing how the start point of a development
process is key to hitting the top of the economic cycle
Reproduced from Chapman, 2008; all rights reserved
Copyright © ICE Publishing, all rights reserved.
Context
68 www.icemanuals.com ICE Manual of Geotechnical Engineering c 2012 Institution of Civil Engineers
7.9 Mitigation steps
The rain came down, the streams rose, and the winds blew and
beat against that house; yet it did not fall, because it had its
foundation on the rock. But everyone who hears these words of
mine and does not put them into practice is like a foolish man
who built his house on sand.
Bible – Matthew 7: 25–26
The standard project work stages (based on RIBA’s Outline
Plan of Work – RIBA, 2007) are shown in Table 7.9, along
with typical geotechnical design activities.
Timely intervention with appropriate advice at the best time
is needed for all projects. Geotechnical advisors need to be
aware of the timescale over which the rest of the project is
being designed and the potential implications of late advice,
particularly for more complex projects where there should be
much early interaction with other parts of the design team,
especially the structural designers.
Often geotechnical elements will present a major constraint
on other parts of the development – for instance on a tight
site, a complete retaining wall system might be up to 2 m wide
(e.g. 0.15 m for guide wall, 0.9 m for hard/soft secant wall,
0.15 m for tolerances and protuberances to specifi cation, reinforced
concrete facing wall of 0.30 m, void of 0.15 m, blockwork
wall of 0.15 m, giving a total thickness of 1.80 m), yet
sometimes architects hope that a 0.3 m reinforced concrete
wall (implicitly expecting an open-cut solution) will be possible.
This will have severe implications for space planning
around the complete perimeter of the building, and may have
worse consequences if a parking layout has been derived for the
smaller wall size. It is therefore important for the geotechnical
team to make sure that their indicative solutions are provided
early. So, for instance, not providing a retaining wall design
until the results of some advanced laboratory tests are known
(and hence relatively late in the overall design process) – while
protecting the geotechnical designer from possibly having to
change his solution – may cause consternation amongst the
rest of the design team.
Whenever a test is carried out on a site, it carries the risk
of not conforming to the specifi cation. It is therefore wise to
test earlier (and ideally off the project’s critical path) instead
of leaving important test results to be measured later and at a
time when a non-conformance will bring greatest disruption.
As examples:
■ A geotechnical baseline report (GBR), as recommended by the
British Tunnelling Society (2003), is a good way of ensuring clarity
on risk sharing for all project types. Rather than giving comprehensive
coverage of the ground conditions as in a geotechnical
interpretative report, a GBR sets out the expected conditions
simply, so that deviations from them are clear. It therefore seeks
to provide clarity regarding when extra payment to a contractor is
due, thereby reducing the likelihood of disputes.
■ A ground investigation should concentrate on the greatest uncertainties
usually revealed by the desk study. A ground investigation
Thus a saving of half of the site investigation budget of ￡22 500
would have to be balanced against a conservative estimate for a
delay of more than ￡800 000 on the fi fth of projects where the risks
materialise, or an average extra cost over all projects of ￡209 000.
In many cases, the consequences are much more severe than the
fairly optimistic assumptions made in this illustration.
7.8 Mitigation not contingency
It is unwise to pay too much, but worse to pay too little; when
you pay too much, you lose a little money, that’s all. When you
pay too little, you sometimes lose everything, because the thing
you bought was incapable of doing the things it was bought to
do. The common law of business balance prohibits paying a little
and getting a lot. It can’t be done. If you deal with the lowest
bidder, it is as well to add something for the risk you run. And
if you do that, you will have enough to pay for something better.
There is hardly anything in the world that someone can’t
make a little worse and sell a little cheaper – and people who
consider price alone are this man’s lawful prey.
John Ruskin (1819–1900)
Often when faced with a low tender price, clients concentrate
on the direct saving and not on the consequential losses to
which they may be subjected. They may believe that a contingency
will provide them with protection, rather than choosing
to resolve the issues directly. This can be valid when the consequences
are ‘mildly random’ but not when they may cause the
sort of severe ‘wildly random’ impact on project fi nances that
were illustrated earlier.
Many developers bear the fi nancial risk themselves, and if things
go wrong in a way that invalidates the assumptions in their fi nancial
model, they are left to sort out the consequences. The main
problems are usually delays, resulting in late handover (delaying
receipt of income streams), or higher costs. Often a disappointed
developer will wish to recover losses from his design and
construction team, for instance by liquidated damages for delay
from the contractor, or via the designers’ professional indemnity
insurance policies. These are not reliable methods for recovering
losses, and especially consequential losses. Better ways of covering
risks are to insure them explicitly by project insurance that
covers the whole team, or by latent defects insurance (LDI) that
covers the fi nal building (although LDI doesn’t always cover consequential
losses). While these insurance products encourage a
shared approach to resolving problems, they don’t let designers
and constructors completely off the hook. For instance, the insurer
may later decide to try to recover some of his losses (by a subrogated
claim) if he suspects negligence was a cause.
LDI should not be seen as a solution by itself, in the same
way that ever tighter liabilities on contractors and designers
are unlikely to resolve a deteriorating position to the benefi t of
a developer. The best way to avoid the sort of issues that result
in delays and losses is to avoid the problems in the fi rst place.
Thus spending more on proper investigation of ground issues
is likely to be more fruitful than engaging lawyer