вход по аккаунту



код для вставкиСкачать
ice | manuals
Chapter 100
doi: 10.1680/moge.57098.1489
Observational method
Dinesh Patel Arup, London, UK
This chapter describes the use of the observational method (OM) on engineering projects.
Traditional ground engineering projects are usually based on a single, fully developed,
robust design. This can lead to conservative but costly designs. The objective of OM is to
achieve greater overall economy by having less conservative designs without compromising
safety. Successful implementation of OM relies on a structured approach to both design and
construction, an assessment of the most realistic design parameters, comparing these with
the cautious design parameters used in traditional designs, having a rigorous monitoring and
observation strategy, having a design/construction which in the light of the monitoring data
can be modified in a timely way, and having a strong management and reporting structure
with well predefined contingency plans in case problems occur. These key ingredients to the
successful implementation of OM are described with the help of some case examples. OM can
also be used when a project, designed by traditional methods, is in difficulty and the ‘best
way out’ approach is required; this is also discussed. Finally, Eurocode 7 (EN1997-1:2004) (EC7)
allows the use of OM in design, but has shortcomings which are also highlighted.
100.1 Introduction
This chapter aims to provide guidance to engineers on the use
of the observational method (OM) on engineering projects.
It describes the important differences between carrying out
designs based on the traditional approach, using ‘characteristic values’ of parameters (defined in Eurocode 7: Geotechnical
Design, Part 1: EN1997-1:2004 (EC7)) and those developed
using realistic parameters (‘most probable’) when using OM.
It also explains that traditional designs are by their very nature
conservative and not flexible to change, but that OM allows
the project team to produce an integrated approach to both the
design and construction, which can yield significant cost and
programme savings. It also fosters good working relationships
between all the team members and can be very satisfying, but
only if it is implemented properly.
The use of OM is permitted in EC7 but engineers should
be aware that there are some shortcomings, which need to be
understood before applying EC7. The use of OM reduces the
safety margins on design compared with traditional designs, but
this can be managed within a rigorous monitoring and observation strategy, implemented within a coordinated team, to still
give safe designs. Before applying OM, the management of
ground risks and pre-agreed contingency plans (should things
not go according to plan) need to be carefully evaluated and
explained to the client/stakeholders. Approval for implementation of OM should be agreed with the client in the knowledge
that it provides benefits and that there may be some drawbacks
(e.g. implementing pre-agreed contingency plans), even if the
risks are low. Approval from other third-party checking engineers may also be required before implementing OM.
Whilst the OM approach is intended to provide overall economy, obviously there are higher associated costs for increased
instrumentation and monitoring, for implementing a stronger
Fundamentals of OM
implementation and pros
and cons of its use 1491
OM concepts and
Implementation of
planned modifications
during construction 1497
‘Best way out’
approach in OM
Concluding remarks 1500
management team to implement the OM, for increased design
services and reviewing of the monitoring data. However, on
complex projects much of this cost may already be accounted
for as instruments and monitoring are required for other reasons, and usually there is already a strong contractor’s management team established on site.
The OM approach is not intended to be used where there is
a risk-averse client, where there is likely to be a lack of total
commitment from any member of the project team, where
there is no thorough investigation of the ground and water conditions, and where there is a likelihood of a rapid or brittle
mode of failure of the ground (including temporary structural
This chapter guides the engineer through the principles of
the use of OM on projects from inception (referred to as the
‘ab initio’ approach) and also, on projects under construction which for unexpected reasons are running into difficulty
(referred to as ‘best way out’ approach). It relies heavily on the
work of some key authors, named below, amongst others:
■ The Observational Method in Ground Engineering, CIRIA C185
(Nicholson et al., 1999);
■ Peck (1969);
■ Powderman and Nicholson (1996).
These authors provide plenty of examples on the structured
implementation of OM on engineering projects. Additional
examples from Europe are also given in a recent study carried out
by a working party on the use of OM in Europe (GeoTechNet,
2005); this includes seven case examples of the ‘structured’ use
of OM on both building and civil engineering projects and also
highlights the overall cost and programme savings made, which
the reader might find useful.
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1489
1/10/2012 6:42:10 PM
Construction verification
100.1.1 Differences between Peck’s and CIRIA’s
approach to OM
The observational method has developed over the last 60 years
and was initially applied on a ‘trial and error’ basis to improve
designs. It was not until Peck’s Rankine lecture (1969) that an
integrated process for predicting, monitoring, reviewing and
modifying designs was advocated operating within a framework
of OM, and without compromising the safety of the structure.
To successfully implement OM, Peck (1969) identified that it
was necessary to have two designs compared with the traditional
single design approach in geotechnics. A range of foreseeable
conditions needed to be considered, which Peck associated with
the most likely condition to happen in practice (‘most probable’)
and the least likely condition to happen (‘most unfavourable’)
(refer to section 100.3.2 for definitions, also illustrated in Figures
100.4 and 100.5). He suggested a design starting with the most
probable (best estimate) condition and varying the design and/or
construction to the planned most unfavourable condition, should
observed behaviour be worse than that predicted based on best
estimate parameters. Peck’s (1969) eight key ingredients for
successful use of OM are given in Table 100.1.
This approach is fundamentally different to the recent work
on OM, published by the Construction Industry Research and
Information Association (Nicholson et al., 1999). The approach
advocated in Nicholson et al. (1999) starts with initial moderately conservative parameters (the same as ‘characteristic’
parameters in EC7), to be relaxed to a likely real situation
(i.e. most probable condition) during construction, should the
observed behaviour warrant it. This approach is also known as
progressive modification to the design and was first suggested
by Powderham (1994). The setting up of a rigorous traffic-light
trigger system (red, amber and green) to deal with uncertainties
in the ground during construction was also established after
Peck’s work. Both these modifications result in improvement
to the use of OM on projects, leading to safer designs.
100.1.2 Definition of OM
The best definition of the OM approach is described in CIRIA
185 (Nicholson et al., 1999) (see Box 100.1). Often engineers
mistake monitoring instruments on a project as following an
OM approach, and Box 100.1 shows that there is a proper
operational framework for carrying out OM and monitoring is
just part of this process.
Box 100.1 Definition of OM (CIRIA 185 (1999))
‘The Observational Method in ground engineering is a continuous,
managed, integrated, process of design, construction control, monitoring and review that enables previously defined modifications to be
incorporated during or after construction as appropriate. All these
aspects have to be demonstrably robust. The objective is to achieve
greater overall economy without compromising safety.’
100.1.3 Comparison between traditional designs
and OM design
Traditional ground engineering projects are usually based on a
single, fully developed, robust design and there is no intention to
vary the design during construction. Instrumentation and monitoring may also be carried out but it plays a very passive role,
to check original predictions are still valid and provide confidence to third-party checkers, e.g. designers for adjacent building owners affected by a development. In CIRIA (1999) this
traditional design is termed ‘predefined design’. In comparison,
in OM the monitoring plays a very active role in both the design
and construction, allowing planned modifications to be carried
out within an agreed contractual framework that involves all the
main parties (client, designer and contractor). The differences in
the two design approaches are illustrated in Table 100.2.
Peck (1969) defined two OM approaches:
■ the ‘ab initio’ approach, adopted from inception of the project;
■ the ‘best way out’ approach, adopted after the project has com-
menced and some unexpected event has occurred that is different to the predefined design or failure occurs, and where OM is
required to establish a way of getting out of a difficulty.
Predefined design process
(traditional design)
The OM process
Permanent works
Temporary works
One set of parameters
Two sets of parameters
One design/predictions
Two designs and predictions
Outline of construction method
Contractor’s temporary works
design/method statement
Integrated design and
construction methods
Methods relate to triggers
Comprehensive and robust
monitoring system
There must be sufficient site investigation
Design is developed on most probable (best estimates) to predict
Monitoring checks predictions not
If checks are exceeded, consider:
(a) best way out approach to design;
or (b) redefine the predefined
design approach reassessing the
geotechnical uncertainties in the
ground (see Table 100.5)
Review and modify process
– Contingency plan
– Improvement plan
Emergency plan
Develop monitoring strategy on calculated values for best case
Perform calculations on most unfavourable conditions
Identify contingency plans for most unfavourable
Monitor and evaluate actual conditions
Modify design to suit actual conditions if triggers are exceeded
OM can only be done if there is adequate time to make decisions
and implement
Table 100.1
Peck’s (1969) eight key ingredients for OM
Emergency plan
Table 100.2 Comparison of the predefined design process and the
observational method
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1490
1/10/2012 6:42:11 PM
Observational method
Case examples presented on the GeoTechNet site give examples
of both the ‘ab initio’ and ‘best way out’ OM approaches. A
paper by Nicholson et al. (2006) suggests a structural framework
for operating the ‘best way out’ approach for recovery of deep,
multi-stage excavation projects when problems occur during construction, like that which occurred at Nicoll Highway, Singapore
(COI, 2005). This operational framework for carrying out the
‘best way out’ is also described in section 100.5 of this chapter.
100.2 Fundamentals of OM implementation
and pros and cons of its use
100.2.1 General
Eurocode 7 does not clearly define the framework to be followed when adopting OM and there are other drawbacks
which can be strengthened if use of CIRIA (1999) is also made
on OM projects, as described below.
100.2.2 OM drawbacks in Eurocode 7 (2004)
The OM method described in Clause 2.7 of EC7 has been
reproduced in Box 100.2 below to illustrate the main drawbacks, which are as follows:
■ Whilst it refers to ‘acceptable limits of behaviour’ it does not define
how these may be derived, since EC7’s premise for design is based
on use of ‘characteristic values’ which present a lower cautious limit
in design, but not the upper limit to represent the most likely behaviour (most probable case), needed to implement the OM approach.
■ No trigger limits are defined to establish planned contingency
actions to check behaviour.
■ There is no operational framework described for management of
the OM within a contract, either within national policy or in a
project organisation.
100.2.3 Operational framework for following OM
The operational framework for implementing OM is illustrated in Figure 100.1. The OM has to be carried out within
the framework of any national and corporate policies governing design codes, specifications, quality management systems and health and safety regulations (e.g. in the UK this
is Health and Safety (HSE) Regulations, and Construction
Design and Management (CDM) Regulations, 1994). This
■ It is primarily aimed at the ab initio approach to OM, although it
does not exclude the ‘best way out’ application of OM.
Box 100.2 Eurocode 7, clause 2.7 (2004)
Observational method
When prediction of geotechnical behaviour is difficult, it can be
appropriate to apply the approach known as ‘the observational
method’, in which the design is reviewed during construction.
The following requirements shall be met before construction is
■ acceptable limits of behaviour shall be established;
■ the range of possible behaviour shall be assessed and it shall
be shown that there is an acceptable probability that the
actual behaviour will be within the acceptable limits;
■ a plan of monitoring shall be devised, which will reveal
whether the actual behaviour lies within the acceptable limits
(the monitoring shall make this clear at a sufficiently early
stage, and with sufficiently short intervals to allow contingency actions to be undertaken successfully);
■ the response time of the instruments and the procedures for
analysing the results shall be sufficiently rapid in relation to
the possible evolution of the system;
■ a plan of contingency actions shall be devised, which may be
adopted if the monitoring reveals behaviour outside acceptable limits.
During construction, the monitoring shall be carried out as
The results of the monitoring shall be assessed at appropriate
stages and the planned contingency actions shall be put into operation if the limits of behaviour are exceeded.
Monitoring equipment shall either be replaced or extended if it
fails to supply reliable data of appropriate type or in sufficient
Figure 100.1 The observational method
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1491
1/10/2012 6:42:11 PM
Construction verification
is represented by the upper box in Figure 100.1 (see also
section 100.3.5 for details).
The second box defines the structure of the key players in
the stakeholder’s organisation (i.e. the client, designer, contractors, third-party checkers and other inspectors), their roles and
responsibilities, and the relationship between organisations
and the individuals. This needs to address the culture of each
organisation, the level of staff training, experience, openness
to communication and management commitment to implementing the OM approach. The stakeholders also need to ‘buy
into’ the technical and commercial risks should any planned
contingency or emergency measures need to be implemented,
even if this risk is considered low.
Once the OM is agreed at ‘project organisation’ level, the
remaining boxes describe the robust management structure
required to implement OM at both design and construction
stage and to control the monitoring and reviewing aspects of the
observational method when the works are on site. The works
have to progress to an agreed plan, with risks being recognised
at each construction phase. Daily construction progress has to
be under the control of a management structure that ensures
any deviation from the method is fully thought through by all
members of the project team and covered by an amendment
to the plan. A monitoring regime has to be set in place, with
competent staff made available to check, review and respond to
all monitoring results within a given timescale from when they
become available. There then needs to be clear instructions to
all involved for all foreseeable situations. Finally, contingency
plans need to be in place that can be rapidly implemented should
preset ‘trigger’ limits be breached or any other unforeseen situation develops.
‘Auditing’, preferably by an independent geotechnical firm,
is essential as it checks that the OM designer and project team
are following established procedures and reaching the correct
technical interpretations. Ideally this should be carried out by a
designer who is unconnected with the OM process.
pre-agreed contingency or emergency plans need to be implemented (see also section 100.4 of this chapter for details). These
plans consisted of increased monitoring, stopping work and
implementing additional propping or berms and/or reverting to
the predefined design. In the event, the measured movements
of the walls during construction were within acceptable limits,
thus allowing considerable savings to be made as the alternative
design allowed the excavation of double height basements to be
sequenced for faster top-down construction.
100.2.5 Pros and cons of OM
OM offers potential savings of time and money and the monitoring provides the needed assurance concerning safety. Some
potential benefits of OM are illustrated in Figure 100.3 and
seven detailed case examples of the benefits provided to clients
are described in GeoTechNet (2005) (see also CIRIA 185).
However, whilst there are significant advantages associated
with the OM application, there are extra costs associated with
prescribing a higher level of management and control on site,
more instrumentation, preparing for contingency plans and
readiness to use back-up plans, e.g. extra propping, and reporting compared with a conventional design situation.
100.3 OM concepts and design
100.3.1 Uncertainty and serviceability
OM is most effective where there is a wide range of uncertainty. Table 100.3 summarises the types of uncertainty that
are often encountered in geotechnical projects.
The OM approach is not suitable where there is a possibility of ‘brittle’ behaviour in the structure or rapid deterioration in the materials which does not allow sufficient warning
to implement any planned modifications (e.g. ‘discovery–
recovery’ contingency plans to be used). Examples of such are
rapid deterioration of soils caused by groundwater or non-ductile failures of structural members (struts/walling connections)
in multi-propped basements.
100.2.4 OM management process on site
At site level, there are usually many layers of contracting organisations involved in a project and all the main players need to
(a) buy into the OM process and (b) have clearly defined responsibility levels. An example of the interaction between these
organisations and the managed reporting of the construction
and monitoring process is given in a paper by Chapman and
Green (2004) for a deep basement project in central London;
this structure chart is illustrated in Figure 100.2. On this project
successful implementation of the OM relied on a clear understanding of the process, roles and responsibilities between the
main contractor (HBG), the groundworks contractor (McGee),
the concreting contractor (Byrne Bros), the instrumentation
contractor (Soil Instruments) and the designer/reviewer (Arup).
The management and reporting structure was defined under
a ‘traffic-light’ system of green, amber and red trigger levels,
so that all parties were clear of their responsibilities, should
100.3.2 Selection of design soil parameters
Where there is a wide range of uncertainty in the soil parameters the OM process in CIRIA 185 uses the terms ‘most probable’ and ‘most unfavourable’ to describe the range of soil
conditions as illustrated in Figure 100.4.
The ‘most probable’ is a set of parameters that represent the
probabilistic mean of all the data, although a degree of engineering judgement must be used in assessing this to take account of
the quality of the data. The ‘most unfavourable’ parameter represents the 0.1% fractile of the data as shown in Figure 100.4,
and this represents the worst value that the designer believes
might occur in practice. The moderately conservative parameter (CIRIA 185) or ‘characteristic value’ of geotechnical
parameters (defined in EC7, clause represents a ‘cautious estimate of the value affecting the occurrence of the limit
state’, and should ideally result in prediction of the upper 5%
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1492
1/10/2012 6:42:12 PM
Observational method
(Arup Geotechnics)
Byrne Bros.
Excavating next
Constructing floor
Soil Instruments
Identifying next
stage construction
Reviewing OM
Correlating OM &
Reviewing values
from monitoring
Taking extra
Adjusting slab
Figure 100.2 OM management and reporting structure on site
Data taken from Chapman and Green (2004)
fractile of the measured deflections as shown in Figure 100.5.
The moderately conservative parameter is therefore not a precisely defined value. It is a cautious estimate of a parameter,
worse than the probabilistic mean but not as severe as the most
unfavourable as shown in Figure 100.4. In assessing these
parameters the designer should carefully consider the quality of
the site investigation data and assess their appropriateness for
use in the OM approach. Often the original data may be appropriate for a more robust ‘predefined design’ approach but may
not be of a higher quality for purposes of implementing OM. In
this case it may be necessary to carry out further investigations,
for instance if there is a lack of groundwater table information.
A typical example of the two sets of OM parameters (most
probable and moderately conservative) which were used for a
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1493
1/10/2012 6:42:12 PM
Construction verification
100.3.3 Serviceability and ultimate limit state
When designing to EC7, checks are required to ensure that the
following ultimate limit states (ULS) are not exceeded:
■ loss of equilibrium of the structure or the ground;
■ internal failure or excessive deformations of the structure or struc-
tural elements;
■ failure or excessive deformations of the ground due to loss of
■ loss of equilibrium of the structure or ground from uplift water
■ hydraulic heave, internal erosion and piping of ground caused by
Figure 100.3 Some potential benefits of OM
hydraulic gradients.
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
Geotechnical uncertainty
Complex geology and hydrogeology
Parameter and modelling
Undrained soil vs drained behaviour
Ground treatment
Grouting, dewatering
Complex temporary work
1 in 1000 1 in 20
Most probable
Moderately conservative
Most unfavourable
Examples of uncertainty in the ground
Characteristic material property (used
in structural engineering)
No of readings
Table 100.3
1 in 2
Soil strength
parameters results
Figure 100.4 Types of soil strength parameters
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
20-m-deep basement wall design into London Clay is illustrated
in Table 100.4 (Chapman and Green, 2004). When selecting
the most probable parameter for use in OM it is important that
the designer can justify the choice; for instance this may be
proven experience from other projects or back analysis of case
EC7 also require checks on ‘serviceability’ limit states (SLS);
states which are less serious than ULS but which are nevertheless undesirable and would need intervention or repair.
In (traditional) predefined designs, calculations are used to
check these states use ‘characteristic values’.
In OM, the acceptable limit of behaviour is a ‘serviceability’ calculation, made using both the ‘most probable’ and
‘characteristic’ parameters and conditions. These provide the
predictions against which the field performance can be monitored and reviewed. Trigger values can be established and contingency plans introduced as illustrated in Figure 100.5 and
Table 100.5.
The OM approach illustrated in Figure 100.5 is for a cantilever wall design but the principle applies to other examples.
The green, amber and red zones represent the trigger limits or
traffic-light control system used in OM. The precise deflections set for the trigger values will depend on the ‘discovery–
recovery’ contingency plans being used and not simply on the
calculated values of predictions made. For instance, if the contingency plans involve a berm in front of the cantilever wall
or introduction of a raking prop, then the time taken to implement these measures will influence the setting of these trigger limits. The triggers may then be based on two criteria: the
first being wall deflection and the second based on the rate of
wall movement to ensure that sufficient time remains to implement the contingency measure. The SLS wall deflection limit
is sometimes used as an easily measured proxy for a range of
undesirable outcomes, including bending moment failure and
buckling of props. It should be noted that OM should not result
in excessive movements and that, if limits were breached, the
risk of a collapse would still be remote.
In respect to ULS predictions, EC7 identifies three sets of
partial factors to apply when assessing the ultimate limit case.
These partial factors are applied to ‘characteristic values’ of
the ground but in essence the ULS design values are then similar to the most unfavourable conditions (see Figure 100.5).
Although, for example in retaining walls, the ULS predictions
are used for assessing structural forces, moments and shear,
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1494
1/10/2012 6:42:12 PM
Observational method
Moderately conservative
Most probable
Conventional design
Made Ground
φ ′ = 25°
φ ′ = 38°
OAP – Broadgate
Taplow Gravel
φ ′ = 36°
φ ′ = 40°
Lehane et al.
London Clay
φ ′ = 23° c′ = 0
φ ′ = 24° c′ = 10 kPa
OAP, Cross Rail
Lambeth Clay
φ ′ = 23° c′ = 0
φ ′ = 28° c′ = 25 kPa
OAP, Cross Rail
Water Pressure in Taplow Gravel
+7.5 mOD
Geotechnical Report
applied from GL
Passive soil
Observations – OAP
Horseferry Road
20 kPa
20 kPa (Perm)
10 kPa (Temp)
0.5 m
Undrained Shear Strength
70 + 7.5z* kPa
+5.5 ≤ d ≥ −11.0 mOD
Results – see Figure 100.4
112 + 5.19z + kPa
London Clay
−11.0 ≤ d ≥ –17.5 mOD
200 kPa
−17.5 ≤ d ≥ –29.2 mOD
400 kPa
Undrained Shear Strength
300 kPa
400 kPa
Results – see Figure 100.4
1000 cu
1500 cu
Back analysis
Lambeth Clay
Stiffness of Clay (Eq)
Table 100.4
Two sets of design parameters used in OM
Reproduced from Chapman and Green (2004)
Predicted most
probable value
No.of readings
“Ideal” distribution of
measured deflections
Predicted EC7
value (SLC)
RED Most
Deflection (δ)
Figure 100.5 Ideal EC7 predicted versus measured performance
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
Most probable
50% likelihood of movement
predictions being exceeded
Characteristic values (EC7) or
moderately conservative (CIRIA185)
5% likelihood of movement
predictions being exceeded
Most unfavourable (CIRIA 185)
0.1% likelihood of movement
predictions being exceeded
Table 100.5 Definitions of most probable, characteristic values and
most unfavourable
Data taken from CIRIA R185 (Nicholson, Tse and Penny, 1999)
the ULS deformations of the wall can also be a useful guide
to determining the maximum predicted movements in the red
zone (Figure 100.5). This ‘upper limit’ of wall deformation (or
curvature) can provide a useful input when developing emergency plans of unexpected behaviour in OM and also with the
‘best way out’ approach in OM, provided the problem has been
identified in time to implement a disaster and recovery plan
before a ULS condition occurs, and with due regard to safety.
100.3.4 Factors of safety
The design values for ULS (stability) calculations are chosen so
that the probability of failure will be acceptably small. It should
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1495
1/10/2012 6:42:13 PM
Construction verification
be noted that the intent of OM is to take out uncertainties in
the ground (see Box 100.2), not reduce factors of safety, when
assessing the ultimate limit state condition in design.
UK standards and those of some other European countries
have traditionally applied ‘factors of safety’ on the design soil
parameter, which can vary depending on the type of foundation
and type of redistribution of load in the ground (e.g. piled rafts
operate differently from single piles per column). EC7, on the
other hand, applies partial factors on the characteristic value of
the ground parameters. Figure 100.6 illustrates how the two
approaches can produce different factors of safety and how
they can vary for different assumed design soil parameters.
100.3.5 Setting OM within the context of a contractual
model and safe design
parties may be severe, and so the application of OM also needs
to consider these factors.
The OM process can be applied to both forms of contract
in the UK model described above. In both cases the design
product comprises:
■ drawings;
■ work specifications and bills of quantities;
■ calculations.
In addition to this, the UK Construction (Design and
Management) (CDM) Regulations 2007 place new duties on
the client, designers and contractors to take health and safety
into account in both the design and construction of a project.
For the designer this means that the design is no longer a set of
calculations but must also:
In the UK, there are essentially two forms of main design
■ address buildability issues;
■ The client appoints a consulting practice to carry out the perma-
■ identify hazards and risks in respect to safety;
nent design (‘engineer design’) and the contractor is responsible
for carrying out the specified works. In this form of contract the
contractor is only responsible for any temporary works design
required to complete the permanent works.
■ The client appoints a ‘design and build’ contractor to complete the
design based generally on an outline or scheme design performed
by a consulting engineering firm.
Other variations to this also occur when a construction manager or project manager is appointed by a client to manage
the overall contract. It should be noted that in some European
countries the contractual model is very different, responsibilities between parties are less clear, the legal obligations on
■ eliminate hazards through good design or, where it is not possible,
to reduce the risk to a low level;
■ show how this process has evolved in the design by producing a
‘risk register’;
■ address impact on adjacent structures (above or below ground).
These regulations are intended to produce stronger links
between the designer and the contractor and minimise risk of
failures, via:
■ production of ‘heath and safety plans’ by both the engineer and
■ appointment of a planning supervisor by the client, who vets these
plans before and during construction;
■ seeking approval from third-party checkers.
Therefore, the CDM Regulations are in line with the OM
objective of integrating the design and construction process to
produce safe designs and construction practices. From work
carried under Work Package 3 of GeoTechNet funded by the
European Commission the UK contractual model with appropriate CDM Regulations and risk assessments do not appear to
exist in all countries in Europe. Again, when carrying out OM
in these countries, the national polices and regulations need to
be fully understood and incorporated within the OM process.
On NATM tunnelling projects, HSC (1996) also describes the
process of managing risk through the use of a ‘discovery–recovery’ model before an unacceptable failure scenario is reached,
and this approach is entirely aligned with the OM approach.
100.3.6 Rapid deterioration
Figure 100.6 Application of factors of safety to different types of
design soil parameters
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
In certain engineering situations, rapid deterioration can be controlled by modifying the construction sequence as follows:
(1) Using the multi-stage construction process – for instance,
an example may be an embankment construction over soft
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1496
1/10/2012 6:42:13 PM
Observational method
clays. In this situation a rapid deterioration in the factor of
safety can be controlled by ‘staged’ filling, between rest
periods, using monitoring to control when the next lift is
done (an example of this is given in Figure 100.7).
(2) Using the incremental construction process – for instance,
an example may be in NATM tunnelling work, where the
rate of advancing the tunnel face and controlling face loss is
a critical component in determining how the ground movements are controlled in the discovery–recovery programme
(see Figure 100.8) using the traffic-light system described
above. This figure shows that the later the problem is discovered, the higher the risk and the longer the structure
remains in a state of reduced stability (red zone). Late instigation of decision-making and recovery would also have the
same effect. In such instances, trigger limits can be set as
both absolute values and/or rate of movements.
The importance of early decision-making to instigate actions
for recovery is an important feature of the UK Health and
Safety Executive (HSC, 1996) discovery–recovery model, and
is a legal requirement for use on all UK construction sites. The
use of trigger values described below, an essential feature of
OM, can also be used in the ‘predefined designs’ (see section
100.1.3) to allow sufficient time for implementation of emergency measures when monitoring is being used.
100.4 Implementation of planned modifications
during construction
100.4.1 Trigger values
As previously mentioned OM uses a ‘traffic-light’ system with
green, amber and red response zones which allow construction
to be controlled, should there be a risk of exceeding the safe
green limit, as follows:
■ green – continue construction;
■ amber – continue with caution and prepare to implement contin-
gency measures, increase rate of monitoring;
■ red – stop progress, do everything possible to slow movements,
implement contingency measures.
In setting the trigger values the following should be noted:
■ The values set may be absolute values or rates of movements or
■ The trigger limits set should also consider the accuracy of the
instrument and whether it is practically measurable.
■ The choice of instruments to measure movements should there-
fore be appropriate for the project and not based simply on lowest
Figure 100.7 Multi-stage construction trigger values
Figure 100.8 Traffic-light system for an incremental excavation
(tunnels) process
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1497
1/10/2012 6:42:14 PM
Construction verification
■ The amber/green limits should be sensibly set based on the design
so that the likelihood of breaching this limit is small (e.g. values
may occasionally stray outside the green zone) but also treated
seriously if it is breached.
■ The contingency/implementation plans should be taken seriously
if the trigger limits are being breached.
■ The choice of the instrumentation should be treated seriously and
the monitoring contractor should be experienced in this type of
The trigger limits should be linked to the ‘most probable’ (green/
amber limit) and the ‘characteristic values’ (amber/red limit) for
implementation of the planned modifications in OM, as illustrated in Figures 100.5, 100.7 and 100.8. A further example, for
use with cantilever retaining wall movements is illustrated in
Figure 100.9. It can be seen that in this example the measured
movements were also well below the most probable conditions.
100.4.2 Monitoring systems
Monitoring systems will vary depending on the type of construction project in which OM is implemented. It is also very
important to define both ‘primary’ and ‘secondary’ monitoring
systems in OM. For example in multi-prop deep basements,
the primary system (e.g. inclinometers and wall-mounted settlement gauges on neighbouring party wall structures) may be
the main instruments relied upon to allow implementation of
any contingency measures in OM, whilst the secondary system (e.g. 3D targets at top of walls or levelling surrounding
ground) might be a more frequent and fast monitoring system
to quickly assess the progress of the excavation works and aid
a broader understanding of the pattern of ground movements
on a site. When considering the amount of instrumentation for
use on a project, considerations should be made in respect to
whether to monitor using remote methods (can be expensive)
or by manual methods (labour-intensive) and if the latter, how
long it will take to carry out a round of monitoring in a single
day. This will then allow assessments of the most important
instruments to be monitored to be made as part of the overall
OM strategy.
An essential part of OM is that the primary system needs to
be immediately repaired if damaged on site, to ensure that OM
can be continued. Chapman and Green (2004) explain the use
of primary and secondary instrumentation for a deep basement
in London, in the context of the trigger limits and the process
owner (see Figure 100.2).
100.4.3 OM quality plans
It is essential to have a quality plan before implementing OM
on a site. This plan would present the designer’s movement
predictions based on the defined construction sequence. Each
stage of construction sequence should also show the acceptable limits of predicted behaviour using the traffic-light system
described above.
Simple graphical outputs that the whole project team can
understand are essential and an example of this is illustrated
in Figure 100.10 for a 20-m-deep basement, excavated in a
top-down manner, in London. This quality plan shows the predicted green and red limits for wall deflection at each dig stage,
for a planned double height excavation technique, to allow
floors to be built quickly. The actual movements recorded
from inclinometers can then be plotted for comparison with
predicted behaviour. These graphs allow the OM reviewer to
make informed judgements and the whole project team are
brought into the decision-making process: whether to continue
to the next dig level, or to implement contingency measures
before continuing. For the example illustrated, contingency
measures involved additional propping, use of natural berms
in front of the wall before proceeding to the next level, and
changing the construction sequence to the original ‘predefined
design’ which involved progressing excavation/floors one after
100.4.4 Construction control
Successful construction control is a vital part of OM; the main
process is as follows.
■ A construction control proforma is used to record all details of
construction operations, strengths of materials exposed during
staged excavations, the fabric and structure of exposed materials
and the deterioration of surfaces exposed to water (see example of
cutting Figure 100.11).
Figure 100.9 Example of trigger limits in retaining walls
Modified with permission from CIRIA R185 (Nicholson, Tse and Penny, 1999).
■ This control has to be fully integrated within the project team,
simple to use and the data easy to read; graphical outputs are
essential for informed decisions to be made (see Figure 100.10).
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1498
1/10/2012 6:42:15 PM
Observational method
Figure 100.10
Typical quality plan developed for a phased top-down basement excavation
Reproduced from Nicholson, Dew and Grose (2006)
Figure 100.11 Example of construction control proforma sheets on
site prepared by a contractor
Reproduced from Nicholson, Dew and Grose (2006)
■ Each process has to have a process owner, with certain levels
of responsibilities and implementing of actions, as illustrated in
Figure 100.2 and described in Chapman and Green (2004). Where
there are many trade sub-contractors involved in a project, the key
organisation for making the decision for reviewing the values
from the site monitoring and implementing planned contingencies
is often the main contractor, who has greater control and powers
to immediately stop works on site.
100.5 ‘Best way out’ approach in OM
The ‘best way out’ approach is used when monitoring checks
in a predefined design (Table 100.1) exceeds, for unexpected
reasons, predicted values but before an emergency condition is
reached. Peck (1969) gives some good examples of the ‘best
way out’ approach in which OM is adopted in response to
some unexpected events or failure to establish a way out of
a difficulty. Nicholson et al. (2006) present a more structured
approach for recovery of deep, multi-stage excavation projects
when problems occur during excavation. These authors studied the post-inquiry reports following the collapse of Nicoll
Highway (COI, 2005), and to avoid such failures occurring
again suggested a ‘best way out’ solution which is described
In the event of a ‘discovery’, the ‘best way out’ approach
would trigger an ‘initial recovery decision-making’ stage as
shown in Figure 100.12.
In all cases this will result in stopping work and/or implementing emergency planned measures to secure the safety of
site staff and the general public while the unexpected event
is fully investigated. This assessment will inevitably be somewhat qualitative rather than quantitative as decisions need to be
made rapidly at this initial stage.
Once the safety of the site has been secured, the project team
can then turn their attention to recovery of the project back to a
fully stable condition which means first carrying out a ‘design
review process’ of the unexpected event, by back analysis of
the actual conditions and comparison with the original design.
This process can be broken down into four processes (termed
‘RADO’) as shown in Figure 100.12. These processes are
briefly described in section 100.5.1 below.
Following this design review, the project team can then consider the following two stages:
■ whether to initiate OM ‘best way out’, in which case this approach
would follow the OM framework illustrated in Figure 100.1 and
described in this chapter; or
■ whether to make a complete re-design based on a traditional, but
revised, predefined design.
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1499
1/10/2012 6:42:15 PM
Construction verification
data should include: soil data and stratigraphy; construction
records; actual sequence of events to inform back analysis process; and observations and physical measurements leading up
to the unexpected event.
PLANNING Process A: back analysis
The purpose of this process is to refine the designer’s understanding of the actual behaviour of the structure and reduce
uncertainty in the design. The process involves: establishing
most probable parameters; developing a satisfactory model
using most probable parameters; comparing results with monitoring data and field observations; reviewing/revising parameters if good agreement is not achieved; and once a reliable
model has been produced, proceeding to design.
EXCEEDED? Process D: verify modified design
This process involves predicting the future behaviour using
the realistic model and set of parameters developed from back
analysis, for the remaining construction stages, but adopting a
level of conservatism into the model. The structure behaviour
should adopt moderately conservative (characteristic) parameters for the serviceability design and ‘worst credible’ or factored parameters for stability checks (see sections 100.3.3 and
SECURE SAFETY Process O: Output plans and triggers
If the OM ‘best way out’ is to be used then the process of OM
as described in this chapter has to be agreed with all stakeholders in the project, with appropriate contingency and monitoring
plans and setting up of trigger values and management teams.
R - Data collection and review
A - Back analysis
D - Modify design-select MC/MP
O - Output and triggers
100.6 Concluding remarks
This chapter provides an overview of the use of OM in engineering projects. It provides a structural framework for carrying out
OM for both the ‘ab initio’ (projects from initial conception)
and ‘best way out’ approaches (traditionally designed projects
running into difficulty). An extensive guide to OM is the UK
CIRIA 185 (1999), and this was consulted widely in preparing
this chapter. The reader is also asked to consult this document
should OM be considered, as it sets out a proper framework
for operating the observational method and is more rigorous
than currently described in Eurocode 7. Other case examples
quoted in this chapter also provide an excellent history of the
use of OM on more recent projects since CIRIA 185.
Figure 100.12
‘Best way out’ operational framework
Reproduced from Nicholson, Dew and Grose (2006)
100.7 References
100.5.1 Four processes of design review (RADO) Process R: data collection and review
This process involves collecting all available data to define
the behaviour of the structure for use in the back analysis.
Particular emphasis should be placed on understanding the
actual conditions and behaviour operating in the field, rather
than justifying the original design assumptions. Sources of
British Standards Institution (2004). Eurocode 7: Geotechnical
Design – Part 1: General Rules. Brussels: Comité Européen de
Normalisation. London: BSI, BS EN1997-1:2004.
Chapman, T. and Green, G. (2004). Observational method looks set
to cut city building costs. Civil Engineering, 157(3), 125–133.
CO1 (2005). Report of the Committee of Inquiry into the Incident at
the MRT Circle Line worksite that Led to the Collapse of the Nicoll
Highway on 20 April 2004. Singapore: Ministry of Manpower.
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1500
1/10/2012 6:42:16 PM
Observational method
GeoTechNet (European Geotechnical Thematic Network) (2005).
Work Package, WP3: Innovative Design Tools in Geotechnics –
Observational Method and Finite Element Method (ed Huybrechts,
N.) BBRI (Nov. 2005). Project with financial support of the
European Commission under the 5th Framework, Project GTC22000-33033.
Health and Safety Commission (HSC) (1994). Managing Construction
for Health and Safety. Construction (Design and Management)
Regulations 1994. Suffolk, England: HSE Books.
Health and Safety Commission (HSC) (1996). Safety of the New
Austrian Tunnelling Method (NATM) Tunnels: A Review of Sprayed
Concrete Lined Tunnels with Particular Reference to London Clay.
Suffolk, England: HSE Books.
Nicholson, D. P. and Penny, C. (2005). The observational method:
application on the railway. Network Rail Earth Works Suppliers
Conference, Birmingham, UK.
Nicholson, D. P., Dew, C. E. and Grose, W. J. (2006). A systematic
‘best way out’ approach using back analysis and the principles of
the observational method. In International Conference on Deep
Excavations, 28–30 June 2006, Singapore.
Nicholson, D., Tse, C. and Penny, C. (1999). The Observational
Method in Ground Engineering: Principles and Applications.
CIRIA Report 185. London: Construction Industry Research and
Information Association.
Peck, R. B. (1969). Advantages and limitations of the observational
method in applied soil mechanics. Géotechnique, 19(2), 171–187.
Powderham, A. J. (1994). The value of the observational method: development in cut and cover bored tunnelling projects. Géotechnique,
44(4), 619–636.
Powderham, A. J. and Nicholson, D. P. (1996). The Observational
Method in Geotechnical Engineering. London: ICE/Thomas
100.7.1 Useful websites
GeoTechNet (European Geotechnical Thematic Network): www.
Health and Safety Executive (HSE):
It is recommended this chapter is read in conjunction with
■ Chapter 78 Procurement and specification
■ Chapter 79 Sequencing of geotechnical works
■ Chapter 94 Principles of geotechnical monitoring
■ Chapter 96 Technical supervision of site works
All chapters in this book rely on the guidance in Sections 1
Context and 2 Fundamental principles. A sound knowledge of
ground investigation is required for all geotechnical works, as set
out in Section 4 Site investigation.
ICE Manual of Geotechnical Engineering © 2012 Institution of Civil Engineers
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1501
1/10/2012 6:42:17 PM
Downloaded by [ University of Liverpool] on [12/11/17]. Copyright © ICE Publishing, all rights reserved.
ICE_MGE_Ch100.indd 1502
1/10/2012 6:42:17 PM
Без категории
Размер файла
1 344 Кб
1489, 57098, moge
Пожаловаться на содержимое документа