Chapter 2 Of Rock Engineering

2

When is a rock engineering design acceptable

2.1 Introduction

When is a design in rock engineering acceptable? The aim of the following text1 is to
demonstrate that there are no simple universal rules for acceptability nor are there
standard factors of safety which can be used to guarantee that a rock structure will be safe
and that it will perform adequately. Each design is unique and the acceptability of the
structure has to be considered in terms of the particular set of circumstances, rock types,
design loads and end uses for which it is intended. The responsibility of the geotechnical
engineer is to find a safe and economical solution which is compatible with all the
constraints which apply to the project. Such a solution should be based upon engineering
judgement guided by practical and theoretical studies such as stability or deformation
analyses, if and when these analyses are applicable.
Tables 1 to 4 summarise some of the typical problems, critical parameters, analysis
methods and acceptability criteria which apply to a number of different rock engineering
structures. These examples have been drawn from my own consulting experience and I
make no claims that this is a complete list nor do I expect readers to agree with all of the
items which I have included under the various headings. The purpose of presenting these
tables is to demonstrate the diversity of problems and criteria which have to be
considered and to emphasise the dangers of attempting to use standard factors of safety or
other acceptability criteria.
In order to amplify some of the items included in Tables 1 to 4, several case histories
will be discussed in terms of the factors which were considered and the acceptability
criteria which were used.

2.2 Landslides in reservoirs

The presence of unstable slopes in reservoirs is a major concern for the designers of dams
for hydroelectric and irrigation projects. The Vajont failure in 1963 alerted the
engineering community of the danger of underestimating the potential for the
mobilisation of existing landslides as a result of submergence of the slide toe during
impounding of the reservoir.

1
Based upon the text of the Müller lecture presented at the 7th Congress of the International Society for Rock
Mechanics held in Aachen, Germany, in September 1991.

ĐƯỢC CHIA SẺ BỞI: WWW.GEOSOFTVN.COM
(Nếu thấy hữu ích hãy vote và chia sẻ nhé bạn)

SHARE BY: WWW.GEOSOFTVN.COM
(If you find useful, please vote and share other)

ACTION PAR: WWW.GEOSOFTVN.COM
(Si vous trouvez utiles, s'il vous plaît vote et d'actions, autres)

SHARE ПО: WWW.GEOSOFTVN.COM
(Если вы найдете полезную, пожалуйста, голосовать и обмениваться другой)

シェア：WWW.GEOSOFTVN.COM
（見つかった場合は、投票を共有、他のご便利です）

共享：WWW.GEOSOFTVN.COM
（如果您发现有用，请投票和分享其他）

20 Chapter 2: When is a rock engineering design acceptable

Landslides in reservoirs 23

During the construction of the Mica and Revelstoke dams on the Columbia River in
British Columbia, Canada, several potential slides were investigated. Two of these, the
Downie Slide, a 1.4 billion cubic metre ancient rock slide, and Dutchman’s Ridge, a 115
million cubic metre potential rock slide, were given special attention because of the
serious consequences which could have resulted from failure of these slides (Imrie
(1983), Lewis and Moore (1989), Imrie, Moore and Enegren (1992)).
The Downie Slide and Dutchman’s Ridge are located in steep, narrow, V-shaped
sections of the Columbia River valley which has been subjected to several episodes of
glaciation. The bedrock at these sites consists mainly of Pre-Cambrian para-gneisses and
schists within or on the fringe of the Shuswap Metamorphic Complex. In both cases, the
potential slide planes, determined by diamond drilling and slope displacement
monitoring, are relatively flat-lying outward-dipping tectonic faults or shears which
daylight in the base of the river valley.
Based on thorough investigation and monitoring programs, British Columbia Hydro
and Power Authority (BC Hydro) decided that remedial measures had to be taken to
improve the stability of both the Downie Slide and Dutchman’s Ridge. These remedial
measures consisted of drainage adits extending within and/or behind the failure surfaces
and supplemented by drainholes drilled from chambers excavated along the adits. Work
on the Downie Slide was carried out in the period 1977 to 1982 (which included a 3 year
observation period) and work on Dutchman’s Ridge was carried out from 1986 to 1988.

Figure 2.1 : Section through Dutchman’s Ridge showing potential slide
surface and water levels before and after drainage.

A section through Dutchman’s Ridge is given in Figure 2.1 and this shows the water
levels in the slope before reservoir filling and after reservoir filling and the construction
of the drainage system. Figure 2.2 shows contours of reduction in water levels as a result
of the installation of the drainage system which consisted of 872 m of adit and 12,000 m
of drainhole drilling. Note that the drawdown area on the right hand side of the potential
slide was achieved by long boreholes from the end of the drainage adit branch.


Figure 2.2 : Contours of water level reduction (in metres) as a
result of the implementation of drainage in Dutchman’s
Ridge.

Comparative studies of the stability of the slope section shown in Figure 2.1, based
upon a factor of safety of 1.00 for the slope after reservoir filling but before
implementation of the drainage system, gave a factor of safety of 1.06 for the drained
slope. This 6% improvement in factor of safety may not seem very significant to the
designer of small scale rock and soil slopes but it was considered acceptable in this case
for a number of reasons:

1. The factor of safety of 1.00 calculated for the undrained slope is based upon a ‘back-
analysis’ of observed slope behaviour. Provided that the same method of analysis and
shear strength parameters are used for the stability analysis of the same slope with
different groundwater conditions, the ratio of the factors of safety is a very reliable
indicator of the change in slope stability, even if the absolute values of the factor of
safety are not accurate. Consequently, the degree of uncertainty, which has to be
allowed for in slope designs where no back-analyses have been performed, can be
eliminated and a lower factor of safety accepted.
2. The groundwater levels in the slope were reduced by drainage to lower than the pre-
reservoir conditions and the stability of the slope is at least as good if not better then
these pre-reservoir conditions. This particular slope is considered to have withstood
several significant earthquakes during the 10,000 years since the last episode of
glaciation which is responsible for the present valley shape.
3. Possibly the most significant indicator of an improvement in stability, for both the
Downie Slide and Dutchman’s Ridge, has been a significant reduction in the rate of
down-slope movement which has been monitored for the past 25 years. In the case of
the Downie Slide, this movement has practically ceased. At Dutchman’s Ridge, the
movements are significantly slower and it is anticipated that they will stabilize when
the drainage system has been in operation for a few more years.

Deformation of rock slopes 25

2.3 Deformation of rock slopes

In a slope in which the rock is jointed but where there are no significant discontinuities
dipping out of the slope which could cause sliding, deformation and failure of the slope is
controlled by a complex process of block rotation, tilting and sliding. In an extreme case,
where the rock mass consists of near vertical joints separating columns of massive rock,
toppling movement and failure may occur.
Figure 2.3 is a section through part of the power tunnel for the Wahleach
hydroelectric project in British Columbia, Canada. A break in the steel lining in this
power tunnel occurred in January 1989 and it is thought this break was caused by a slow
down-slope gravitational movement caused by block rotations within a near-surface zone
of loosened jointed rock.

Figure 2.3 : Cross-section through a section of the Wahleach
power tunnel showing the original tunnel alignment and the
location of the replacement conduit. The deformed shape of the
hillside is shown (exaggerated) and the replacement pressure
tunnel was located in a zone of small deformations.

The Wahleach project is located 120 km east of Vancouver and power is generated from
620 m of head between Wahleach Lake and a surface powerhouse located adjacent to the
Fraser River. Water flows through a 3500 m long three metre diameter unlined upper
tunnel, a rock trap, a 600 m two metre diameter concrete encased steel lined shaft
inclined at 48° to the horizontal, a 300 m long lower tunnel and a 485 m long surface
penstock to the powerhouse.
The tunnels were excavated mainly in granodiorite which varies from highly fractured
and moderately weathered in the upper portions of the slope to moderately fractured and
fresh in both the lower portions of the slope and below the highly fractured mass. Two
main joint sets occur in the rock mass, one set striking parallel to the slope and the other
perpendicular to it. Both dip very steeply. Average joint spacings range from 0.5 to 1 m.


A few joints occur sub-parallel to the ground surface and these joints are most well
developed in the ground surface adjacent to the inclined shaft. Thorough investigations
failed to reveal any significant shear zones or faults oriented in a direction conducive to
sliding.
The toe of the slope is buried beneath colluvial and fan deposits from two creeks
which have incised the Fraser Valley slope to form the prominence in which the inclined
shaft was excavated. This prominence is crossed by several linear troughs which trend
along the ground surface contours and are evidence of previous down-slope movement of
the prominence. Mature trees growing in these troughs indicate a history of movement of
at least several hundred years (Moore, Imrie and Baker (1991)).
The water conduit operated without incident between the initial filling in 1952 and
May 1981 when leakage was first noted from the upper access adit located near the
intersection of the inclined shaft and the upper tunnel (see Figure 2.3). This leakage
stopped when two drain pipes embedded in the concrete backfill beneath the steel lining
were plugged at their upstream ends. Large holes had been eroded in these drainage pipes
where they were not encased in concrete and it was concluded that this corrosion was
responsible for the leakage. This conclusion appeared to be valid until 25 January, 1989
when a much larger water flow occurred.
Investigations in the dewatered tunnel revealed a 150 mm wide circumferential tension
crack in the steel lining of the upper tunnel, about 55 m from its intersection with the
inclined shaft. In addition, eight compressional buckle zones were found in the upper
portion of the inclined shaft. Subsequent investigations revealed that approximately 20
million cubic metres of rock are involved in down-slope creep which, during 1989-90,
amounted to several centimetres per year and which appears to be ongoing. This down-
slope creep appears to be related to a process of block rotation rather than to any deep
seated sliding as was the case at both the Downie Slide and Dutchman’s Ridge.
While discrete element models may give some indication of the overall mechanics of
this type of slope deformation, there is no way in which a factor of safety, equivalent to
that for sliding failure, can be calculated. Consequently, in deciding upon the remedial
measures to be implemented, other factors have to be taken into consideration.
After thorough study by the BC Hydro and their consultants, it was decided to
construct a replacement conduit consisting of an unlined shaft and tunnel section and a
steel lined section where the rock cover is insufficient to contain the internal pressure in
the tunnel. This replacement conduit, illustrated in Figure 2.3, will remove the steel lined
portions of the system from zones in which large displacements are likely to occur in the
future. This in turn will minimise the risk of a rupture of the steel lining which would
inject high pressure water into the slope. It was agreed that such high pressure water
leakage could be a cause for instability of the overall slope. Further studies are being
undertaken to determine whether additional drainage is required in order to provide
further safeguards.
Careful measurements of the displacements in the inclined shaft, the length of the steel
lining cans as compared with the original specified lengths and the opening of the tensile
crack in the upper portion of the steel lined tunnel, provided an overall picture of the
displacements in the rock mass. These observed displacements were compared with
displacement patterns computed by means of a number of numerical studies using both

Structural failures in rock masses 27

continuum and discrete element models and the results of these studies were used in
deciding upon the location of the replacement conduit.
In addition to the construction of this replacement conduit to re-route the water away
from the upper and potentially unstable part of the slope, a comprehensive displacement
and water pressure monitoring system has been installed and is being monitored by BC
Hydro (Baker (1991), Tatchell (1991)).

2.4 Structural failures in rock masses

In slopes, foundations and shallow underground excavations in hard rock, failure is
frequently controlled by the presence of discontinuities such as faults, shear zones,
bedding planes and joints. The intersection of these structural features can release blocks
or wedges which can fall or slide from the surface of the excavation. Failure of the intact
rock is seldom a problem in these cases where deformation and failure are caused by
sliding along individual discontinuity surfaces or along lines of intersection of surfaces.
Separation of planes and rotation of blocks and wedges can also play a role in the
deformation and failure process.
An analysis of the stability of these excavations depends primarily upon a correct
interpretation of the structural geological conditions in the rock mass followed by a study
of the blocks and wedges which can be released by the creation of the excavation.
Identification and visualisation of these blocks and wedges is by far the most important
part of this analysis. Analysis of the stability of the blocks and wedges, and of the
reinforcing forces required to stabilize them, is a relatively simple process once this
identification has been carried out.
The Río Grande Pumped Storage Project is located in the Province of Córdoba in the
Republic of Argentina. Four reversible pump-turbines operating at an average head of
170 m give the project a total installed capacity of 750 MW. These turbines are installed
in a 25 m span, 50 m high, 105 m long cavern at an average depth of 160 m .
The rock in which the underground excavations are situated is a massive tonalitic
gneiss of excellent quality (Amos et al (1981)). The gneiss has an average uniaxial
compressive strength of 140 MPa. The maximum principal stress, determined by
overcoring tests, is 9.4 MPa and is almost horizontal and oriented approximately normal
to the cavern axis. In massive rocks, this 15:1 ratio of uniaxial strength to maximum
principal stress is unlikely to result in any significant failure in the rock and this was
confirmed by numerical stress analyses (Moretto (1982), Sarra Pistone and del Río
(1982)). The principal type of instability which had to be dealt with in the underground
excavations was that of potentially unstable blocks and wedges defined by intersecting
structural features (Hammett and Hoek, (1981)). In one section of the cavern, the axis
of which is oriented in the direction 158-338, four joint sets were mapped and were found
to have the following dip/dip direction values:

Table 5. Dip and dip direction values for joints in one location in the Río Grande cavern
N. Dip Dip dir. Comments
1 50 131 infrequently occuring joints
2 85 264 shear joint set
3 70 226 shear joint set
4 50 345 tension joint set


Figure 2.4 : Perspective view of Río Grande
power cavern showing potentially unstable
wedges in the roof, sidewalls and bench.

Figure 2.4 is a perspective view of the Río Grande power cavern showing typical wedges
which can be formed in the roof, sidewalls, bench and floor by joint sets 2, 3 and 4.
These figures represent the maximum possible sizes of wedges which can be formed and,
during construction, the sizes of the wedges were scaled down in accordance with
average joint trace lengths measured in the excavation faces. In Figure 2.4 it is evident
that the roof and the two sidewall wedges were potentially unstable and that they needed
to be stabilised. This stabilisation was achieved by the placement of tensioned and
grouted rockbolts which were installed at each stage of the cavern excavation. Decisions
on the number, length and capacity of the rockbolts were made by on-site geotechnical
staff using limit equilibrium calculations based upon the volume of the wedges defined
by the measured trace lengths. For those wedges which involved sliding on one plane or
along the line of intersection of two planes, rockbolts were installed across these planes
to bring the sliding factor of safety of the wedge up to 1.5. For wedges which were free to
fall from the roof, a factor of safety of 2 was used. This factor was calculated as the ratio
of the total capacity of the bolts to the weight of the wedge and was intended to account
for uncertainties associated with the bolt installation.
The floor wedge was of no significance while the wedges in the bench at the base of
the upstream wall were stabilised by dowels placed in grout-filled vertical holes before
excavation of the lower benches.
Early recognition of the potential instability problems, identification and visualization
of the wedges which could be released and the installation of support at each stage of
excavation, before the wedge bases were fully exposed, resulted in a very effective
stabilisation program. Apart from a minimal amount of mesh and shotcrete applied to
areas of intense jointing, no other support was used in the power cavern which has
operated without any signs of instability since its completion in 1982.

Excavations in weak rock 29

2.5 Excavations in weak rock

In contrast to the structurally controlled failures in strong rock discussed in the previous
section, there are many cases where tunnels and caverns are excavated in rock masses
which are weak as a result of intense jointing or because the rock material itself has a low
strength. Rocks such as shales, mudstones, siltstones, phyllites and tuffs are typical weak
rocks in which even moderate in situ stresses are likely to induce failure in the rock
surrounding underground excavations.
Progressive failure of this type, which can occur in the rock surrounding an
underground excavation in a weak rock mass, is a difficult analytical problem and there
are no simple numerical models nor factor of safety calculations which can be used to
define acceptable limits to this failure process. Judgement on the adequacy of a support
design has to be based upon an evaluation of a number of factors such as the magnitude
and distribution of deformations in the rock and the stresses induced in support elements
such as grouted cables, steel sets or concrete linings. This design process is illustrated by
means of an example.
The Mingtan pumped storage project is located in the central region of the island of
Taiwan and utilizes the 400 m head difference between the Sun Moon Lake and the
Shuili River to generate up to 1600 MW at times of peak demand. The power cavern is
22 m wide, 46 m high and 158 m long and a parallel transformer hall is 13 m wide, 20 m
high and 17 m long. The caverns are 45 m apart and are located at a depth of 30 m below
surface in the steep left bank of the Shuili river (Liu, Cheng and Chang (1988)).
The rock mass consists of weathered, interbedded sandstones, siltstones and shales
dipping at about 35° to the horizontal. The Rock Mass Ratings (RMR) (Bieniawski
(1974)) and Tunnelling Quality Index Q (Barton, Lien and Lunde (1974)) and
approximate shear strength values for the various components of the rock mass are given
in Table 6 below.

Table 6. Rock mass classifications and approximate friction angles φ and cohesive strengths c for
the rock mass in which the Mingtan power cavern is excavated
Rock type RMR Q φ ' degrees c’ MPa
Jointed sandstone 63-75 12-39 50 1.0
Bedded sandstone 56-60 7-31 45 0.8
Fautle or shears 10-33 0.1-1.1 30-40 0.15-0.3

Weak beds of siltstone, up to 2 m thick, appear to have caused a concentration of shear
movements during tectonic activity so that fault zones have developed parallel to the
bedding. The common feature observed for all these faults is the presence of continuous
clay filling with a thickness varying from a few mm to 200 mm. The cavern axis is
intentionally oriented at right angles to the strike of these faults, as shown in Figure 2.5.
The measured in situ stresses in the rock mass surrounding the cavern are
approximately
Maximum principal stress (horizontal) σ max = 10.9 MPa
Minimum principal stress (vertical) σ min = 7.5 MPa
The maximum principal stress is parallel to bedding and normal to the cavern axis.


Figure 2.5 : Orientation of the underground underground excavations in relation
to the faults in the bedded sandstone surrounding the power cavern and
transformer hall of the Mingtan Project. The red plane indicates the dip and strike
of the faults.

Bedding faults of significant thickness which were intersected in the roof of the cavern
were treated by using high pressure water jets to remove the clay and then filling the
cavities with non shrink cementitious mortar (Cheng (1987), Moy and Hoek (1989)). This
was followed by the installation of 50 tonne capacity untensioned grouted cables from a
drainage gallery 10 m above the cavern roof in order to create a pre-reinforced rock mass
above the cavern. All of this work was carried out from construction adits before the
main contract for the cavern excavation commenced.
The initial design of the reinforcing cables was based upon experience and precedent
practice. Figures 2.6 and 2.7 give the lengths of rockbolts and cables in the roof and
sidewalls of some typical large powerhouse caverns in weak rock masses. Plotted on the
same graphs are empirical relationships suggested by Barton (1989) for bolt and cable
lengths for underground powerhouses.
During benching down in the cavern, 112 tonne capacity tensioned and grouted
cables were installed on a 3 m x 3 m grid in the sidewalls. The final layout of the cables
in the rock surrounding the power cavern and the transformer hall is illustrated in Figure
2.8. Five metre long grouted rockbolts were installed as required at the centre of the
squares formed by the cable face plates A 50 mm layer of steel fibre reinforced micro-
silica shotcrete was applied within 5 to 10 m of the face. This shotcrete was later built up
to a thickness of 150 mm on the roof and upper sidewalls and 50 mm on the lower
sidewalls where it would eventually be incorporated into the concrete foundations.


Figure 2.6 : Lengths of rockbolts and cables used for roof support in
some large caverns in weak rock. Equations defining trend lines
were suggested by Barton (1989).

Figure 2.7 : Lengths of rockbolts and cables used for sidewall
support in some large caverns in weak rock. Equations defining
trend lines were suggested by Barton (1989).


A key element in the decision making process on the adequacy of the support system was
a monitoring and analysis process which involved the following steps :

1. Displacements in the rock surrounding the excavations monitored by means of
convergence arrays and extensometers, some of which had been installed from
construction galleries before excavation of the caverns commenced.
2. Numerical modelling of each excavation stage using non-linear multiple-material
models. The material properties used in the models of the early excavation stages
were adjusted to obtain the best match between predicted and measured
displacements.
3. Prediction of displacements and support loads during future excavation stages and
adjustment of support capacity, installation and pre-tensioning to control
displacements and cable loads.
4. Measurement of displacements and cable loads (using load cells on selected cables
which had been de-bonded) and comparison between measured and predicted
displacements and cable loads.
5. Installation of additional cables or adjustment of cable loads to control unusual
displacements or support loads.

The aim of this program was to maintain as uniform a displacement pattern around the
excavations as possible and to keep the loads on the cables at less than 45% of their yield
load. The intermediate rockbolts and the shotcrete were not accounted for in the
numerical modelling since it was assumed that their role was confined to supporting the
rock immediately adjacent to the excavations and that the overall stability was controlled
by the 10 to 15 m long grouted cables.
Figure 2.8 shows the results of a two-dimensional finite element analysis of the power
and transformer caverns. The combination of materials used in analysing this section
through the caverns, assuming that the bedding faults can be represented by horizontal
layers in the two-dimensional model, is indicated. The model was excavated in seven
stages to simulate the excavation stages used in the field. Support was installed at each
excavation stage.
In order to match the measured and predicted displacements in the rock mass, it was
found that a 2.5 m thick zone of softened and weakened material had to be wrapped
around the excavations to account for blast damaged material (The properties of this
damage zone are E = 3000 MPa, φ = 45 degrees and c = 0.32 MPa).
In Figure 2.9, the predicted and measured displacements along six extensometers
installed in the power cavern sidewalls are compared. The overall agreement is
considered to be acceptable. Maximum sidewall displacements were of the order of 100
mm at the mid-height of the upstream wall, adjacent to one of the major faults.
Elsewhere, displacements were of the order to 25 to 46 mm.
Figure 2.10 shows the results of monitoring at seven stations along the axis of the
power cavern. Before excavation of the cavern commenced, extensometers were installed
at each of these stations from a drainage gallery above the roof arch and from
construction galleries as shown in the upper part of Figure 2.10. In addition, load cells
were installed on cables adjacent to some of the extensometers.


Rapid responses were recorded in all extensometers and load cells as the top heading
passed underneath them. Further responses occurred as the haunches of the cavern arch
were excavated and as the first bench was removed. As can be seen from the plots, after
this rapid response to the initial excavation stages, the displacements and cable loads
became stable and showed very little tendency to increase with time. The difference in
the magnitudes of the displacements and cable loads at different stations can be related to
the proximity of the monitoring instruments to faults in the rock above the cavern arch.
The rapid load acceptance and the modest loading of the cables together with the
control of the displacements in the rock mass were the goals of the support design.
Measurements obtained from the extensometers and cable load cells indicate that these
goals have been met.

Figure 2.8 : Displacement contours obtained from a two-dimensional
finite element analysis of the seven excavation stages of the Mingtan
power and transformer caverns. The layout of cables used to support
the rock surrounding the caverns and the location and properties of the
rock units are indicated in the figure. Shear failure of the rock mass is
represented by the × symbols while tensile failure is shown by the ο
symbols. Note that almost all of the failures are within the envelope
defined by the ends of the cables. Different types of cables are shown
in different colours and failure of the cables (under the powerhouse
drainage gallery) is shown in red. The deformed cavern boundaries are
shown and these indicate a uniform distribution of deformations
abound the caverns. The maximum sidewall displacement for the in situ
stresses and material properties used in this analysis is 90 mm.


Figure 2.9: Comparison between calculated and measured
displacements along six extensometers installed in the
sidewalls of the Mingtan power cavern.

Figure 2.10 : Surface displacements and cable loads measured
at seven stations along the power cavern axis.

Factor of safety 35

2.6 Factor of safety

The four case histories, discussed in previous sections, have been presented to
demonstrate that a variety of criteria have to be considered in deciding upon the adequacy
of a rock structure to perform its design objectives. This is true for any design in rock
since the performance of each structure will be uniquely dependent upon the particular
set of rock conditions, design loads and intended end use.
In one group of structures, traditional designs have been based upon a `factor of
safety’ against sliding. These structures, which include gravity and fill dams as well as
rock and soil slopes, all involve the potential for sliding along well defined failure
surfaces. The factor of safety is defined as the factor by which the shear strength
parameters may be reduced in order to bring the slope (or dam foundation) into a state of
limiting equilibrium (Morgenstern (1991)). The numerical value of the factor of safety
chosen for a particular design depends upon the level of confidence which the designer
has in the shear strength parameters, the groundwater pressures, the location of the
critical failure surface and the magnitude of the external driving forces acting upon the
structure.

Figure 2.11 : Hypothetical distribution curves representing the
degree of uncertainty associated with information on driving
stresses and shear strengths at different stages in the design of a
structure such as a dam foundation.

Figure 2.11 illustrates a set of hypothetical distribution curves representing the degree
of uncertainty associated with available information on shear strength parameters and
disturbing stresses for different stages in the design of a rock or soil structure. The factor
of safety is defined as A/B where A is the mean of the distribution of shear strength
values and B is the mean of the distribution of driving stresses. For the purpose of this
discussion, the same factor of safety has been assumed for all three cases illustrated.
During preliminary design studies, the amount of information available is usually very
limited. Estimates of the shear strength of the rock or soil are generally based upon the


judgement of an experienced engineer or geologist which may be supplemented, in some
cases, by estimates based upon rock mass classifications or simple index tests. Similarly,
the disturbing forces are not known with very much certainty since the location of the
critical failure surface will not have been well defined and the magnitude of externally
applied loads may not have been established. In the case of dam design, the magnitude of
the probable maximum flood, which is usually based upon probabilistic analysis,
frequently remains ill defined until very late in the design process.
For this case, the range of both available shear strength and disturbing stresses, which
have to be considered, is large. If too low a factor of safety is used, there may be a
significant probability of failure, represented by the section where the distribution curves
overlap in Figure 2.11. In order to minimise this failure probability, a high value for the
factor of safety is sometimes used. For example, in the 1977 edition of the US Bureau of
Reclamation Engineering Monograph on Design Criteria for Concrete Arch and Gravity
Dams, a factor of safety of 3.0 is recommended for normal loading conditions when ‘only
limited information is available on the strength parameters’. This value can be reduced to
2.0 when the strength parameters are ‘determined by testing of core samples from a field
investigation program or by past experience’.
During detailed design studies, the amount of information available is usually
significantly greater than in the preliminary design stage discussed above. A
comprehensive program of site investigations and laboratory or in situ shear strength tests
will normally have been carried out and the external loads acting on the structure will
have been better defined. In addition, studies of the groundwater flow and pressure
distributions in the rock mass, together with modifications of these distributions by
grouting and drainage, will usually have been carried out. Consequently, the ranges of
shear strength and driving stress values, which have to be considered in the design, are
smaller and the distribution curves are more tightly constrained.
The case histories of the Downie Slide and Dutchman’s Ridge, discussed earlier, are
good examples of designs based upon back-analyses. In both of these cases, very
extensive site investigations and displacement monitoring had established the location of
the critical failure surfaces with a high degree of certainty. Careful monitoring of the
groundwater in the slopes (256 piezometer measuring points were installed in
Dutchman’s Ridge) had defined the water pressures in the slopes and their fluctuations
over several years. Some shear testing on fault material recovered from cores was carried
out but, more importantly, the mobilized shear strength along the potential failure
surfaces was calculated by back-analysis, assuming a factor of safety of 1.00 for existing
conditions.
Figure 2.11 illustrates the hypothetical distribution curves for the range of values for
shear strength and driving stresses for the case of a structure in which an existing failure
has been carefully back-analyzed. Depending upon the degree of care which has been
taken with this back-analysis, these curves will be very tightly constrained and a low
factor of safety can be used for the design of the remedial works.
This discussion illustrates the point that different factors of safety may be appropriate
for different stages in the design of a rock structure. This difference is primarily
dependent upon the level of confidence which the designer has in the values of shear
strength to be included in the analysis. Hence, a critical question which arises in all of
these cases is the determination or estimation of the shear strength along the potential

Factor of safety 37

sliding surface. In a paper on the strength of rockfill materials, Marachi, Chan and Seed
(1972) summarize this problem as follows: `No stability analysis, regardless of how
intricate and theoretically exact it may be, can be useful for design if an incorrect
estimation of the shearing strength of the construction material has been made’.
Except in simple cases involving homogeneous soils or planar continuous weak
seams, determination of the shear strength along potential sliding surfaces is a
notoriously difficult problem. This is particularly true of the determination of the
cohesive component, c’, of the commonly used Mohr-Coulomb failure criterion.
Laboratory test specimens tend to be too small to give representative results while in situ
tests are difficult and expensive and, unless carried out with very great care, are liable to
give unreliable results.

Table 7: Factors of safety for different loading conditions in the design of earth and rockfill dams.
Loading condition S.F. Remarks
End of construction porewater pressures in the 1.3
dam and undissipated porewater pressures in
the foundation. No reservoir loading.

Reservoir at full supply level with steady state 1.3 Possibly the most critical (even if
seepage in the dam and undissipated end-of- rare) condition.
construction porewater pressures in the
foundation.

Reservoir at full supply level with steady state 1.5 Critical to design.
seepage.

Reservoir at probable maximum flood level 1.2
with steady state seepage conditions.

Rapid reservoir drawdown from full supply 1.3 Not significant in design. Failures
level to minimum supply level very rare and, if they occur, usually
shallow.

For failure surfaces which involve sliding on rough or undulating rock surfaces such as
joints or bedding planes, the methodology proposed by Barton (1976) is appropriate for
estimating the overall shear strength of the potential sliding surface. This involves adding
a measured or estimated roughness component to the basic frictional strength which can
be determined on sawn and polished laboratory shear test specimens.
For heavily jointed rock masses in which there are no dominant weakness zones such
as faults or shear zones, a crude estimate of the shear strength of the rock mass can be
obtained by means of the use of rock mass classification systems as proposed by Hoek
and Brown (1988).
In all cases, a greater reliance can be placed upon the frictional component, φ, of the
Mohr-Coulomb shear strength equation and extreme care has to be taken in the
estimation of the cohesive strength, c’. Where no reliable estimates of this value are
available from carefully conducted shear tests or from back-analysis of existing failures,


it is prudent to assume a cohesive strength of zero for any stability analysis involving
structures such as dam foundations.
In the design of fill and gravity dams there is a tendency to move away from the high
factors of safety of 2 or 3 which have been used in the past, provided that care is taken in
choosing sensible conservative shear strength parameters, particularly for continuous
weak seams in the foundations. An example of the range of factors of safety which can be
used in the design of earth or rockfill dams is given in Table 7.

2.7 Probabilistic analyses

The uncertainty associated with the properties of geotechnical materials and the great
care which has to be taken in selecting appropriate values for analyses has prompted
several authors to suggest that the traditional deterministic methods of slope stability
analyses should be replaced by probabilistic methods (Priest and Brown (1983),
McMahon (1975), Vanmarcke (1980), Morriss and Stoter (1983), Read and Lye (1983)).
One branch of rock mechanics in which probabilistic analyses have been accepted for
many years is that of the design of open pit mine slopes. This is because open pit planners
are familiar with the concepts of risk analysis applied to ore grade and metal price
fluctuations. Probabilistic methods are used in estimating the economic viability of
various options in developing an open pit mine and so it is a small step to incorporate the
probability of a geotechnical failure into the overall risk assessment of the mine. The
mine planner has the choice of reducing the probability of failure by the installation of
reinforcement, reducing the angle of the slope or accepting that failure will occur and
providing for extra equipment which may be needed to clean up the failure. Since the
mine is usually owned and operated by a single company and access to the mine benches
is restricted to trained personnel, accepting a risk of failure and dealing with the
consequences on a routine basis is a viable option.
On the other hand, the emotional impact of suggesting to the public that there is a
finite risk of failure attached to a dam design is such that it is difficult to suggest the
replacement of the standard factor of safety design approach with one which explicitly
states a probability of failure or a coefficient of reliability. The current perception is that
the factor of safety is more meaningful than the probability of failure. Even if this were
not so, there is still the problem of deciding what probability of failure is acceptable for a
rock structure to which the general public has access.
In spite of these difficulties, there does appear to be a slow but steady trend in society
to accept the concepts of risk analysis more readily than has been the case in the past. The
geotechnical community has an obligation to take note of these developments and to
encourage the teaching and practical use of probabilistic as well as deterministic
techniques with the aim of removing the cloak of mystery which surrounds the use of
these methods.
Fortunately, there is a compromise solution which is a form of risk analysis used
intuitively by most experienced engineers. This is a parametric analysis in which a wide
range of possibilities are considered in a conventional deterministic analysis in order to
gain a ‘feel’ for the sensitivity of the design. Hence, the factor of safety for a slope would
be calculated for both fully drained and fully saturated groundwater conditions, for a
range of friction angles and cohesive strengths covering the full spectrum which could be

Probabilistic analyses 39

anticipated for the geological conditions existing on the site, for external forces ranging
from zero to the maximum possible for that slope. The availability of user-friendly
microcomputer software for most forms of limit equilibrium analysis means that these
parametric studies can be carried out quickly and easily for most designs.

Chapter 2 Of Rock Engineering

More Related Content

What's hot (19)

Viewers also liked (20)

Similar to Chapter 2 Of Rock Engineering (20)

More from Ngo Hung Long (13)

Chapter 2 Of Rock Engineering