UNIVERSIDADE DO MINHO

JORNADAS DE ENGENHARIA CIVIL

Guimares, Portugal

GEOTECHNICAL RISK IN
RAILWAYS

Author: Antnio Campos e Matos

April 2009

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

INTRODUCTION:
In Civil Engineering, we teach and learn how to safely build bridges, buildings, dams, railways,
sanitary landfills, water treatment plants, etc. First we learn how to design and then how to
build.
We quickly forget that all that we design and build is surrounded and founded in an environment
of great variability, which interferes with all of our work.
Almost nothing (or even nothing at all) of what we calculate is exact. The parameters, the
models, the construction, even the maintenance and operation, are based in variables or
contain uncertainties or errors.

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

INTRODUCTION:
This variability, which leads to the reality of having to assume that the safety is not unlimited, is
a source of Risk. Structural Risk is a concept that is generally known and accepted with
reasonable consensus.
The necessary safety of the constructions and operations, which the society demands of Civil
Engineering, is provided by Science (Mechanics) and by the Codes. This way, Risk is controlled
both during construction and service.
The width of the concept of Risk, subject that we will not approach at the moment, implies that it
should not be associated only with the negative aspect of the accident. The financial Risk due to
delay or unforeseen events, the Social and Environmental Risk are surely some of the most
important regarding the activity of Civil Engineering.

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

INTRODUCTION:
It happens that the mandatory usage of natural parameters by Geomechanics and Geotechnics
introduces a strong component of variability and Risk in the system, with no equivalent in
Structural Mechanics.
In these cases, for Civil Engineers, as well as for other professions that constantly deal with
Risk, the professional responsibility can assume extremely high proportions. Knowing how to
manage and control this Risk is an obligation of our activity.
It is curious that, even in the meanders of Civil Engineering, the idea that the Risk is over once
the construction of a building is complete seems to linger. The service associated Risk can be
very relevant, both the Risk of defective behavior and the one of loss of performance with time.
This Risk is called Residual Risk and its characterization is, nowadays, an important aspect of
Civil Engineering investigation.

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

INTRODUCTION:
The construction and service of railways, along with geomechanical aspects of the natural or
artificial surroundings, are an extremely important and interesting example of analysis of this
type of Risk. In this presentation, it is intended to use a particular and specific phenomenon of
Geotechnical Risk to expose more general aspects of Risk and Residual Risk areas, in Civil
Engineering.
The presentation is based on a conference which took place in November 2008 in Sabratha,
Lybia, with relatively small updates and adaptations.

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

TOPICS::
TOPICS
1 - Risk General Concept
2 - Risk Engineering Concept
3 - Risk Evaluation
4 - Residual Risk
5 - Residual Risk Management
6 - Residual Risk Control
7 - Model Limitations and Future Developments
8 - Conclusions

1 RISK GENERAL CONCEPT

Risk is a mental perception of danger, not necessarily rational, that spreads across the
societies, changing in concept and acceptable values in time and space, as societies do.
Humans learn risk since the beginning, since they are born. Societys evolution determines
the rules.

Risk is something that affects people and we should learn how to live with it. Civilizations
evolution determines more and more responsibilities related with non-correct risk control
and mitigation. Modern societies with increasing ethics conscience determine that citizens
are more and more concerned with public discussion concerning this subject. This is
important for social and individual risk perception, key of human risk management.

1 RISK GENERAL CONCEPT

If danger perception is an efficient way of risk management and mitigation, then we must
be strong and persistent, informing and disseminating in the society that even after a
construction, intervention, reinforcement or rehabilitation, a Residual Risk remains on the
System. Zero Risk doesnt exist. Immediately after the construction, this risk could be
below codes level, but with time things can change.
In this communication, I mainly would like to transmit that in Geotechnical Engineering, the
Residual Risk can be much higher than one could expect, during and even immediately
after the construction. If so, Risk Control Implementation is an obligation of owners,
contractors and engineers.

1 RISK GENERAL CONCEPT

The self risk perception, and the society risk diffusion, and not only the technical or expert
knowledge, is an absolute need to human safety in our societies. Thats why the discussion
of these matters on a regular basis is an obligation that we need to fulfil, promoting and
increasing citizens perception and specialized experts reaction capacity.
The natural fact that everyone has an idea about risk level, but not about risk concept and
acceptable values, difficults an analytical and more accurate approach, more and more
required by insurance companies, owners, and our society. Thats the engineer case, now.
Civil engineering professionals take permanent decisions. The Risk is there.
But Risk in engineering is not only related with potential problems that directly put in
danger the human life. Risk is also financial, environmental, sustainability, contractual.
Just as an example one could say that, actually, the Risk of Non Control of the Initial Cost
(contractual cost) during underground construction is certainly the biggest risk influence on
geotechnical systems.
Finally and because engineering is not only a safety control activity but also a cost
management activity, one needs to visualise also the opposite risk situation, this is the
Opportunity.

1 RISK GENERAL CONCEPT

Example Project Pittsburgh North Shore
Connector (USA)

Example Results Updates or

Risk Management

2 RISK ENGINEERING CONCEPT

On a mathematical, mechanical and engineering point of view, Risk is absolutely related
with existing unknowns in the system. Graduation of the mathematical level of approach to
real solutions can be made:
Statistic Approaches
If we assume that the unknown is related with controlled parameters and also that it is
possible to recognize and control the parameters variability, then the Risk is correctly
assumed on the system and, on these cases, Statistic Approaches are usable. This is
used on special structures, analysis, long tunnels, in comparison solutions between bridges
and tunnels, etc.

2 RISK ENGINEERING CONCEPT

Deterministic Approaches
However, if we accept that unknowns exist but we dont need, or we cannot, recognize
completely the system`s parameters as variables, we assume a Deterministic Approach
on the identification and characterization in time and space and, as consequence,
uncertainty acceptation remains on the system. Thats the case of some environment risk,
social risk, etc.

Risk Analysis Matrix

2 RISK ENGINEERING CONCEPT

No risk concept
If the system exists without unknowns, then we say that there is no risk on the system. In
some cases, if the risk is very, very small, one could assume that the system has no risk,
but this is not mathematically correct and could be mechanically dangerous. Of course this
is only a practical common sense definition, because there are no exact systems in
engineering, neither in medicine, economy, etc. The variability lives with us, no doubt about
this.

3 RISK EVALUATION
In many engineering issues and particularly in geotechnical issues, where solutions live
together with enormous impact (urban construction, underground metro, railways, dams),
high vulnerability (human presence), it is difficult to control and dominate the variability,
before, during and after the construction. A common risk evaluation is:
RISK= PROBABILITY (SOMETHING NEGATIVE) X VULNERABILITY X IMPACT
- Something negative can be, for instance, the structural strength being lower than the
structural action, and this is a probability (0 - 1)
- Vulnerability is the degree of non protection between the occurrence and the
accidental target (human, environment, etc.) (0-1)
- Impact is the direct and indirect consequence, is the value that occurs if the
vulnerability is total (1). Its a value, or a relative value (human loss, environment affects,)
added to the restitution of the previous situation (indemnity, assurance cover,
reconstruction, etc.)
This definition is important to postulate that, for us, risk:
- Is not only a probability
- Neither a vulnerability
- Neither a consequence

3 RISK EVALUATION

Practical solutions
There is a an important amount of theories related with uncertainty and engineering
systems. The research on these areas is now deep and strong, with conclusions
becoming utilized on some practical applications. However, the proper implementation in
real systems is not so easy, particularly when the variability or the number of parameters
is too high, as is the case of geotechnical systems.
Thats why today I want to talk about possible practical and applicable solutions that show
the way to safety decisions, meanings better risk control in our engineering profession.
Because, on the end of the day, we have the responsibility and intention of reducing the
risk - and so we use organized procedures and methodologies - the sequence of
activities, transverse to Risk Control, can be divided, in my opinion, in three steps:
-

Risk assessment (before construction: project, design, decision support)

Risk management (during construction)

Residual risk management (after the construction, when the operation starts)

3 RISK EVALUATION
On the first step, the risk is not yet a reality, but here we define the correct or incorrect
way we approach the risk issue, therefore the importance of this phase is determinant.
On the second step, unknown, variability, uncertainty remain on the system. However the
system includes now a methodology to control the risk.
On the third step one must find and define methodologies and technologies but always
supported by a good understanding of remaining unknowns and uncertainties on the
system.
For practical reasons, I prefer using the following division of unknown aspects of the
system (on these cases, the parameters):
-

Uncertainties with recognized random variability distribution (variabilities)

Uncertainties recognized, but without a random distribution (uncertainties)

Surprises (if one can not know or suppose the existence)

3 RISK EVALUATION
Common sense speaking:
- Variables, if we know, could know or should know, the existence of such
characteristics and related parameter existing values. Also required the possibility of
random values for parameters distribution (with more or less approximation).
- Uncertainties, if we know, could know or should know, the existence, or the high
potential of existence of such aspects, phenomena, parameters, but because we cannot
see it, factually confirm it, we never can relate them by random distribution. We know that
they exist but we cannot say where they are or when they happen.
-

Surprises, if we dont know or suspect that they could occur.

3 RISK EVALUATION
WHAT WE MAY NOT FORGET IN PRATICAL
Examples:
Risk transmit by information
The geological and geotechnical mass characterization (for instance, obtaining friction
angle, cohesion, water level, in space) is based on drills, geophysical site tests, samples
and laboratories tests. All of these parameters are absolutely random in each case. So,
the final interpretation (geotechnical maps for instance) transmitted to the next person
involved (engineer) is absolutely a prediction or prevision (statistical, deterministic, etc).
The probability that could occur different values, exists always and can be higher than the
engineer could expect and this is Risk based and transmitted by geotechnical prevision.
Geotechnical parameters are Variables.

3 RISK EVALUATION

Third Ring Road Deep Foundations

On the TRR, the design of deep foundations considered ground conditions
variability and also different bearing capacity methods (Bustamante / Reese
and ONeil) in order to reduce Risk.
Th ird Ring Road
IC 11 (pk 0+ 450 to 0+610)
B H-2 (1977)

BH -A1 (2006)

BH -A2 (2006)

SC -3(2007)

N S PT
0

10

20

30

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

Depth (m )

19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

uncementedmateria l
cem ented m ate rial

40

50

60

70

80

90

100

3 RISK EVALUATION

Third Ring Road Deep Foundations

TRR - I C11 (BH_A2)
Axi al Capac ity (SLS) (Bustamante, 1993)

TRR - I C11 (BH_A2)

Axial Capacity (SLS) (Reese & O'Nei l, 1999)

Rcd (kN)

R cd (kN)
0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

8 000

9 000

10 000

Depth (m)

15
20
25

3 000

4 000

5 000

6 000

7 000

8 000

9 000
Diameter (m)

Diameter (m)

0.60

0.60
0.80

0.80

10

1.00

1.00
1.20

1.20

15

Depth (m)

10

2 000

0
5

1 000

20
25

30

30

35

35

40

40

Bearing capacity of drilled shafts computed with two different methods and
four diameters.
Static and dynamic load tests will be performed to access actual capacity.

10 000

3 RISK EVALUATION
WHAT WE MAY NOT FORGET IN PRATICAL
Examples:
An uncertainty, not a surprise, neither a random variable
The case of a tunnel: The previous geological studies indicate that the potential of
faults occurrence is high, and suggest some possible localization.
During the construction, and after 300m of tunnel excavation, any fault occurs. The
optimism was installed on the technical staff, and people start thinking that the geological
prevision was too conservative. Safety and site supervision relaxes on the site. But
suddenly a collapse comes about, with equipment damage.
The mistake was in considering this fact as a surprise. But really, it was only an
uncertainty. They should know, based on the transmitted information, that the faults
system was there, even if not on the exact position.
The optimism is not recommended in geotechnics. We never know all before we face
the problems, and even after that, frequently we go on with a high degree of unknown
(residual risk).

3 RISK EVALUATION

Uncharted fault

3 RISK EVALUATION

3 RISK EVALUATION

3 RISK EVALUATION
WHAT WE MAY NOT FORGET IN PRATICAL
Examples:
A code reduction by risk control objective
The case of the concrete strength acceptable on similar
concrete in different constructions.

The reasons of such strong cut in the stress (approx. 50% !),
and apparently this is an economical error, are related with
two aspects:
- Unknowns remain on the system of geotechnical
characterizations (residual risk)
- Non control (neither visual) of the concrete structure
(pile), and this is completely different from visible building
construction (column)

3 RISK EVALUATION
Example:

3 RISK EVALUATION
RISK CONTROL METHODOLOGIES
Summary
Risk Approaches
- Statistic
- Deterministic
- No risk
Risk Control
- Risk assessment
- Risk management
- Residual risk management
Category of parameters
- Variables
- Uncertainties
- Surprises

4 RESIDUAL RISK
On the core of risks geotechnical complex problems, the rock stability analysis is
perhaps one of the most interesting issues. That is because:
I) Characteristics:
- Rock falls occur suddenly, many times without warning or signals
- Rock masses surfaces are very heterogeneous. The variability is very high
- Collapsing masses are strong and often affect railways, roads, houses, etc.
- The trigger factor is mostly associated with rain effect, seismic effect and human
activity.
II) Stabilization measures
- Important limits are imposed to the designers, mainly related with environment and
economical aspects
-Typical solutions are bolts, shotcrete, anchors, meshes, cable meshes, and dynamic
barriers
Thats why it makes sense to discuss the performance achieved by each solution. If we
assume an important Residual Risk, after the stability intervention, then the evaluation of
Performance of the solutions is an important step towards the understanding of the risk.
It is dangerous to believe -by optimism or ignorance that the implemented solution
absolutely protects us from anything.

4 RESIDUAL RISK

4 RESIDUAL RISK
The main purpose of this presentation is to discuss the Residual Risk Issue in geotechnical
interventions and, particularly, in rock fall stabilizations.
To forget or ignore the residual risk is assuming zero risk situation, and this, even being
deterministically possible, is a dangerous position.

Of course, on scientific and engineering activities, one understands that zero risk is not a
statistical or reliability possible condition. However, in other activities, the zero risk is
assumed as a possible condition, but this common sense opinion is associated with the
non responsibilities opinions.

On the domain of assumed responsibilities, the question is quite different and its
relationship with the idea of non-existence of other risks after a convenient intervention to
control the main risk.

It happens sometimes the owners have a difficulty in managing this potential residual risk,
because the investments in geotechnical stabilizations are understood as a final point on
the matter.

4 RESIDUAL RISK
But it must also be clearly assumed by all the intervenients in geotechnical or geological
processes that in many situations one cannot guarantee the complete resolution of the
previously existing risk. Typically, this RR happens because:
- economical, environmental, social, limited acceptable impacts;
-

the strong rock masses variability and potential surprises;

the designed performance level is not attained on construction end.

At FEUP/GEG, and in association with consultant companies and contractors, the

investigation of some of these questions is going forward. This document presents rock
falls cases of important RR, that resulted either by an assumed impossibility of significant
risk eliminations, or by not reaching the contractual level of stabilization.

4 RESIDUAL RISK
In Structural Engineering, the use of synthetic and fabric materials (concrete, steel,
carbon fibber, etc.) width controlled fabrication, determine a insignificant level of
variability and thats why the reliability safety calculation is more a quasi
deterministic system than a statistical one, except in what concerns with some
actions variability.

Also, we well understand that designer engineers have no responsibility by the material
quality confirmation. He only imposes the desirable parameters values, he doesnt
verify. So the Risk is much more focused on others aspects (mistakes, modulation,
construction, degradation, extreme actions, etc.), than on the materials variations.

Contrary, in typical geotechnical designing projects, engineers have no possibility to

impose or determine the parameters (geomechanics parameters). He needs to identify
and characterize the parameters, and so, he also assumes relevant part of
responsibility by the correct model performance.

As the geotechnical parameters variability is quite strong, the Risk associated can
reach relevant values and appropriated control methodologies needs to be
implemented.

4 RESIDUAL RISK
The rock masses are normally characterized by important anisotropy and heterogeneity
conditions. By comprehensive reasons, surfaces bands of rock masses, and rock
slopes, among all other issues, have the most intensive variability natural conditions
that one can find. As slopes are directly expose to meteorological conditions, the
actions are also difficult to estimate and the quality time degradation is an assumed
fact. All this aspects converge to a huge issue. The consequent Risk increases in
magnitude if one looks to the extension and vulnerability of the influence area.

4 RESIDUAL RISK
Interesting, is the specific rock slope stability behaviour coming from different
geological materials. In fact, granite, sandstone, calcareous schist, presents typical
water behaviour, in relation with the shear strength joint water sensibility, and
accumulated joint water pressure and the drainage. Also some empiric information, as
localized worst characteristics, potential dangerous surprises, etc., is very important on
a global Risk control.

4 RESIDUAL RISK
Deviating a little bit of this presentation, it could be curious to look in parallel for other
consequence of this residual risk and some possible contractual conflict between owner
and contractor in design and build projects, concerning the financial management and
this risk transfer owner objective.

The owner, first responsible by assure safety conditions with limit cost investment, is
many times in difficult decisions conditions mostly because the non-capacity to control a
contract on a strong variability media.
On the relationship with the contractor, the Risk management solutions can be listed on
the following cases:

- Assume (A);
- Share (S);
- Transfer (T)

4 RESIDUAL RISK
The Financial Risk Transfer is perhaps the most important or frequent aspect of the risk
management in geotechnical contracts. The connection with geological variability its
evident.
The owner or the contractor can invest important values in identification and
characterization parameters, however, always remains a unknown volume of
uncertainties and surprises, that must be controlled. Of course this is not easy, but a
balanced equilibrium can be take on the construction contract between both parts.

At all of these cases, the Residual Risk (RR), or the risk that remains after the
intervention, must be identified and characterized. If it remains in low level and
acceptable by the society and the codes, one can forget the question, otherwise the
owner is the first entity that has to assure the resolution.

4 RESIDUAL RISK

The RR evaluation can be made in different steps of the process, depending on the
objective of the study.
The Arrbida rock falls present on the following slides is an example of a Posconstructed RR evaluation and, on this case, it must be used the final characteristics
and parameters, and not the designers prior parameters. Typically an important amount
of geotechnical information came to support this phase.
Other important step, is the tender phase proposal preparation, and the contract
agreement discussion, in which someone must evaluate the remains risk to be
transfered without the support of relevant information. This requires experience,
geological studies, and risk analysis process.
Last but not the least, this kind of studies and its convenient divulgation, can increase
the social perception of risk and this will be one of the most effective actions to control
dangerous situations.

4 RESIDUAL RISK
Among others, we use on this document the common engineer definition:

RISK = (COLLAPSE PROBABILITY) X (VULNERABILITY) X (IMPACT)

The mainly question of the Arrabida Atlantic coast rock fall issue, is that calcareous,
carsick, masses are tremendous heterogeneous, with large extension of potential
blocks detach and long possible trajectories. On this morphology and environment,
nobody would decide by a stabilization based on concept of total, exhaustive active
solutions, the correct is the passive solution concept.

This concept permites the block detachment and intercepts the trajectories by
mechanic systems. Instead of this, the active solution resolves all the stabilization of
possible detachment without permit displacements and movements. The passive
solution assumes that its not possible to guaranty the total interception of potential
trajectories and cinematic energy and so, an RR remains on the final implemented
solution.

4 RESIDUAL RISK
During this studies an innovation methodology had been introduced, quite useful for
rock falls stabilizations risk control:
- assume Vulnerability V as a statistical value, in complete independence with
Collapse probability P;
- assume detach probability P as the detachment block (or local masses)
probability;
- connect the passive solutions (barriers, mesh, etc.) to the Vulnerability;
- the V value (as a probability) is the non interception trajectories probability;
- admit no active solutions implemented (injections, bolts, etc..) ;
- use appropriated models to evaluate P and V;
- RR= P x V x I

4 RESIDUAL RISK
So, the Collapse = Detach Probability P, that we call P, is related to the local collapse,
of small masses, or blocks, or ignition rock fall, and depends on rain intensity, soil water
sensibility, clayed joints, winds, temperatures, and earthquakes. The internal and external
system variability is the well comprehensive parameters variability. P can be evaluated
with stability models and variability on the strength parameters (cohesion, friction), and
external forces (particularly water pressure).
The Vulnerability (V), is the statistical 01 capacity for interception and eliminate the
energy protection against block impact and can be evaluate by rock fall models, assuming
variability in morphology, geology and strength parameters.

RESIDUAL RISK = (Detach Probability) X (Vulnerability Probability) X (IMPACT)

RR = P x V x I

4 RESIDUAL RISK
In soil, rocks and soil and rock slope materials, first the stability evaluation, and after
the stabilization solutions, has both a double behaviour concept (internal and
external), and this is a source of frequently confusion.
Logically, the danger existence is independent of the kind of these phenomena. In spite
of the risk associated with the internal instability is usual greater than the surface or
external instability, one must concern about the strong risk associated with rock surface
instability in roads, railways or urban areas.

4 RESIDUAL RISK
The complete risk control must analyse the two possibility of potential danger.
Since some years, and by geomechanics reasons, we separate the external
stabilization solutions in Active and Passive, being passives if the mass
displacements or block movements are authorized, like protection barriers, steel net
and cables, tunnels, etc., while the actives doesnt permit displacements
(displacements sense associated with discontinuity deformation), like, anchors, bolts,
etc.

What we use now, and after the P & V risk model in rock falls, is the
association of:
- Active like a solution that transform and reduces
significantly P, by increment the local stability.
- Passive like a solution that reduces V, by controlling the
trajectories and energy falls, avoiding hit humans or
properties.

5 RESIDUAL RISK MANAGEMENT

One important intervention has been made in the centre of Portugal Arrbida coast, in
calcareous and sandstones masses, very similar with Libya geological situations. We
study the risk and develop news methodologies for residual risk evaluation.

5 RESIDUAL RISK MANAGEMENT

The Arrbida Natural Park

Atlantic Coast

5 RESIDUAL RISK MANAGEMENT

The Residual Risk, associated to problematic natural slopes, results from:
-Extension and magnitude of the problem;
-High heterogeneity of the rocky slopes;
-Exposition to environmental factors;
-Possible climacteric alterations;
-Vulnerability in case of Railways, Roadways or Sidewalks, etc;
-Geological or geotechnical variability, dimensions, block detaching methods;
- Difficulties in the analysis, problem numerical modelling and risk evaluation.

Opposing to the other geotechnical works, the extension, the economical investment and the
environmental and landscape reasons dont allow the stabilization and treatment of all the slope.

RESIDUAL RISK

5 RESIDUAL RISK MANAGEMENT

- Evaluate the risk level which is covered by the passive systems of protection (Dynamic
Barriers, protection Steel Wire Net);
- Confirm the performance levels and reliability of the systems concerning the land masses
detaching;
- Through statistical analysis and probabilities, evaluate the percentage of occurrences
which are not covered by the installed systems;
- Propose complementary systems based on warnings methods to control this residual risk,
based in realistic monitored measures, (displacements have no real time utility).

5 RESIDUAL RISK MANAGEMENT

RESIDUAL RISK (R.R.) = P x V x Iv
Detach Probability
Assumed non controllable

P Detach Probability [0;1]

V System Vulnerability [0;1]
I v Consequent Impact Value

Energy Failure

P Controlled by ACTIVE solutions

Energy Failure

V Controlled by PASSIVE solutions

R.R.

5 RESIDUAL RISK MANAGEMENT

Local Geology
RECENT Debris deposits formed by reddish silty material with rocky blocks already
consolidated by carbonated cement.

MIOCNE - Areias da Torre Coarse grain carbonated sandstones. 100m thickness

and 25 dip towards north. Medium to low rock strength (c = 6 a 20MPa)

LOWER JURASSIC
- Calcrio de pedreiras Limestones strongly carsted. High strength (c= 60 a
150MPa)
- Dolomitos do Convento Dolomites and limes with big dolomitic cavernous
blocks. High strength rocks (c= 25 a 100MPa)
Several transversal faults, with N-S to NNE-SSE and NNW-SSE orientation.
Rock mass highly fractured. Joints without filling, exposed to weathering processes,
or joints filled with low strength materials.
The rock mass has many caves and empty spaces inside and outside

5 RESIDUAL RISK MANAGEMENT

Rocky slopes with a high number of fractures and joints

5 RESIDUAL RISK MANAGEMENT

Different kinds of material releasing and

instabilities

Potential Rock Instabilities

4

54

5 RESIDUAL RISK MANAGEMENT

Different kinds of material releasing and instabilities

5 RESIDUAL RISK MANAGEMENT

Variability in very decomposed

rock mass

Soil with a wide ranged granulometry

5 RESIDUAL RISK MANAGEMENT

Complex geotechnical
characterization and modelling

Conjugation of several
instability factors

Mixed terrains of soil and rock

5 RESIDUAL RISK MANAGEMENT

The stability is highly dependent on the humidity ratios

Friction Angle vs. Humidity Ratio

Cohesion vs. Hum idity Ratio

80

45

70

40

60

35

Friction Angle ()

Cohesion, c (kPa )

50

50
40
30
20

30
25
20
15

10

10

10

15

20
Humidity Ra tio, w(% )

25

30

0
10

15

20
Humidity Ratio, w (% )

25

30

Studies from another case developed in similar geological conditions

Detach Mechanism Resistance Reduction (Water, accumulated rain) and Water Pressure (Instant rain)

5 RESIDUAL RISK MANAGEMENT

RESIDUAL RISK (R.R.) = P x V x Iv

P-

Analysis of detach probability or rock fall phenomenon beginning based on

software SLIDE (without active control systems)
Stability slope program, based in the limit equilibrium theory of the vertical slice. Determination of
the safety factors or slope instability probability with Monte Carlo Method.

V -

Analysis of vulnerability - probability of installed passive systems failure based on software ROCFALL
Statistical analysis program based in Monte Carlo Method, which analyses the rock fall in slopes.
Permitting the simulation along the slope of an elevated number of rocks, it allows a complete
study of energies, heights, velocities, trajectories, and immobilizations points.

5 RESIDUAL RISK MANAGEMENT

Determination of the critical points of block releasing to use in the cinematic
modelling - Rocfall.
Determination of the detach probability (P) and safety factors of block
releasing or land masses sliding.
R.R. = P x V x Iv
Used Method:
Model SLIDE
Monte Carlo
Variables:
Cohesion cm;
Friction Angle - m ;
Humidity conditions
introduces mechanic variation
Non-circular sliding surface
Calculations to be made in all
the analysed cross-sections.

5 RESIDUAL RISK MANAGEMENT

Results:

Material Properties:
Cohesion cm =50kPa; = 10 kPa
Friction Angle - m = 33; = 5

F.S.m = 1,30
Detach Failure (P) = 17,1%

5 RESIDUAL RISK MANAGEMENT

Characterization
Generally, the following conditions can be seen:
- Existence of large-sized blocks that can generate huge shock energies upon their
release;
- Significant percentage of small and medium-sized blocks that can be controlled with
Dynamic Barriers;

5 RESIDUAL RISK MANAGEMENT

- Very irregular rock fall trajectories, unable to follow a vertical plan - The
detached block never falls on 2D vertical trajectories;
- Complex models of land masses detaching;
- Impossibility to define the fragmentation of the blocks during their falls;
- Highly dependent instabilities with the rainfall (water pressure in short term
and strength reduction in long term);
- Impossibility to install the dynamic barriers, steel wire nets and soil nails in all
the natural slope;

5 RESIDUAL RISK MANAGEMENT

Statistical distribution of the detached blocks dimension

64

5 RESIDUAL RISK MANAGEMENT

Distribuio
Blocksdo
Volume
volume
Distribution
dos blocos [m3]
200
100, 00%

180

% acumulada

160

Frequncia

80, 00%

140
120

60, 00%

100
80

40, 00%

60
40

20, 00%

20

33

42
9

ai
s

52

12
90

% ocornica >90%

3,
08

2,
60

2,
12

32

08

06

3
Vdimensionamento
V
design4m

28
5

Barriers Energy
Energia deDesign
dimensionamento
das[kJ]
Barreiras [kJ]
200
160

Frequncia

80, 00%

140
120

60, 00%

100
80

40, 00%

60
40

% acumulada

100, 00%

180

20, 00%

20

M
ais

90
33 4
42
95
2

85
5

Edimensionam ento
3,
08

2,
60

32
12

80
6
2,
12

30
82

75
7
1,
64

29
52

70
9
1,1

62

69
2

62

0,6

82

25

0,

82
2

66

89

E mdia

20

67
3
75

, 27
-0

, 75
-0

, 23
-1

97
4

86
76

27
4

35
78

57
5
79

, 71
-1

, 19

80

87
5

84

32
-2

, 67

82

17
6

47
6
83
, 15
-3

37

, 00%

81

-2

52
8

27
5

29

1,
64

82
2

1,
16

0,
68

26

92
7

26

09

56

0,
20

67

38

43

75

-0

,2
7

76
,7 5

-0

m dio 3
VV
m 1m

97

48

27

,2
3
-1

,7
1

78

79

57

53

58

-1

,1
9
-2

,6
7

80

82

17

87

63

68
47
83
-2

,1
5
-3

,00%
1

% f alha ourutura < 10%

65

5 RESIDUAL RISK MANAGEMENT

Modelling
Statistical determination of block trajectory after the detaching;
Statistical determination of the energies, velocities, heights and immobilization
points distributions;
Statistical estimative of the vulnerability (V) of the designed protection systems.

R.R. = P x V x Iv

Calculations to be made in all

the analysed cross-sections

5 RESIDUAL RISK MANAGEMENT

Surface Variability Pseudo 3D trajectories analysis according with biggest
dip slope lines and considering a statistic distribution of the crosses vertices
points.

25 handmade cross-sections in a 100m extension zone

according to curve levels and probable trajectories

5 RESIDUAL RISK MANAGEMENT

Materials Variability Restitution coefficients parameter assigned with a
medium value and statistical normal distribution ()
Normal restitution Coef.,
Rn

Tangential restitution
Coef., Rt

Terrain friction angle,

A) Sliding

0,35 0,02

0,85 0,02

25 2

B) Mixed

0,35 0,02

0,85 0,02

17,5 2

C) Rolling

0,35 0,02

0,85 0,02

10 2

Zone

Volumetric Variability Statistical blocks volumetric distribution

Blocksdo
Volume
Distribution
Distribuio
volume
dos blocos [m3]
200
100,00%

180

Frequncia

80,00%

140
120

60,00%

100
80

40,00%

60
40

20,00%

20

52

M
ai
s

29

04

34

12

85

08

% ocornica >90%

3,

2,

60

32

82
30

12
2,

V
ensionam
ento3
Vdim
4m
design

06

7
1,

64

95

28

75
22

70
28

92
16
1,

26

0,

68

25
62

67

0,
20

66

38
9

43
7

-0
,2
7

-0

,7
5

76

97

27
78

Vmdio
3
V
m 1m

75

48
6

53
5
57
-1
,2
3

-1
,7
1

79

87

63
2
,1
9
-2

,6
7

82

80

17

47
83
-2

-3
,1
5

58
4

,00%

68
1

% acumulada

160

5 RESIDUAL RISK MANAGEMENT

Calculation Method for each cross-section analysis:
- Identification of detach local points (blocks, local mass);
- Statistical distribution of the vertical cross-section points;
- Division in different detach methods (slide, roll, both);
- Parameters assign to different zones;
- Distinct develop analysis (energies and maximum height of trajectories);
- Assign different material types along the slope (intact rock, vegetation,
soil, etc);
- Run the calculation software program with 5000 blocks in each analysed
cross-section.

5 RESIDUAL RISK MANAGEMENT

Results:

Trajectories

5 RESIDUAL RISK MANAGEMENT

Results:

Energies

Immobilization Points

5 RESIDUAL RISK MANAGEMENT

Results:
Section
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Nr. of blocks
released
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0
5 00 0

Valid b lock
re lea se d
49 45
49 34
44 58
45 30
48 23
37 17
49 27
45 83
46 18
48 16
48 39
48 37
44 53
40 43
43 44
47 35
40 76
47 83
47 60
47 88
47 85
47 36
45 18
49 34
41 05

Zone
A) Sliding
B) Mixed
C) Rolling

Barrier im pact s
33 08
38 85
11 24
14 56
13 28
16 66
34 13
30 05
26 45
29 53
29 60
27 05
31 18
13 22
26 16
37 59
30 46
43 13
39 63
40 20
40 77
33 62
44 55
46 71
33 86

Nr. Of b lo cks o ver th e

ba rrier
0
0
0
0
0
0
0
21
1 23
42
13
69
4 46
1 86
46
2 85
3 21
5
2 89
3 71
1 94
2 70
8
0
21

% of blocks over the barrier

0 ,00
0 ,00
0 ,00
0 ,00
0 ,00
0 ,00
0 ,00
0 ,70
4 ,65
1 ,42
0 ,44
2 ,55
14 ,3 0
14 ,0 7
1 ,76
7 ,58
10 ,5 4
0 ,12
7 ,29
9 ,23
4 ,76
8 ,03
0 ,18
0 ,00
0 ,62

Weig ht o f the block th at reaches

the barrier capa city (kg )
9 600
10 00 0
8 500
8 500
7 000
10 10 0
9 900
6 500
6 100
7 000
7 600
6 200
5 500
3 800
3 600
3 400
5 100
4 100
5 000
4 900
5 200
2 800
5 700
5 100
6 000

Calculation Example:
To a specific crosssection we obtained

Venergy failure= 5,3%

V trajectory failure = 7,6%

% of blocks that pass over the W eight of the block that reaches the
barriers
barrier capacity (kg)
0,1
8763
6,4
5367
3,8
4850

Conclusions Results:

1 R.REnergy Failure= P x V x I = 0,17x0,053xIv = 0.009 Iv

2 R.RTrajectory Failure = P x V x I = 0,17x0,076xIv = 0.013 Iv

6 RESIDUAL RISK CONTROL

This methods should always be sustained by a data acquisition system,
based on the monitoring of several factors existent on the slopes:
- Deformations / Movements
- Inclinations;
- Vibrations;

Long Time Observation

Cant be used in Alert Systems

- Winds,
- Temperatures;
- Accelerations;
- Humidity ratios;
- Rainfall.

Can support the Alert Systems and

control de Residual Risk

6 RESIDUAL RISK CONTROL

As seen, the principal detachment factor is the rainfall and the water
presence in the local land mass, due to the long time and instant effects.
Being so, we propose a system to monitor, in real time, the local
pluviometric conditions, permitting the control of the R.R. and increasing
the safety through traffics sign or other automatic systems that are
activated when some established maximum limits are reached.

6 RESIDUAL RISK CONTROL

The propose to control the Residual Risk is based in a system that interrupts the
traffic when the humidity ratio (W) or the water level in the local land mass
reaches maximum values. This maximum values are determined throw the
following steps:
1 Laboratory tests in calibrating joints (A);
2 Local Monitoring (B e C)
3 Measuring the local pluviometric curves;
4 Study of the historical relation between the collapses and detachments and the
rainfall occurred;
5 Observation of actual small collapses and relation with cohesion and humidity

7 MODEL LIMITATIONS AND FUTURE DEVELOPMENTS

- The detachment control should also attend to more exact block detachment
phenomenon and not only to local land mass slope collapse;
- Split the rainfall detach factor in water/impulse and water/resistance
reduction;
- Investigation of meteorological historic (maximum rainfall, return period, etc)
- Mathematical and statistical studies of Residual Risk [ R.R = f (P;V) ];
- Developments and implementation of R.R control technologies.

8 CONCLUSIONS

8 CONCLUSIONS
Analysing the results it is clear that the RESIDUAL RISK is still high;
The conducted study is very dependent on the huge variability involved
and the parameters values, reason why they should be very well studied;
The water is the principal factor that promotes the detachment of rocks;
The studies based on probabilities and statistical distributions of the
parameters seems to be the most real approximation to this kind of
problems in rocky slopes;
The RESIDUAL RISK can be controlled, raising the security, with the
installation of some warning systems.

UNIVERSIDADE DO MINHO
JORNADAS DE ENGENHARIA CIVIL
Guimares, Portugal

Thank you very much!

Author: Antnio Campos e Matos

April 2009