11.2 Semi-empirical Approach
161
Table 11.4 (continued)
Category
Design
Build
Operate
D1
D2
D3
CO
OM
Investigation Testing
Analyses and
documentation
Construction Operations
and
monitoring
III Average
Evaluate performances of
nearby
structures
Index tests
on samples
from site
Rational
analyses
using
parameters
inferred from
index tests
Informal
construction
supervision
Annual
inspection by
qualified
engineer
Estimate
subsoil
profile from
existing data
and borings
No filed
measurements
Maintenance
limited to
emergency
repairs
IV Poor
No field
investigation
No
laboratory
tests on
samples
obtained at
the site
Approximate
analyses
using
assumed
parameters
No
construction
supervision
by qualified
engineer
Occasional
inspection
No
construction
control tests
No field
measurements
Adapted from SLM and (Altarejos et al. 2015)
11.2.2 Some Advisable Limits
By using the empirical relationships described above, data that were known before the
accidents, we evaluated the following ranges of probabilities for two recent failures,
which we used as a further bench-marking element in the portfolio risk assessment
(Fig. 10.5):
• Mount Polley: Operations were going to add a buttress at the toe, a bulldozer and
people were working at the toe. Nothing was noted until short before the failure.
Estimated p f = 0.08 − 0.1.
• Fundao Dam: Plans to build a buttress were laid out. Leaks were noted before
failure. Estimated p f = 0.11 − 0.17.
Again, it is strongly recommended to comply with the “credibility threshold” (see
Sect. 7.3) and therefore not to state probabilities of failure lower than 10
−5 to 10
−6 .
Consider as well what very large FoS may mean and avoid blindly accepting
values of the FoS > 2, as they may be meaningless.
161
Table 11.4 (continued)
Category
Design
Build
Operate
D1
D2
D3
CO
OM
Investigation Testing
Analyses and
documentation
Construction Operations
and
monitoring
III Average
Evaluate performances of
nearby
structures
Index tests
on samples
from site
Rational
analyses
using
parameters
inferred from
index tests
Informal
construction
supervision
Annual
inspection by
qualified
engineer
Estimate
subsoil
profile from
existing data
and borings
No filed
measurements
Maintenance
limited to
emergency
repairs
IV Poor
No field
investigation
No
laboratory
tests on
samples
obtained at
the site
Approximate
analyses
using
assumed
parameters
No
construction
supervision
by qualified
engineer
Occasional
inspection
No
construction
control tests
No field
measurements
Adapted from SLM and (Altarejos et al. 2015)
11.2.2 Some Advisable Limits
By using the empirical relationships described above, data that were known before the
accidents, we evaluated the following ranges of probabilities for two recent failures,
which we used as a further bench-marking element in the portfolio risk assessment
(Fig. 10.5):
• Mount Polley: Operations were going to add a buttress at the toe, a bulldozer and
people were working at the toe. Nothing was noted until short before the failure.
Estimated p f = 0.08 − 0.1.
• Fundao Dam: Plans to build a buttress were laid out. Leaks were noted before
failure. Estimated p f = 0.11 − 0.17.
Again, it is strongly recommended to comply with the “credibility threshold” (see
Sect. 7.3) and therefore not to state probabilities of failure lower than 10
−5 to 10
−6 .
Consider as well what very large FoS may mean and avoid blindly accepting
values of the FoS > 2, as they may be meaningless.