Environmental Engineering Reference
In-Depth Information
2. The role of fragility curves to evaluate the uncertainty in probability estimates
3. How to mathematically and numerically obtain conditional probabilities for slope sta-
bility failures related to factor of safety (FS) and fragility curves capturing natural and
epistemic uncertainties
4. How to use the estimated fragility curves to obtain annualized probabilities used in
the most common practices of risk analysis
5. Overall conclusions on applicability, transparency, robustness, and engineering foot-
print over the process
11.2 eStIMatIon oF ConDItIonal ProbabIlItY
aS a FunCtIon oF SaFetY FaCtor
Risk assessments complement, but do not replace, other engineering analyses such as stabil-
ity, flow, or deformation analyses. One of the main benefits of performing a risk assessment
results from the detailed examination of engineering fundamentals required to prepare
rational event trees. These frequently highlight the importance or irrelevance of mecha-
nisms that could have been overlooked or overemphasized, respectively. Since engineering
analyses shed light on which mechanisms strongly affect the safety of a facility, one could
reasonably conclude that engineering analyses should precede risk assessment. However, the
risk assessment process can help decide which aspects of performance, and thus which type
of engineering analyses, should receive the bulk of the effort. Thus, as is so often the case in
engineering, an iterative process provides the most practical alternative. In our practice, we
find that starting with deterministic analyses, or what many would term “traditional engi-
neering analyses,” provides a solid base for the risk assessment, particularly if event prob-
abilities come from expert elicitation. During the risk assessment, the need for additional
or more refined deterministic or probabilistic analyses could become evident. Ultimately,
the engineer needs to combine both traditional engineering analyses and risk assessment to
obtain a good understanding of the structure's expected performance and risk.
11.2.1 FS versus p(f) charts for slope instability and soil transport
For slope stability problems, a correct slope stability analysis provides a desirable starting
point for a risk assessment. The engineer should attempt to understand the likely values and
variability of all relevant information. For most slope stability problems, these include
• Geometry
• Stresses
• Pore pressures
• Strengths
If the determination of one of these key parameters is not possible, the engineer should
at least evaluate the likely impact of any uncertainty or knowledge gap in the results of the
stability evaluation. Pitfalls to avoid when performing stability analyses include
1. Not determining the field drainage conditions—drained, undrained, or partially drained.
2. Using an incorrect strength for the stress path in the field—Lambe and Silva (2003)
show examples of the influence of stress path on strength for stability analyses.
3. Confusing undrained conditions and total stresses—engineers can and do perform sta-
bility analyses for undrained conditions using strengths determined in terms of effec-
tive stresses. While this requires knowledge about the corresponding pore pressures,
Search WWH ::




Custom Search