48
4 Historic Failures “Statistics”
the higher estimate from the most recent data (Oboni and Oboni 2016b; Bowker and
Chambers 2015), for the Serious and Very Serious failures.
Finally, Case (D) with peer review and inspections reaches the lower bound of
the interval, i.e., the value we published for the decade of the year 2000, and very
similar to the lower estimate from the most recent data (Oboni and Oboni 2016b;
Bowker and Chambers 2015) for the Very Serious failures.
Absent or botched mitigations M1 and/or M2 can of course increase the value of
the probability of failure to historic high (decade around 1979) due to CCF.
To sum up, based on this example, the biggest reduction in the probability of
failure of the e-IDC chain is obtained through thorough inspections with sensible
RBDM procedures and risk assessment from project inception. Peer review has also
a beneficial effect, of course, but smaller, probably because the most deviances and
shortcuts intervene during the long term service life rather than during design. All
together the implementation of both mitigation reduces the e-IDC chain probability
of failure by almost one order of magnitude.
Prioritizing Risks in a Portfolio of Dams
Notwithstanding the assumptions made, which could be perfected in a real-life
portfolio study, the model is capable of reconstructing first estimates (a priori) of the
probability of failure in good agreement with the last one hundred years of tailing
dams failure history by looking at data (records) that should still be available for
many structures, possibly in corporate, governmental or regulators’ offices archives.
Thus, based on an examination of those records it is possible to determine, dam by
dam, the first estimate of the probability of failure which, paired with the potential
consequences each dam failure (to be determined using a multi-dimensional consequence analysis, see Chap. 12), will give the total risk and finally a dam portfolio
(corporate, national, regional) quantitative risk prioritization. That quantitative risk
prioritization would be the first step of what the Auditor General for the Province of
British Columbia recommends:
1.10 Risk-based approach. We recommend that government develop a risk-based approach
to compliance verification activities, where frequency of inspections are based on risks, such
as industry’s non-compliance record, industry’s financial state, and industry’s activities (e.g.,
expansion), as well as risks related to seasonal variations. (Bellringer 2016)
We have already demonstrated (Oboni and Oboni 2012) how to include societal
and corporate tolerance (See Chap. 13) into the risk prioritization techniques and
shown how decision makers’ focus can be enhanced rather than obfuscated by unclear
risk assessments (Oboni and Oboni 2016a).
Given the public and corporate capital investments required to reduce future risks
generated by dams, it is paramount to first be able to address the situations of greater
risk that lie above corporate and social tolerance. In order to be able to use more
efficient prioritization it will be necessary to define multi-dimensional tolerance
levels, an exercise that we have already performed from local to country-wide scale
but cannot be discussed here due to limits of space.
4 Historic Failures “Statistics”
the higher estimate from the most recent data (Oboni and Oboni 2016b; Bowker and
Chambers 2015), for the Serious and Very Serious failures.
Finally, Case (D) with peer review and inspections reaches the lower bound of
the interval, i.e., the value we published for the decade of the year 2000, and very
similar to the lower estimate from the most recent data (Oboni and Oboni 2016b;
Bowker and Chambers 2015) for the Very Serious failures.
Absent or botched mitigations M1 and/or M2 can of course increase the value of
the probability of failure to historic high (decade around 1979) due to CCF.
To sum up, based on this example, the biggest reduction in the probability of
failure of the e-IDC chain is obtained through thorough inspections with sensible
RBDM procedures and risk assessment from project inception. Peer review has also
a beneficial effect, of course, but smaller, probably because the most deviances and
shortcuts intervene during the long term service life rather than during design. All
together the implementation of both mitigation reduces the e-IDC chain probability
of failure by almost one order of magnitude.
Prioritizing Risks in a Portfolio of Dams
Notwithstanding the assumptions made, which could be perfected in a real-life
portfolio study, the model is capable of reconstructing first estimates (a priori) of the
probability of failure in good agreement with the last one hundred years of tailing
dams failure history by looking at data (records) that should still be available for
many structures, possibly in corporate, governmental or regulators’ offices archives.
Thus, based on an examination of those records it is possible to determine, dam by
dam, the first estimate of the probability of failure which, paired with the potential
consequences each dam failure (to be determined using a multi-dimensional consequence analysis, see Chap. 12), will give the total risk and finally a dam portfolio
(corporate, national, regional) quantitative risk prioritization. That quantitative risk
prioritization would be the first step of what the Auditor General for the Province of
British Columbia recommends:
1.10 Risk-based approach. We recommend that government develop a risk-based approach
to compliance verification activities, where frequency of inspections are based on risks, such
as industry’s non-compliance record, industry’s financial state, and industry’s activities (e.g.,
expansion), as well as risks related to seasonal variations. (Bellringer 2016)
We have already demonstrated (Oboni and Oboni 2012) how to include societal
and corporate tolerance (See Chap. 13) into the risk prioritization techniques and
shown how decision makers’ focus can be enhanced rather than obfuscated by unclear
risk assessments (Oboni and Oboni 2016a).
Given the public and corporate capital investments required to reduce future risks
generated by dams, it is paramount to first be able to address the situations of greater
risk that lie above corporate and social tolerance. In order to be able to use more
efficient prioritization it will be necessary to define multi-dimensional tolerance
levels, an exercise that we have already performed from local to country-wide scale
but cannot be discussed here due to limits of space.