4.2 A Systemic View on Tailings Dams Failure Processes
47
Table 4.4 Attributed causality scenario stemming from project inception
Attributed
causality scenario
ICOLD
World-wide 1910–2009 data
Recorded failures Failures in %
Recorded failures Failures in %
Poor investigation
9
13
12
13
Engineering error
and omission,
excessive audacity
44
63
32
35
Poor construction
17
24
48
52
Total
70
100
92
100
Results
One Dam, Various Levels of Mitigation
Figure 4.9 depicts the results of the analyses for the four levels of mitigation (A),
(B), (C) and (D) (Figs. 4.3, 4.6, 4.7, 4.8). In Fig. 4.9 the horizontal dotted/dashed
lines “1979” and “2000” depict the framing estimates thresholds we published in
(Oboni and Oboni 2013) for those decades. It is interesting to compare the results of
the various mitigative levels (A)–(D) to those thresholds.
Base Cases (A) and (B): Base Case and Peer Review (including sensible RBDM
procedures and risk assessment from project inception) have calculated probabilities
of failure in the vicinity of 10
−3 , i.e., the factual estimated rate we published in
(Oboni and Oboni 2013) for the decade around 1979.
Case (C): Base case and inspections (over the life of the structure paired with
sensible RBDM procedures and risk assessment from project inception) lead to a
value near the mid-point of the values for the decades 1979 and 2000, bordering with
Fig. 4.9 Results of the annual probability of failures derived from the model for mitigation level
A, B, C, D including attributed causalities (Tables 4.2, 4.3) derived from factual data (Figs. 4.4, 4.5)