Civil Engineering Reference
In-Depth Information
Qualitative
Likelihood classed as 'unlikely', 'likely', etc.
■
Consequences classed as 'minor', 'severe', etc.
■
Risk determined from a qualitative combination of likelihood and consequence, sometimes with a risk matrix (e.g. Figure 12.9)
■
Semi-quantitative
Likelihood prescribed values on say a fi ve-point scale based on description of how likely it is that the hazard will occur (e.g. once per year, once during
■
the lifetime of the building)
Consequence prescribed values on a similar scale, often based on descriptions of the potential fatalities/injuries (e.g. few minor injuries, several minor
■
injuries/few major injuries, one fatality, several fatalities), fi nancial loss, extent of collapse or some other metric
Risk determined from a risk matrix combining the likelihood and consequence (e.g. Figure 12.9)
■
Quantifi ed risk assessment
Probability of the hazard occurring calculated and expressed quantitatively (e.g. 1E-4/yr)
■
Consequences calculated, usually based on fatalities and injuries expressed as fi nancial 'cost'
■
Individual and population risk calculated for each hazard and aggregated across all hazards
■
Table 12.3
Types of risk assessment
Common to all forms of risk assessment is the concept of a
tolerable threshold, beyond which mitigation must be applied
to bring the residual risk back within tolerable bounds. The bold
line in
Figure 12.9
shows an example of this threshold, but it
must be noted that no one likelihood and consequence scale
will be universally applicable, and similarly the risk appetite
must be discussed and agreed with the client and building con-
trol authority at the outset. The example in
Figure 12.9
shows
a relatively risk-averse scale skewed against hazards resulting
in major consequences, refl ecting an aversion to such hazards
which is often observed in society when such hazards do materi-
alise. This aversion is observed despite the risk associated with
the hazard being lower than for some more frequent hazards
when assessed using the conventional type of risk matrix dis-
cussed in Chapter 3:
Managing risk in structural engineering
-
in effect, the perception of risk becomes one of conditional
probability: should the hazard materialise, it will automatically
be perceived as disproportionate. As such, the societal percep-
tion of risk is perhaps irrational: we do not expect buildings
to collapse, and when they do it is rarely viewed favourably
by society. The structural engineer must therefore be careful
to avoid the 'one size fi ts all' risk matrix and design a matrix
which is suited to the particular circumstances of the project
and incorporates the client's and building control authority's
attitude to risk. It is the responsibility of the structural engin-
eer to ensure that the client's decision on tolerability of risk is
an informed one, based both on the engineering consequences
of a particular hazard and the legal, societal and other implica-
tions of the client's decisions as the risk owner.
Fully quantifi ed risk assessments are usually only used in
particular circumstances, typically in the nuclear, petrochem-
ical and other low likelihood/high consequence industries. In
a semi-quantitative risk assessment that would be typical for a
Class 3 risk assessment, the values ascribed to the likelihood
and consequence of the hazard would be determined by the
designer in a systematic fashion. (See
Tables 12.4
and
12.5
for
an example of each.)
Once the likelihood and consequence are defi ned, the risk
can be determined from the risk matrix. Any risk assessment
process requires the threshold between tolerable and intoler-
able risks to be identifi ed: this is represented as an example by
the dark line in
Figure 12.9
but needs to be agreed with the cli-
ent and control authority at the outset of the risk assessment.
The principle that applies to mitigation of the risks identifi ed
is that intolerable risks must be mitigated, but more widely
all
risks should be mitigated so far as reasonably practicable. The
ERIC
(Eliminate, Reduce, Isolate, Control) hierarchy defi ning
an approach to the reduction of risk (Institution of Structural
Engineers, 2012) is particularly suitable for engineering
Consequence
Frequent
Common
Likely
Unlikely
Rare
Improbable
Figure 12.9 Example risk matrix adapted from Harding
et al
., 2009.
Courtesy of
The Structural Engineer
