527 / 648
CDOIF
Chemical and Downstream Oil
Industries Forum
CDOIF is a collaborative venture formed to agree strategic areas for
joint industry / trade union / regulator action aimed at delivering
health, safety and environmental improvements with cross-sector
benefits.
These can then be compared to the tolerability criteria as follows:
Frequency per establishment per
receptor per year
(
mitigated
)
Frequency at which
CDOIF Consequence
Level
is equalled or
exceeded
10
Ͳ8
–10
Ͳ7
10
Ͳ7
–10
Ͳ6
10
Ͳ6
–10
Ͳ5
10
Ͳ5
–10
Ͳ4
10
Ͳ4
–10
Ͳ3
10
Ͳ3
–10
Ͳ2
>10
Ͳ2
D Ͳ MATTE
Intolerable
C Ͳ MATTE
TifALARP
B Ͳ MATTE
Broadly Acceptable
X
A Ͳ MATTE
X
Sub MATTE
Tolerability not
considered by CDOIF
The mitigated
risk
is depicted above by
X
.
It can now be seen that the mitigated risk is TifALARP. Further risk reduction needs to
be considered and implemented so far as is reasonably practicable (but an ALARP
demonstration may show the cost of further risk reduction is grossly disproportionate).
6.2.1.5 Interdependent scenarios
When summing frequencies it is important that this should only be done for independent
events.
For example, from the four tank example above (6.2.1.2), consider a further possible
level C scenario of a multi-tank fire arising from a spill followed by escalation. The
overall escalated scenario frequency would be made up from the chance of any of the
other events occurring (spills) and then escalating (ignition). The frequency of the
escalation scenario would need to be compared to the level C tolerability criteria.
However, when considering the frequencies for A and B tolerability (all events with
outcomes at or exceeding level A or B), the risk assessor would not in this case sum the
A and B spill frequencies with the escalated event (level C) frequency. This is because
the level C event is not independent from the level A and B initiating events. The
escalated scenario frequency is derived from the frequencies of the lesser events and
their probabilities of escalation (the spill frequency includes the frequency of both un-
ignited and ignited events). Summing the spill events and the escalated fire events
would result in double counting of the same initiating events.
Conversely, if the level C scenario was caused by an event independent to the level A
and B events (e.g. explosion from adjacent site) then the frequencies would be summed
when examining level A or B tolerability.
Consideration of bowtie diagrams often helps to avoid errors in logic.
Guideline – Environmental Risk Tolerability for COMAH Establishments v1.0
Page 41 of 88