Page 450 - Pipeline Pigging Technology

Risk assessment and inspection for integrity

Fig. 5. Schematic of economic risk assessment.
While the estimates of the outage probabilities required for the life-loss
analysis are considered to have acceptable accuracy, more work is needed to
improve our confidence in methods used to estimate the probabilities of
associated consequences resulting from outage involving fire. The approach
that shows some promise in this regard focuses on developing risk spectra for
specific locations in which a pipeline facility passes close to a centre of
population.
Fig.4 illustrates the results that would be obtained for a specific site where
the probability of an event causing N or more fatalities is a function of the
number of fatalities N. This particular form of presentation appears to be
useful, in that it considers the full range of potential consequences. While it
is recognized that the concept of risk spectra has been used in other
industries[13,14], as well as in the gas-transmission industry[15] in Europe,
some fundamental questions will have to be addressed before this concept
finds a broader acceptance. Namely, what is the acceptable level of risk, what
assumptions should go into the estimates, since the results are sensitive to the
input, and how do we evaluate the benefits of remedial action. We need to be
able to decide, for example, whether it is worth spending $10million to
reduce outage probability by a factor of 2.

Economic risk

Economic risks have had a strong influence on our pipeline integrity
programme because the safety risks are estimated to be very low, and are
consistent with the excellent safety record. The fault tree used to estimate
economic risk is illustrated schematically in Fig.5, which shows that:

431