Page 362 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Risk Criteria 151337
per year are acceptable would be a part of such criteria but is Qualitative criteria
controversial for obvious reasons.
As previously noted, the notion of acceptable risk is central Some practitioners have developed qualitative matrices to help
to risk management. Because acceptability is often linked to decision makers evaluate risks. An example is shown in Figure
numerical criteria, the use of risk values expressed in absolute 15.2. While such charts have their limitations, they do provide
terms is often required. However, more qualitative criteria are a framework from which decision makers can agree on termi-
also commonly used, as illustrated later. nology and assign relative risk levels. Therefore, they do pro-
Areas where absolute risk acceptability criteria are more vide a tool to remove at least some subjectivity from the process
commonly seen include of risk evaluation.
Sometimes risk levels are classified using qualitative terms.
Regulatory approvals and standards For example:
New designs deviating from accepted practices
Instances where conventional mitigation does not “appear to Negligible-where the occurrence of the event is very
be” adequate. improbable and the consequence minor. No further action is
required for this level of risk beyond regular reviews.
Risk criteria bridge the gap between numerical risk estimates Low-where the risks are considered manageable through
and decision criteria such as “insignificant risk” or “acceptable appropriate mitigations measures that are in place to keep the
risk,” which incorporates a value judgment. risk at this level.
Some countries and local governing agencies have estab- 0 Intermediate-where the risks are higher than desired and
lished numerical risk criteria while others have avoided them. actions are required to reduce the risk to low. negligible, or
As previously discussed an argument can be made that, even in ALARP
the absence of numerical criteria, acceptable risk levels are still High-risks that are considered intolerable and must be
established implicitly via regulations that dictate design, opera- reduced to intermediate or lower.
tions, and maintenance activities.
When a regulatory authority establishes risk criteria or estab- Additional qualitative terms include tolerable. intolerable or
lishes design and operations requirements, it is making a social tolerable ifALARF: and broadly acceptable. Sometimes terms
and political decision, which can be guided but not replaced by like these are also coupled with numeric risk criteria (see Figure
technical advice. It is impossible for risk criteria to represent 15. I) to define the boundaries between these regions. When
with precision what is or is not acceptable to the public. Such this is done, it seems reasonable to assume that little further
value judgments vary within and between societies, and alter attention is paid to the qualitative terms.
with time, recent accident experience, and changing societal The concept of “as low as reasonably practical” (ALARP) is
values. widely used throughout risk assessment and management.
Risk criteria can be established for risk as a whole or for its Safety regulators worldwide require hazardous industries to
components of probability and consequence. In some cases, it evaluate the risks associated with manufacturing plants or
is more appropriate to focus on a component. For example, if processes of those industries. Generally, the philosophy is that
there is no opportunity to change the consequence portion of the risks should be minimized wherever possible.
the risk, a probability-only criterion might be more useful. Another approach to qualitative criteria attempts to avoid
In addition to the complex socioeconomic considerations, criteria all together. That philosophy de-emphasizes risk crite-
establishing pipeline risk criteria is an exercise that may draw ria in favor of a “continuous improvement” approach. Under
from this philosophy, the risk manager is continuously evaluating
and ranking risks and working to improve risks according to
0 Pipeline risk assessments some predetermined strategy. There are no “passifail” crite-
Analyses of historical pipeline failure and consequence rates ria. No risk will ever be acceptable because attempts to
Comparisons of risks from other similar and/or common improve are ongoing. Lower risk portions will just get less
activities attention. This approach is appealing in many ways, particu-
0 Comparisons with existing criteria in other areas (different larly in that it avoids some “tough” decisions, and can indeed
countries, regions, etc.) be useful in budget setting and other company internal efforts.
0 Comparisons with existing criteria for other structures or However, it will not support certain decisions and will not
other industries. shield the practitioner from the fact that his actions can be
used to infer a risk tolerance expressed in absolute terms
Criteria can be established on the basis of human life safety, (see page 335).
potential environmental damages, economic considerations
(including the costs of failure), or other factors. Some factors ALARP principle
can be seen to dominate certain types ofpipelines. For example,
a natural gas pipeline is perhaps best judged on the basis of life The ALARP (as low as reasonably practical) principle is
safety, whereas a crude oil pipeline is perhaps better held to an derived from the U.K. Health and Safety at Work. .. Act of 1974,
environmental damages criterion. which requires “every employer to ensure, so far as is reason-
Because so many nontechnical issues are embedded in risk ably practicable, the health, safety and welfare of all his
criteria and no clear-cut guidance can be given from a purely employees.”This is interpreted as requiring employers to adopt
technical standpoint, it is useful to review some existing safety measures unless the cost is grossly disproportionate to
criteria. the riskreduction [91].