Page 28 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Basic concepts 117
As a common denominator, the monetary value of losses sions. This makes accurate failure predictions almost impossi-
is often used to quantify consequences. Such “monetizing” of ble. So, modem risk assessment methodologies provide a sur-
consequences-assigning dollar values to damages-is straight- rogate for such predictions. Assessment efforts by pipeline
forward for some damages. For others, such as loss of life and operating companies are normally not attempts to predict how
environmental impacts, it is more difficult to apply. Much has many failures will occur or where the next failure will occur.
been written on the topic of the value of human life, and this is Rather, efforts are designed to systematically and objectively
further discussed in absolute risk quantification (see Chapter capture everything that can be known about the pipeline and its
14). Placing a value on the consequences of an accident is a key environment, to put this information into a risk context, and
component in society’s determination of how much it is willing then to use it to make better decisions.
to spend to prevent that accident. This involves concepts of Risk assessments normally involve examining the factors or
acceptable risk and is discussed in Chapter 15. variables that combine to create the whole risk picture. A com-
The hazards that cause consequences and are created by the plete list of underlying risk factors-that is. those items that
loss of integrity of an operating pipeline will include some or add to or subtract from the amount of risk-can be identified
all ofthe following: for a pipeline system. Including all of these items in an assess-
ment, however, could create a somewhat unwieldy system and
Toxicityiasphyxiation threats from released products- one of questionable utility. Therefore, a list of critical risk indi-
contact toxicity or exclusion of air from confined spaces cators is usually selected based on their ability to provide use-
Contaminatiodpollution from released productsdamage ful risk signals without adding unnecessary complexities. Most
to flora, fauna, drinking waters, etc. common approaches advocate the use of a model to organize or
0 Mechanical effects from force of escaping product- enhance our understanding of the factors and their myriad pos-
erosion, washouts, projectiles, etc. sible interactions. A risk assessment therefore involves trade-
0 Firehgnition scenarios involving released products-pool offs between the number of factors considered and the ease of
fires, fireballs, jet fires, explosions use or cost of the assessment model. The important variables
are widely recognized, but the number to be considered in the
These hazards are fully discussed in following chapters, model (and the depth ofthat consideration) is a matter ofchoice
beginning with Chapter 7. for the model developers.
The concept of the signal-to-noise ratio is pertinent here. In
Risk assessment risk assessment, we are interested in measuring risk levels-the
risk is the signal we are trying to detect. We are measuring in a
Risk assessment is a measuring process and a risk model is a very “noisy” environment. in which random fluctuations and
measuring tool. Included in most quality and management high uncertainty tend to obscure the signal. The signal-to-noise
concepts is the need for measurement. It has been said that “If ratio concept tells us that the signal has to be of a certain
you don’t have a number, you don’t have a fact-you have an strength before we can reliably pick it out of the background
opinion.” While the notion of a “quantified opinion” adds noise. Perhaps only very large differences in risk will be
shades of gray to an absolute statement like this, most would detectable with our risk models. Smaller differences might be
agree that quantifying something is at least the beginning of indistinguishable from the background noise or uncertainty
establishing its factual nature. It is always possible to quantify in our measurements. We must recognize the limitations of
things we truly understand. When we find it difficult to our measuring tool so that we are not wasting time chasing
express something in numbers, it is usually because we don’t apparent signals that are, in fact, false-positives or false-
have a complete understanding of the concept. Risk assessment negatives. The statistical quality control processes acknowl-
must measure both the probability and consequences of all edge this and employ statistical control charts to determine
of the potential events that comprise the hazard. Using the risk which measurements are worth investigating further.
assessment. we can make decisions related to managing those Some variables will intuitively contribute more to the signal;
risks. that is, the risk level. Changes in variables such as population
Note that risk is not a static quantity. Along the length of a density, type of product, and pipe stress level will very obvi-
pipeline, conditions are usually changing. As they change, the ously change the possible consequences or failure probability.
risk is also changing in terms of what can go wrong, the likeli- Others. such as flow rate and depth of cover will also impact the
hood of something going wrong, andor the potential conse- risk, but perhaps not as dramatically. Still others, such as soil
quences. Because conditions also change with time, risk is not moisture, soil pH, and type of public education advertising, will
constant even at a fixed location. When we perform a risk eval- certainly have some effect, but the magnitude of that effect is
uation, we are actually taking a snapshot of the risk picture at a arguable. These latter are not arguable in the sense that they
moment in time. cannot contribute to a failure, because they certainly can in
There is no universally accepted method for measuring risk. some imaginable scenarios, but in the sense that they may be
The relative advantages and disadvantages of several more noise than signal, as far as a model can distinguish. That
approaches are discussed later in this chapter. It is important to is, their contributions to risk may be below the sensitivity
recognize what a risk assessment can and cannot do, regardless thresholds ofthe risk assessment.
of the methodology employed. The ability to predict pipeline
failures-when and where they will occur-would obviously Risk management
be a great advantage in reducing risk. Unfortunately, this can-
not be done at present. Pipeline accidents are relatively rare and Risk management is a reaction to perceived risks. It is practiced
often involve the simultaneous failure of several safety provi- everyday by every individual. In operating a motor vehicle,