Page 23 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

1/2 Risk: Theory and Application
diffuse and mix in irreversible processes, unmaintained build- methodologies and has made appropriate use of information
ings eventually crumble, and engines (highly ordered systems) and logic to create a model that can reliably produce such theo-
break down without the constant infusion of maintenance ries. It is hoped that the theory is a fair representation of actual
energy. risks. To be judged a superior theory by the scientific commu-
Here is another way of looking at the concept: “Mother nity, it will use all available information in the most rigorous
Nature hates things she didn’t create.” Forces of nature seek to fashion and be consistent with all available evidence. To be
disorder man’s creations until the creation is reduced to the judged a superior theory by most engineers, it will additionally
most basic components. Rust is an example-metal seeks to have a level of rigor and sophistication commensurate with its
disorder itself by reverting to its original mineral components. predictive capability; that is, the cost of the assessment and its
If we indulge ourselves with this line of reasoning, we may use will not exceed the benefits derived from its use. If the
soon conclude that pipeline failures will always occur unless an pipeline actually behaves as predicted, then everyone’s confi-
appropriate type of energy is applied. Transport of products in a dence in the theory will grow, although results consistent with
closed conduit, often under high pressure, is a highly ordered, the predictions will never “prove” the theory.
highly structured undertaking. If nature indeed seeks increas- Much has been written about the generation and use of theo-
ing disorder, forces are continuously at work to disrupt this ries and the scientific method. One useful explanation of the
structured process. According to this way of thinlang, a failed scientific method is that it is the process by which scientists
pipeline with all its product released into the atmosphere or into endeavor to construct a reliable and consistent representation of
the ground or equipment and components decaying and revert- the world. In many common definitions, the methodology
ing to their original premanufactured states represent the less involves hypothesis generation and testing of that hypothesis:
ordered, more natural state of things.
These quasi-scientific theories actually provide a useful way 1. Observe a phenomenon.
of looking at portions of our world. If we adopt a somewhat 2. Hypothesize an explanation for the phenomenon.
paranoid view of forces continuously acting to disrupt our 3. Predict some measurable consequence that your hypothesis
creations, we become more vigilant. We take actions to offset would have if it turned out to be true.
those forces. We inject energy into a system to counteract the 4. Test the predictions experimentally.
effects of entropy. In pipelines, this energy takes the forms of
maintenance, inspection, and patrolling; that is, protecting the Much has also been written about the fallacy of believing
pipeline from the forces seeking to tear it apart. that scientists use only a single method of discovery and that
After years of experience in the pipeline industry, experts some special type of knowledge is thereby generated by this
have established activities that are thought to directly offset special method. For example, the classic methodology shown
specific threats to the pipeline. Such activities include patrolling, above would not help much with investigation of the nature of
valve maintenance, corrosion control, and all of the other the cosmos. No single path to discovery exists in science, and
actions discussed in this text. Many of these activities have no one clear-cut description can be given that accounts for all
been mandated by governmental regulations, but usually only the ways in which scientific truth is pursued [56,88].
after their value has been established by industry practice. Common definitions of the scientific method note aspects
Where the activity has not proven to be effective in addressing a such as objectivity and acceptability of results from scientific
threat, it has eventually been changed or eliminated. This evalu- study. Objectivity indicates the attempt to observe things as they
ation process is ongoing. When new technology or techniques are, without altering observations to make them consistent with
emerge, they are incorporated into operations protocols. The some preconceived world view. From a risk perspective, we
pipeline activity list is therefore being continuously refined. want our models to be objective and unbiased (see the
A basic premise of this book is that a risk assessment discussion of bias later in this chapter). However, our data
methodology should follow these same lines of reasoning. All sources often cannot be taken at face value. Some interpretation
activities that influence, favorably or unfavorably, the pipeline and, hence, alteration is usually warranted, thereby introducing
should be considered-even if comprehensive, historical data some subjectivity. Acceptability is judged in terms ofthe degree
on the effectiveness of a particular activity are not yet available. to which observations and experimentations can be reproduced.
Industry experience and operator intuition can and should be Of course, the ideal risk model will be accurate, but accuracy
included in the risk assessment. may only be verified after many years. Reproducibility is
another characteristic that is sought and immediately verifiable.
The scientific method If multiple assessors examine the same situation, they should
come to similar conclusions if our model is acceptable.
This text advocates the use of simplifications to better under- The scientific method requires both inductive reasoning and
stand and manage the complex interactions of the many vari- deductive reasoning. Induction or inference is the process of
ables that make up pipeline risk. This approach may appear to drawing a conclusion about an object or event that has yet to be
some to be inconsistent with their notions about scientific observed or occur on the basis of previous observations of sim-
process. Therefore, it may be useful to briefly review some ilar objects or events. In both everyday reasoning and scientific
pertinent concepts related to science, engineering, and even reasoning regarding matters of fact, induction plays a central
philosophy. role. In an inductive inference, for example, we draw conclu-
The results of a good risk assessment are in fact the advance- sions about an entire group of things, or a population, on the
ment of a theory. The theory is a description of the expected basis of data about a sample of that group or population; or we
behavior, in risk terms, of a pipeline system over some future predict the occurrence of a future event on the basis of observa-
period of time. Ideally, the theory is formulated from a risk tions of similar past events; or we attribute a property to a
assessment technique that conforms with appropriate scientific nonobserved thing on the grounds that all observed things of