Publications
Nuclear Systems Enhanced Performance (NSP) Program
Enhancing the Competitiveness of Nuclear Power Plants Through Risk Informed Decision Making
G.E. Apostolakis, M.W. Golay, J.P. Koser, A. Knight, G. Sato, and M.E. Silsdorf,
MIT-NSP-TR-016 (November 2002)
Abstract
Nuclear Power Plants are
faced with many new challenges in a deregulated energy market. To stay
competitive, expenses must be reduced while maintaining safety and stakeholder
confidence. To enhance competitiveness, the decision making process must
consider the impact of every decision on economics, safety, and stakeholders.
When considering plant
changes, the impact on both, the public and workers, must be considered. While
workers in other industries are generally only subjected to industrial accidents,
nuclear workers are also subjected to radiological hazards. Decisions that
involve the risk of an industrial accident only tend to be trivial; avoid
decisions that increase the risk of an industrial accident. However, when
decisions are made that involve nuclear safety, the problem becomes more
complex. The radiological hazard not only can affect the workers, but the
public as well.
Nuclear power plants also must consider the impact of decisions on the stakeholders. Stakeholders are external parties that have the potential to have a negative or positive impact on the continued operation of a nuclear power. Stakeholders include government and the public. Public opinion has the power to destroy even the safest, best-run nuclear plants. Even if a decision has no significant impact on the safety of the public and workers, the potential impact on the stakeholders must be considered.
It becomes clear that every decision made at nuclear power plants must simultaneously consider the impacts on economics, safety, and stakeholders. However, humans have difficulty making decisions that involve multiple objectives (attributes). Furthermore, different people put different emphasis (weights) on objectives. To further complicate the matter, the same person may vary the emphasis on the objectives with each problem being considered. Therefore, for decision makers to make consistent, reliable decisions, a structured methodology is needed.
A two-step methodology is presented to guide nuclear power plant staff in making better decisions. The goal of the methodology is to reduce the subjectivity in the decision making process as much as possible. All possible decision options are formulated and applied to the process. The first step eliminates those decision options that are clearly not acceptable. The second step rank-orders the remaining decision options using multi-attribute utility theory (MAUT). A numerical index called the Performance Index (PI) is assigned to each remaining decision option (this is the expected disutility of MAUT). The decision option with the lowest PI is considered the "best" option.
To demonstrate the usefulness of the decision making methodology, two case studies are presented. The first case study, presented in Part 1 of this report, involves the frequency of the Containment Integrated Leak Rate Test (ILRT). Three decision options are formulated representing different ILRT frequencies. MAUT is applied to the three decision options yielding performance indices. The decision maker can then compare the performance indices to maker a better decision.
Part 2 of this report
presents the second case study, which investigates Hydraulic Control Unit (HCU)
maintenance. Decision options are presented that involve both online and
offline maintenance strategies. As in the first case study, MAUT is applied to
the decision options.
The third part of this report provides an overview of current online maintenance practices in the United States. A discussion on the benefits of the inclusion of risk information is presented.
