Measuring Nuclear Power Plant Performance in the 1990s

With competition in electricity markets, here is concern that reliability
and safety at nuclear power plants (NPPs) will be compromised. Some argue
that safety shortcuts will be taken by NPP operators to reduce cost. But
without rate regulation (and the ability to pass through the costs of
repairs and replacement power to cover commitments during outages due
to unsafe practices), NPP operators are learning that both high safety
and low cost are required under competition. Without guaranteed revenues,
extended regulatory outages for safety problems or an outage due to an
accident could lead to a total loss of plant equity.

To make a profit in a competitive market, base-load nuclear power plants
must attempt to

(1) maximize the number of megawatt hours (MWh) generated (output) during
the year to cover fixed costs while

(2) minimizing the cost per MWh to cover variable costs.

This is done by maximizing the capacity factor and minimizing total
cost. In this white paper I’ll discuss the primary performance indicators
(output and cost) and how they are functionally related to several secondary
performance indicators. (In the next section I discuss the NRC’s safety
Performance Indicators.)

First, the annual capacity factor (CF) is output divided by the maximum
megawatts that could be generated during a year (annual capacity, e.g.,
Maximum Dependable Capacity, MDC). (So, output is equal to capacity times
the capacity factor.) The capacity factor is equal to the capacity utilization
rate (the capacity factor while the plant is operating) times the service
(or availability) factor (the percent of the time the plant is operating).
The service factor is equal to scheduled availability rate times the reliability
factor. To maximize output, the plant operator must: 1) increase capacity
utilization (CU) by minimizing production losses during operation (1 –
CU), i.e., by pushing the plant closer to its designed capacity limits;
2) increase the scheduled availability (SA) factor by minimizing the scheduled
outage rate (1 – SA), i.e., by reducing outage maintenance time; and 3)
increase the reliability factor (RF) by minimizing the Forced Outage Rate
(1 – RF).

To summarize, CF = CU x SA x RF, i.e., the capacity factor is a function
of several underlying performance indicators, including reliability.

During the last 10 years, nuclear power plants have increased the median
industry capacity factor from 68 percent in 1989 to 88 percent in 1998
(because several plants were experiencing unusual outages in 1997, the
data in that year is skewed). See Table 1.

Table 1 – Increase in median industry performance

This was done by:

• Reducing the median production losses during operation (1 – CU) from
  5.9 percent to 0.7 percent

• Reducing the median scheduled outage rate (1 – SA) from 22.5 percent
  to 10.3 percent

• Reducing the median forced outage rate (1 – RF) from 1.5 percent
  to 0.5 percent

As can be seen in Figure 1, if the reliability factor (RF) declined from
the previous year, the change in capacity utilization (CU) decreased until
reliability improved, creating a counter-cyclical relationship. With reliability
positively related to underlying safety, the nuclear power industry was
able to improve productivity while carefully monitoring reliability, hence
safety.

Figure 1 – Capability utilization vs. reliability

Second, under competition the plant operator must minimize cost/MWh.
Here, cost/MWh is total operating cost divided by output. Operating cost
is composed of three components: 1) Operations and Maintenance expenses;
2) fuel expenses, and 3) capital additions (O&M costs that must be depreciated).

To analyze real changes in cost during the last decade, we can decompose
changes in cost/MWh into decreases in cost and increases in the capacity
factor. This can be done by considering changes in the “minimum” cost/MWh
(this is cost/MWh at a 100 percent capacity factor) and changes in the
capacity factor.

From 1989 to 1998, cost/MWh declined 4 percent per year on average. (See
Table 2 and Rothwell, 2000.) However, minimum costs declined only 2 percent
per year, while the capacity factor improved 2 percent per year. Further,
while fuel expenses (25 percent of cost) declined 3 percent per year and
capacity additions (12 percent of cost) declined 4 percent per year, O&M
(63 percent of cost) declined only 1 percent per year. Therefore, with
capacity utilization and reliability approaching 100 percent and fuel
expenses and capacity additions approaching lower limits, further declines
in cost/MWh must come from decreases in scheduled outage time (measured
by scheduled availability) and decreases in O&M expenses without compromising
safety. This can be done in several ways, including:

1) benchmarking industry performance to determine possibilities (see
Maidment and Rothwell, 1998);

2) cross-training employees (see the Limerick case study in IAEA 1999);
and

3) by using “Risk-Informed Management” (RIM). Before discussing RIM systems,
I review changes in NRC regulation.

Table 2 – Decline in MWh usage

Risk-Informed and Performance-Based Regulation

The US NRC is now exploring “Risk-Informed and Performance-Based Regulation.”
According to NRC Commissioner Diaz, “Risk-informed regulation is a set
of deterministic criteria, operating experience, defense-in-depth, engineering
judgments, and probabilistic risk assessments” (PRA) conducive to safety-focused
decision-making. (See Diaz, 1999, p. 5.) Performance-Based Regulation
involves:

1) defining several safety-related “Performance Indicators” (these are
mathematical representations of various aspects of nuclear plant performance
based on “objective data that have been shown to accurately measure performance
or provide a positive correlation to performance in specific areas” (see
www.nrc.gov/NRR/OVERSIGHT /OVERVIEW/faq.html); and

2) determining whether plants are operating within “thresholds of acceptable
risk” for each safety Performance Indicator.

“Performance-Based Regulation” began with the Performance Indicator program
in 1986. It was expanded and refined for a dozen years. In 1999 the NRC
tested a new Oversight Process based on its review of its Performance
Indicators, regular (baseline) inspections, and inspections beyond the
baseline at plants with performance below established thresholds. Oversight
focuses on seven areas: Initiating Events, Mitigating Systems, Barrier
Integrity, Emergency Preparedness, Occupational Radiation Safety, Public
Radiation Safety, and Physical Protection (security). Each area is measured
with one or more Performance Indicators. (See www. nrc.gov/OPA/ primer.htm
for a list of the indicators.)

“Risk-Informed Regulation” has its roots in the first US NRC (1975) PRA,
WASH-1400, also known as the “Rasmussen Report.” This was the NRC’s first
systematic attempt to estimate accident sequences, probabilities, and
consequences for Pressurized and Boiling Water Reactors (PWRs and BWRs).
After the accident at Three Mile Island, the NRC ordered utilities with
plants in highly populated areas to conduct plant-specific PRAs. The NRC
updated its PRA methodology with NUREG-1150 (US NRC, 1991) looking at
PWRs with three types of containment and BWRs with two types of containment.
While US NRC (1991) was being reviewed and redrafted, the NRC issued a
Generic Letter (US NRC 1988) requesting nuclear power plant operators
(licensees) to perform a plant-specific search for vulnerabilities to
severe internally-initiated accidents under the Individual Plant Examination
(IPE) program. Most licensees used the PRA methodology in performing their
IPE.

During the 1990s, the NRC began to rely more and more heavily on the
PRA approach, for example, in setting priorities for NRC staff inspection
activities. In July 1998, the NRC issued Regulatory Guide 1.174 (US NRC
1998) and has since been modifying its basic nuclear power plant regulations
contained in Volume 10 of the Code of Federal Regulations Part 50 (10
CFR 50), “Domestic Licensing of Production and Utilization Facilities.”

This revision of regulations has not been without controversy. For example,
on the one hand, the Advisory Committee on Reactor Safeguards (ACRS) in
a letter to NRC Chair Richard Meserve (February 14, 2000) on the subject
of “Impediments to the Increased Use of Risk-Informed Regulation” stated,
“We consider the more significant of the technical impediments to be:

• PRA inadequacies and incompleteness in some areas

• The need to revisit risk-acceptance criteria

• Lack of guidance on how to implement defense in depth and on how
  to impose sufficiency limits

• Lack of guidance on the significance and appropriate use of importance
  measures

• Variation of PRA quality and scope and the need for Standards

NRC has requested the American Society of Mechanical Engineers and the
American Nuclear Society to develop Standards to ensure that the technical
quality of PRAs is sufficient to support the regulatory review and approval
of licensee risk-informed applications. We believe that development of
appropriate PRA Standards is important to risk-informing the regulations.”

On the other hand, some in the nuclear power industry believe that licensees
should not be required to have a PRA that conforms to a specific Standard
as a condition for using PRA insights. In contrast, at least one representative
(see Riccio, 2000) of the “anti-nuclear power community” believes, “The
nuclear industry is not operating better. The NRC is just regulating less.
That is why Public Citizen has concluded that ‘risk-informed’ regulation
means that the public is exposed to more risk while the nuclear industry
is exposed to less regulation.”

How should nuclear power plant managers respond to these changes in NRC
regulation?

Risk-Informed Management

Nuclear plant managers must decide whether the benefits outweigh the
costs of acquiring the data and building models necessary to make risk-informed
decisions. As one example in determining these economics, South Texas
NPP managers have categorized plant components with their PRA as high,
medium, low, and non-risk significant. They are requesting the NRC to
change regulations that require special maintenance for components determined
to be of low or non-risk significance. The South Texas Project, as a pilot
plant in the NRC’s test of “Risk-Informed and Performance-Based Regulation,”
is requesting the NRC (through exemption requests) to change regulations
requiring rigorous testing, maintenance, surveillance, and qualification
procedures on those safety-related and non-safety-related components in
the low-risk and no-risk significant categories (see Stellfox, 1999).
If these exemptions are approved, South Texas O&M costs could decline
as much as $1 million annually. Further, even without approval, South
Texas estimates another $1 million annual savings on discretionary program
and procedure changes. If annual savings are greater than the annual cost
of acquiring data and building models, South Texas increases its value
by adopting this form of risk-informed management (RIM).

If the South Texas approach proves successful, it is likely that benefits
will exceed costs in implementing RIM systems at other plants. However,
regardless of whether PRA standards are required or ever approved by the
NRC, the industry should take the initiative to extend the PRA methodology
to other optimization goals. PRA focuses on accident sequences, probabilities,
and consequences. The PRA can be embedded within probability models that
focus on business decision sequences, probabilities, and consequences,
including profitability. These models would identify maximum profit strategies
under safety constraints, such as release standards (this is a constrained
optimization problem that leads to the calculation of “shadow” prices
for changes in the safety constraints). The success of these business
decisions must be measured with appropriate business-oriented Performance
Indicators, such as cost/MWh and net revenue/MWh, as discussed above,
not simply those used by the NRC. These performance indicators should
be functionally related to avoid redundancy and provide insight into key
underlying relationships.

“Profitability Risk Assessment” models could be used to analyze the following
questions facing nuclear power decision makers in competitive electricity
market (these are only a few):

• What is the maximum capacity of the plant within acceptable safety
  thresholds given available upgrade options?

• What is the optimal nuclear fuel burn up without compromising safety
  or the fuel cycle length without decreasing reliability?

• What is the optimal level of generation above the maximum dependable
  capacity (not “tech specs”) for very short periods with high market
  prices, e.g., above $100/MWh?

• What is the optimal allocation of capacity between electricity generation
  and ancillary services provision, for example, to provide voltage
  support, and how quickly can be plant ramp up to full capacity safely?

• What risk premium should be charged on firm power in a bilateral
  contract?

• What is the optimal sequence of upgrades to extend the life of the
  plant?

Conclusion

Risk-Informed Management models will enable nuclear power plant owners
to make profit-maximizing decisions while maintaining reliability and
safety. While the industry has been able to increase output and lower
cost/MWh, not all plants have been equally successful. While poorly performing
plants must increase productivity and safety, Risk-Informed Management
systems will enable all nuclear plant owners and operators to apply Information
Technology and real-time plant data to business decisions as these decisions
become more complex with the introduction of competition into electricity
markets.

References Diaz, N.J. (1999). “Benefits of Safety-Focused Regulation,”
presented at the 1999 ANS Winter Meeting, Long Beach, California.

www.nrc.gov/OPA/nrarcv/s99-38.html

International Atomic Energy Agency (1999). Evaluating and Improving Nuclear
Power Plant Operating Performance. (G.S. Rothwell, editor)IAEA-TECDOC-1098.

Maidment, J. and G.S. Rothwell (1998). “All Nuclear Power Plants Are
Not Created Equal: Preparing for Competition and Deregulation,” Public
Utilities Fortnightly (April 1).

Riccio, J.P. (2000). “The Deregulation of Nuclear Safety Standards Otherwise
Known As ‘Risk-Informing’: The Technical Requirements of 10 CFR Part 50,”
Statement of James P. Riccio, Public Citizen’s Critical Mass Energy Project
before the US NRC (June 20). www. citizen.org/CMEP/nuclearsafety/Deregnukesafety.htm

Rothwell, G.S. (2000). “Nuclear Power Plant Operating Costs: What’s Next,”
presented at the American Nuclear Society’s 2000 Annual Meeting, San Diego
(June 4-8, 2000).

Stellfox, D. (1999). “STP (South Texas Project) Finds Ranking Risk Components
Pays, Even Before NRC Changes,” Inside NRC 21 (26): 3-4 (December 20).

U.S. Nuclear Regulatory Commission (1998). An Approach for Using Probabilistic
Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to
the Licensing Basis (Regulatory Guide 1.174). www. nrc.gov/NRC/RG/01/01-174.html

U.S. Nuclear Regulatory Commission (1991). Severe Accident Risks: An
Assessment for Five U.S. Nuclear Power Plants, Final Report (NUREG-1150).

U.S. Nuclear Regulatory Commission (1988). Individual Plant Examination
for Severe Accident Vulnerabilities (Generic Letter No. 88-20, 10 CFR
50.54). www .nrc.gov/NRC /GENACT/GC/GL/1988/gl88020.txt

U.S. Nuclear Regulatory Commission (1975). Reactor Safety Study (WASH-1400,
NUREG-75/014).