NPP: Power Grid Mutual Safety Assessment - Nuclear Power Plant Instrumentation and Control Systems for Safety and Security - page 422

modes, systems composition and structure,
etc.);

Analysis of literature shows a lack of attention
given to the issues of development of approaches
to integration of different safety assessment meth-
ods Thus, the work (Leech et al., 2008) suggests
an idea to combine qualitative and quantitative
methods. The main premise is that qualitative
methods should prepare base data for quantita-
tive methods.
The work (Johnson et al., 2008) offers the idea
of “methodological triangulation” - an extended
model of methods integration. The integration
discussed allows receiving information as to
the extent of the results obtained using different
methods agree or disagree. A common limitation
in the known works is a lack of methods compat-
ibility analysis, analysis of integration techniques,
scaling of input and output parameters, choice of
results aggregation rules, etc.
Consequently, NPP safety assessment methods
integration must ensure both validity check for
results obtained and enhanced assessment validity
as a result of maximum coverage of the whole set
of input data by a minimum set of methods and
information technologies used.
Fuzzy technologies are actively used for NPP
safety assessment. Thus, for example, in nuclear
industry Fuzzy Logic and Intelligent Technologies
intensively are applied for solving fuzzy control
problems , which cannot be solved using existing
methods and approaches.
BBNs are also widely used in system safety
assessment tasks characterized by uncertainty,
imperfect knowledge, influence of a variety of
random factors. Thus, e.g., BBNs are used as a
basis for creation of the expert diagnostics system
for NPP operators (Kang et al., 1999), for modeling
complex industrial facilities (Weber et al., 2006),
as well for evaluation of reliability and safety as-
sessment in complex systems (Weber et al., 2001).

• Statistical (Historical) Data (D s ):

Accumulated by observation of NPP sys-
tems parameters throughout their life cy-
cle. Into this category reliability and safety
characteristics, operating environment
conditions, external systems it. Many pa-
rameters (quantitative attributes) describ-
ing NPP operation processes are in fact
random values (RV). The need to consider
a number of random factors leads to sto-
chastic uncertainty in safety assessment
and assurance;

• Linguistic Data (D L ): Represented as

natural language expressions, obtained
from professional experts in this ield. A
part of information about NPP behavior,
relationships of parameters of its operation
and environment may be represented in the
form of expert knowledge, which should
also be taken into account in NPP safety
assessment.

The problem of NPP safety assurance cannot
be solved within the scope of one disciplinary
approach. Use of any group of methods of risk
analysis due to, for example, the expert's prefer-
ences, leads to loss and/or disregard of a part of
input data describing NPP operation.
Consequently, in order to obtain a reliable
estimate of NPP safety it is expedient to use all
the above groups of input data (deterministic,
statistical, linguistic).
The problem of NPP safety assessment cannot
be solved within the scope of one disciplinary ap-
proach. Consequently, in order to obtain reliable
safety values it is reasonable to use all the above
groups of input data.