<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Analysis of the actions of NPP personnel in making decisions</article-title>
      </title-group>
      <contrib-group>
        <aff id="aff0">
          <label>0</label>
          <institution>ANO «Russian New University»</institution>
          ,
          <addr-line>Moscow</addr-line>
          ,
          <country country="RU">Russia</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>ANO «Scientific and Research Center for Information in Physics and Technique»</institution>
          ,
          <addr-line>Nizhny Novgorod</addr-line>
          ,
          <country country="RU">Russia</country>
        </aff>
        <aff id="aff2">
          <label>2</label>
          <institution>M.A. Berberova</institution>
        </aff>
        <aff id="aff3">
          <label>3</label>
          <institution>MIREA - Russian Technological University</institution>
          ,
          <addr-line>Moscow</addr-line>
          ,
          <country country="RU">Russia</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>The operational experience of nuclear facilities shows that personnel reliability has a significant impact on safety. We consider the reliability of the staff as the property (ability) of the staff to accurately and timely perform the necessary actions prescribed by the operating instructions, both during normal operation and in an accident. Personnel Reliability Analysis (PRA) is one of the significant tasks of probabilistic safety analysis. There are problem situations at NPP: the analysis methods and techniques used, the variety of solving tasks, the wide range and diverse nature of errors that a human operator can potentially make, the numerous factors affecting the probability of a particular error. All this makes it necessary to develop an automated personnel reliability analysis system that can support a specialist performing probabilistic safety analysis (PSA) in terms of performing the reliability analysis of the human operator, to ensure the required accuracy, quality and completeness of the results of the personnel reliability analysis.</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>The current practice of performing personnel
reliability analysis during PSA involves the use of a
number of methods of qualitative and quantitative PRAs,
as part of a structured procedure for their application.
PRA contains numerous levels and stages. In particular,
the formation of a list of personnel actions, a selection
analysis and a detailed analysis of the most significant
actions of the personnel, the determination of the
probabilities of an error in the performance of actions,
the analysis of the relationships between the actions of
the personnel, the analysis of the possibilities to correct
the mistake made, and the assessment of the uncertainty
of the personnel reliability indicators. The reliability of
the human operator depends on a combination of factors,
and the most important for analysis used to be the
available time, stress level, and staff qualification level.</p>
      <p>Currently, the application of these methods to
perform various stages of the PRA requires manual
processing by the executor of large amounts of
information presented in the form of tables and graphs,
which increases the analysis timeline and reduces the
quality of its implementation.
2.</p>
    </sec>
    <sec id="sec-2">
      <title>Decision making by the NPP operator</title>
      <p>The operator's activity consists of a set of sequential
actions to develop a solution and implement
management.</p>
      <p>
        The development of a solution is a process that
consists of three stages [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]:
1. Information search (perception of information)
(table 1).
2. Information preparation of decision making (table
2).
3. Decision making (table 3).
      </p>
      <p>Actions</p>
      <p>Signal Detection
Comparison with the standard</p>
      <p>Identification</p>
      <p>The inclusion of psychological processes
Attention, signal and background perception</p>
      <p>Visual thinking, random access memory
Operational thinking, long-term memory</p>
      <p>Actions
Information processing about the problematic</p>
      <p>situation
Search for essential information. Highlighting
from background information. Situation</p>
      <p>classification
Compilation of information. Building an FCM
situation.</p>
      <p>The inclusion of psychological processes
Long-term and random access memory. Conceptual and image thinking
Operational thinking. Theoretical thinking
Long-term memory. Theoretical thinking</p>
      <p>At the first stage, it is very important to respond to
alarms as quickly as possible and to find the source of the
accident.</p>
      <p>Table 3 demonstrates the next stage.</p>
      <p>If at the first stage it is important to notice the
accident, then at the second stage operator will have to
evaluate the current situation. At this stage, it is
extremely important to analyze the situation quickly and
correctly.</p>
      <p>Table 4 demonstrates the next stage.</p>
      <p>Table 4. Areas of error probabilities considering the types of
behavior
Operator Behavior Types Error Probability Areas
1. Based on experience Pr = 10-4 - 10-2
2. Based on rules (instructions) Pr = 10-3 – 0.1
3. Knowledge based Pr = 10-2 - 1</p>
      <p>Before performing actions aimed at eliminating the
emergency, the operator compares various solutions to
the current situation (from possible) and selects the
optimal.</p>
    </sec>
    <sec id="sec-3">
      <title>Types of staff behavior</title>
      <p>
        For rapid assessment of personnel reliability, used
primarily for screening analysis (screening) to determine
dominant events, error probability domains they apply
considering the types of behavior (cognitive modeling)
required for each operation [
        <xref ref-type="bibr" rid="ref2 ref3">2, 3</xref>
        ]:
1. Skill-based actions - Automated actions. They are
performed strictly according to the scenarios that are
set out in the instructions (similar, for example,
driving a car along a familiar route). With this type
of action, errors are usually associated with a
combination of various forced or temporary factors.;
2. Rule Based Actions - Actions taken in familiar
situations. Based on previously worked out
sequences of actions that are detailed in the
instructions and / or obtained from experience or
training. Actions by rules are less developed than
actions based on skill, because they are performed
less frequently and are more complex and often
require increased conscious control. Errors in this
type of action are associated with incorrect diagnosis
of the situation, which leads to the use of the wrong
rule or instruction;
3. Knowledge-based actions - Actions performed in
emergency situations when existing rules and
experience cannot be applied directly. There are no
ready-made instructions for these actions, they are
based on general knowledge, and they must be
executed in real time. Errors in this type of actions
are the result of the limited time to make decisions
and perform actions, as well as insufficient or
incorrect knowledge.
      </p>
      <p>
        Table 4 presents probability domains for the
described types of behavior [
        <xref ref-type="bibr" rid="ref2 ref3">2, 3</xref>
        ].
      </p>
      <p>
        Preliminary assessments of personnel post-emergency
actions (in response-response to an initial event) used in
PSA are expert screening [
        <xref ref-type="bibr" rid="ref2 ref3 ref4 ref5">2-5</xref>
        ].
      </p>
      <p>A logical scheme to help in deciding what type of
mental activity a person’s actions are shown in Fig. 1.</p>
    </sec>
    <sec id="sec-4">
      <title>Types (mechanism) and forms of personnel errors</title>
      <p>Personnel errors modeled in the PSA are usually
divided into the following types related to different types
of behavior:
− oversight / omission - an action leading to an
unintended result due to an error during performing.
This kind of error usually refers to skill-based
actions;
− delusion is an intentionally performed action, but the
intention is incorrect. This is a typical mistake in
assessing a situation or in planning a response,
leading to the execution of incorrect actions. This
type of error refers to actions based on rules or
knowledges.</p>
      <p>Personnel errors modeled in PSA are subdivided into
such forms as non-fulfillment errors and execution
errors:
− Failure to comply: personnel error consisting in
failure to perform one or more prescribed actions.
Typically, a failure error is a failure by the operator
to take the required security system into operation;
− execution error: personnel error consisting in
incorrectly performing one or more actions, or
performing any other action / actions instead of
prescribed ones. Typically, a runtime error is either
an inappropriate termination of a required security
function, or an inappropriate activation of a system
or initiation of a security function.</p>
      <p>The relationship between the actions of the staff takes
place if the result of a specific action depends on the
previous action or actions. The probability of personnel
errors in dependent activities should be higher than if
they were considered independent. All alarm sequences,
which include several personnel errors, are analyzed for
their potential dependencies. Whenever possible,
completely dependent personnel errors are redefined and
modeled as a single event.</p>
    </sec>
    <sec id="sec-5">
      <title>5. The effect of stress on decision making</title>
      <p>The main factors that affect the behavior of personnel
and the reliability of their actions are considered in a
detailed analysis of personnel errors:
− instructions (the following aspects of instructions are
evaluated, which are considered to be factors that
influence behavior: simplicity / complexity of
operations performed, number of steps, availability
of a written instruction, its format (descriptive or
symptom-oriented), quality of presentation, degree
of detail, etc.);
− training / experience;
− human-machine interaction;
− interaction / size (composition) of the group of
operators;
− information exchange;
− load;
− stress.</p>
      <p>Not all of these factors are directly applicable to each
of the analyzed personnel errors. Additional factors not
included in this list may be considered.</p>
      <p>Depending on the likelihood of the consequences of
the accident and the time available to the operator for the
intervention, personnel are exposed to different levels of
stress. This directly affects the likelihood of correct or
erroneous actions by personnel to troubleshoot in the
event of an emergency. Stress can occur for a number of
reasons: unpredictability of an accident, workload (its
level and duration), high noise level (for example, when
an alarm has triggered), conflicts during work.</p>
      <p>Stress is not necessarily accompanied by negative
reactions. The optimal level of stress will allow for better
concentration, increased attention, which will favorably
affect the work of the operator. At the same time,
“unfavorable” (high) stress will negatively affect the
work of even very experienced operators. The time limit,
the difficulty of eliminating the consequences of an
accident, errors, in the case of incorrect diagnosis of an
accident, will only increase the level of staff stress,
which will entail an increase in the likelihood of
erroneous actions.</p>
      <p>
        Fig. 2 shows a graph of the dependence of the
operator's performance on the level of stress experienced
by him [
        <xref ref-type="bibr" rid="ref6 ref7 ref8">6-8</xref>
        ].
      </p>
      <p>
        The following types of motivation for activity are
distinguished [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]:
      </p>
      <p>Cognitive motivation - a manifestation of interest in
the content of the activity. The process of work attracts a
person, the result is important for him of every task.</p>
      <p>Conflict avoidance is the pursuit of a conflict-free
style of behavior. They distinguish motivation for
avoiding conflicts with the social environment and with
their own principles.</p>
      <p>The prestige motivation is the realization that the
activity carried out is respected and gives weight to the
individual. It is expressed in desire to assert oneself.</p>
      <p>Utilitarian motivation - the possibility of obtaining
good earnings or other material wealth.</p>
      <p>Human activity is mobilized in real activity in the
form of a system of motives leading to actions.</p>
      <p>Motives are formed and rebuilt in the process of
professional development of the employee and in his
daily life.</p>
      <p>There are some differenties between situational
components of activity that is external, objective and
internal, and mental.</p>
      <p>External, objective components are the complexity of
the task and its compliance (or inconsistency) with the
conditions of execution:
– equipment and the possibility of its application for
solving problems;
– deadlines (sufficient or short);
– mission statement.</p>
      <p>Internal, mental components of activity:
– professional training;
– functional state (mood, health, fatigue, etc.) for the
duration of the task.</p>
      <p>A person who is not ready for psychophysiological
qualities to work that requires high emotional stability,
endurance, self-control and a strong nervous system,
under the influence of poor health or conflict, is inclined
to dishonestly perform the tasks assigned to him, thereby
causing irreparable damage to the task and the project as
a whole.</p>
      <p>When solving habitual, everyday duties, situational
and target settings may coincide in time and not be fixed
in the human mind.</p>
      <p>If decisive actions are required to solve a
nonstandard problem with a shortage of time, the formation
of the target setting becomes stressful. The more
controversial the situational and target settings, the
higher the level of stress experienced.</p>
      <p>For example, the desire to remove the driven unit
from a dangerous state and the realization that it is
impossible to do this for objective, situational reasons or
when the employee does not know how to do this in this
situation. This is a typical situation preceding a forced
shutdown of a reactor when there is not enough time or
information to understand what needs to be done to
prevent shutdown.</p>
      <p>The opposite picture may also occur - the operator
knows what is being done in the specific threat of block
exit from the given mode, but if there is even the smallest
risk, he prefers to provide work to an automatic
protection system that will stop the block.</p>
      <p>The knowledge gained as a result of long preparation
and training predetermines the success of the activity and
its productivity, sets such characteristics as accuracy,
infallibility, speed and overall reliability of the operator’s
actions.</p>
      <p>
        In the long term, the creation and application of
systems for the operational assessment of the situation
and the development of recommendations in the
conditions of uncertainty implemented on the basis of
studies developed in [
        <xref ref-type="bibr" rid="ref10 ref9">9,10</xref>
        ].
      </p>
    </sec>
    <sec id="sec-6">
      <title>7. Conclusion</title>
      <p>Further planns:
1. To continue work on the assessment of personnel
actions in the most dangerous accidents with the
emission of thermal neutron sources with a low flux
density.
2. To develop a methodological approach to solving the
problems of assessing doses of external and internal
exposure and assessing damage to the population
(taking into account the age composition of the
population) living around nuclear power plants
during the most dangerous accidents involving the
emission of thermal neutron sources with a low flux
density.
3. To develop an atlas of risk metrics and a program for
monitoring (control) the safety of nuclear power
plants.</p>
    </sec>
    <sec id="sec-7">
      <title>Acknowledgment</title>
      <p>The reported study was funded by RFBR according to
the research projects № 18-07-00225, 18-07-00909,
1807-01111, 19-07-00455, 20-04-60185.</p>
    </sec>
    <sec id="sec-8">
      <title>About the authors</title>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <surname>Abramova</surname>
            <given-names>V.N.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Belekhov</surname>
            <given-names>V.V.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Belskaya</surname>
            <given-names>E.G.</given-names>
          </string-name>
          <article-title>Psychological methods in working with personnel at nuclear power plants</article-title>
          . -M.:
          <string-name>
            <surname>Energoatomizdat</surname>
          </string-name>
          ,
          <year>1988</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <surname>Haunaman</surname>
            <given-names>G.W.</given-names>
          </string-name>
          and
          <string-name>
            <surname>Spurgin A.J. Systematic</surname>
          </string-name>
          <article-title>Human Action Reliability Procedure (SHARP)</article-title>
          ,
          <source>EPRI NP-3583</source>
          , Electric Power Research Institute, Palo Alto California,
          <source>Project 2170-3, Interim report. June</source>
          <year>1984</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>Per</given-names>
            <surname>Holmgren</surname>
          </string-name>
          «
          <article-title>Enchancing Conditions for Ignalina Nuclear Power Plant in Lithuania (RBMK-Reactor Type)</article-title>
          .
          <source>RELCON AB, Box 1288, S-172 25 Sundbyberg</source>
          , Sweden.
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <article-title>[4] Consumers power company and pickard, lowe and garrick, Midland Nuclear Plant Probabilistic Risk Assessment, Irvine</article-title>
          ,
          <string-name>
            <surname>С A</surname>
          </string-name>
          ,
          <year>1984</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>Peter</given-names>
            <surname>Jacobsson end Per</surname>
          </string-name>
          <article-title>Holmgren. «Human Interactions Analysis V3.3», BRP (93) 6</article-title>
          . (
          <year>1993</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <surname>Syngaevskaya</surname>
            <given-names>A.A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Berberova</surname>
            <given-names>M.A.</given-names>
          </string-name>
          ,
          <article-title>Derevyankin A.A. Studying the effect of stress on the probability of errors by NPP personnel as part of personnel reliability analysis /</article-title>
          / CPT-2018
          <source>: proceedings of the International Scientific Conference</source>
          . - ProtvinoMoscow: Publishing House. IFTI, p.
          <fpage>282</fpage>
          -
          <lpage>288</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>M.A.</given-names>
            <surname>Berberova</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.S.</given-names>
            <surname>Zolotarev</surname>
          </string-name>
          , «
          <article-title>NPP risk assessments results dependence study on the composition of the population living around the NPP (on the example of Rostov and Kalinin NPP)»</article-title>
          ,
          <source>GraphiCon 2019 Computer Graphics and Vision</source>
          .
          <source>The 29th International Conference on Computer Graphics and Vision</source>
          . Conference Proceedings (
          <year>2019</year>
          ), Bryansk, Russia,
          <source>September 23-26</source>
          ,
          <year>2019</year>
          , Vol-
          <volume>2485</volume>
          , urn:nbn:de:
          <fpage>0074</fpage>
          -
          <lpage>2485</lpage>
          -1, ISSN 1613- 0073, DOI: 10.30987/graphicon-2019-2-
          <fpage>285</fpage>
          -289, http://ceur-ws.
          <source>org/</source>
          Vol-
          <volume>2485</volume>
          /paper66.pdf, p.
          <fpage>285</fpage>
          -
          <lpage>289</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>M.A.</given-names>
            <surname>Berberova</surname>
          </string-name>
          ,
          <string-name>
            <surname>K.I.Chernyavskii</surname>
          </string-name>
          , «
          <article-title>Comparative assessment of the NPP risk (on the example of Rostov and Kalinin NPP)</article-title>
          .
          <article-title>Development of risk indicators atlas for Russian NPPs»</article-title>
          ,
          <source>GraphiCon 2019 Computer Graphics and Vision</source>
          .
          <source>The 29th International Conference on Computer Graphics and Vision</source>
          . Conference Proceedings (
          <year>2019</year>
          ), Bryansk, Russia,
          <source>September 23-26</source>
          ,
          <year>2019</year>
          , Vol-
          <volume>2485</volume>
          , urn:nbn:de:
          <fpage>0074</fpage>
          -
          <lpage>2485</lpage>
          -1, ISSN 1613-0073, DOI: 10.30987/graphicon-2019-2-
          <fpage>290</fpage>
          -294, http://ceurws.org/Vol-
          <volume>2485</volume>
          /paper67.pdf, p.
          <fpage>290</fpage>
          -
          <lpage>294</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <surname>Osipov</surname>
            <given-names>V.P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Chetverushkin</surname>
            <given-names>B.N.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Baluta</surname>
            <given-names>V.I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Nechaev</surname>
            <given-names>Yu.I.</given-names>
          </string-name>
          <article-title>A formal apparatus for modeling and interpreting antagonistic conflicts based on an electronic training ground</article-title>
          . KIAM
          <string-name>
            <surname>Preprints</surname>
            <given-names>M.V.</given-names>
          </string-name>
          <string-name>
            <surname>Keldysh</surname>
          </string-name>
          .
          <year>2018</year>
          . No. 181.P. 1-
          <fpage>28</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <surname>Baluta</surname>
            <given-names>V.I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Osipov</surname>
            <given-names>V.P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Chetverushkin</surname>
            <given-names>B.N.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Yakovenko</surname>
            <given-names>O.</given-names>
          </string-name>
          <string-name>
            <surname>Yu</surname>
          </string-name>
          .
          <article-title>Conceptual issues of model representation of conflicts / In collection: CPT2019 International Scientific Conference of the Nizhny Novgorod State University of Architecture and Civil Engineering and the Research Center for Physical and Technical Informatics</article-title>
          .
          <source>Proceedings of the International Scientific Conference</source>
          .
          <year>2019</year>
          .S.
          <volume>27</volume>
          -
          <fpage>39</fpage>
          . Berberova Maria A.,
          <source>PhD</source>
          , docent, MIREA - Russian Technological University,
          <article-title>Е-mail: maria.berberova@gmail.com Khakimova Aida Kh</article-title>
          .,
          <source>PhD</source>
          , docent, Kama Institute (Naberezhnye Chelny, Russia),
          <article-title>ANO «Scientific and Research Center for Information in Physics and Technique» (Nizhny Novgorod, Russia), Е-mail: aida_khatif@mail</article-title>
          .ru Zolotarev Oleg V.,
          <string-name>
            <surname>Ph</surname>
            .D., Docent,
            <given-names>ANO HE</given-names>
          </string-name>
          «Russian New University» (Moscow, Russia),
          <article-title>E-mail: ol-zolot@yandex</article-title>
          .ru
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>