Risk measures in living probabilistic safety assessment

Jan Holmberg, Gunnar Johanson, Ilkka Niemelä

Research output: Book/ReportReportProfessional

Abstract

Risk measures in various applications of probabilistic safety assessment (PSA) are defined and their use to support operational safety management of nuclear power plants is discussed. The basic risk measures represent the plant risk frequency (core damage frequency) either as a static value - nominal risk frequency or reference risk frequency - or as a plant configuration dependent, dynamic measure - instantaneous risk frequency. Generated risk measures are used in applications. The most significant application, or rather an objective of PSA, is the risk contributor identification in which risk importance measures are practical. Other applications as well as generated risk measures can be seen as advanced forms of risk contributor identification and risk importance measures, respectively. In the living use of PSA, plant status knowledge is used to represent actual plant safety status in monitoring or follow-up perspective. The PSA model must be able to express the risk given a time and plant configuration. This requires a development of time-dependent basic event models. The basic events can be divided to evident and hidden events, and the hidden events can be further divided to time-independent and time-dependent events. One problem is to develop realistic common cause failure models in this context. The applicability of the measures should be tested in order to find a practical set of quantities for the result presentation. There might be a danger that too many measures and results are presented. On the other hand, while some measures do not seem to be applicable at core damage frequency level evaluations, they might be descriptive at system and component level evaluations. The risk measures do not only characterize the plant safety status, but they reflect the structure of the PSA model. A proper use of the measures requires that decision making criteria are established. Probability or frequency based criteria are not sufficient in complex decision making situations. They might, however, give guidance or first indication about the acceptability of the decision alternative. An absolute criterion or measure is perhaps the correct quantity from the rationality point of view, but a relative criterion can be justified due to uncertainties in PSA.
Original languageEnglish
Place of PublicationEspoo
PublisherVTT Technical Research Centre of Finland
Number of pages76
ISBN (Print)951-38-4384-X
Publication statusPublished - 1993
MoE publication typeNot Eligible

Publication series

NameVTT Publications
PublisherVTT
No.146
ISSN (Print)1235-0621
ISSN (Electronic)1455-0849

Fingerprint

Decision making
Nuclear power plants
Monitoring
Uncertainty

Keywords

  • risk analysis
  • probability theory
  • safety engineering
  • assessments
  • evaluation
  • utilization
  • PSA
  • operations research
  • nuclear power plants
  • nuclear energy
  • reactor cares
  • damage
  • failure
  • frequencies
  • models
  • configurations (mathematics)
  • decision making
  • criteria

Cite this

Holmberg, J., Johanson, G., & Niemelä, I. (1993). Risk measures in living probabilistic safety assessment. Espoo: VTT Technical Research Centre of Finland. VTT Publications, No. 146
Holmberg, Jan ; Johanson, Gunnar ; Niemelä, Ilkka. / Risk measures in living probabilistic safety assessment. Espoo : VTT Technical Research Centre of Finland, 1993. 76 p. (VTT Publications; No. 146).
@book{4c3323ca040d4ab19d867ab19b8cc9d9,
title = "Risk measures in living probabilistic safety assessment",
abstract = "Risk measures in various applications of probabilistic safety assessment (PSA) are defined and their use to support operational safety management of nuclear power plants is discussed. The basic risk measures represent the plant risk frequency (core damage frequency) either as a static value - nominal risk frequency or reference risk frequency - or as a plant configuration dependent, dynamic measure - instantaneous risk frequency. Generated risk measures are used in applications. The most significant application, or rather an objective of PSA, is the risk contributor identification in which risk importance measures are practical. Other applications as well as generated risk measures can be seen as advanced forms of risk contributor identification and risk importance measures, respectively. In the living use of PSA, plant status knowledge is used to represent actual plant safety status in monitoring or follow-up perspective. The PSA model must be able to express the risk given a time and plant configuration. This requires a development of time-dependent basic event models. The basic events can be divided to evident and hidden events, and the hidden events can be further divided to time-independent and time-dependent events. One problem is to develop realistic common cause failure models in this context. The applicability of the measures should be tested in order to find a practical set of quantities for the result presentation. There might be a danger that too many measures and results are presented. On the other hand, while some measures do not seem to be applicable at core damage frequency level evaluations, they might be descriptive at system and component level evaluations. The risk measures do not only characterize the plant safety status, but they reflect the structure of the PSA model. A proper use of the measures requires that decision making criteria are established. Probability or frequency based criteria are not sufficient in complex decision making situations. They might, however, give guidance or first indication about the acceptability of the decision alternative. An absolute criterion or measure is perhaps the correct quantity from the rationality point of view, but a relative criterion can be justified due to uncertainties in PSA.",
keywords = "risk analysis, probability theory, safety engineering, assessments, evaluation, utilization, PSA, operations research, nuclear power plants, nuclear energy, reactor cares, damage, failure, frequencies, models, configurations (mathematics), decision making, criteria",
author = "Jan Holmberg and Gunnar Johanson and Ilkka Niemel{\"a}",
note = "Project code: S{\"A}H370012",
year = "1993",
language = "English",
isbn = "951-38-4384-X",
series = "VTT Publications",
publisher = "VTT Technical Research Centre of Finland",
number = "146",
address = "Finland",

}

Holmberg, J, Johanson, G & Niemelä, I 1993, Risk measures in living probabilistic safety assessment. VTT Publications, no. 146, VTT Technical Research Centre of Finland, Espoo.

Risk measures in living probabilistic safety assessment. / Holmberg, Jan; Johanson, Gunnar; Niemelä, Ilkka.

Espoo : VTT Technical Research Centre of Finland, 1993. 76 p. (VTT Publications; No. 146).

Research output: Book/ReportReportProfessional

TY - BOOK

T1 - Risk measures in living probabilistic safety assessment

AU - Holmberg, Jan

AU - Johanson, Gunnar

AU - Niemelä, Ilkka

N1 - Project code: SÄH370012

PY - 1993

Y1 - 1993

N2 - Risk measures in various applications of probabilistic safety assessment (PSA) are defined and their use to support operational safety management of nuclear power plants is discussed. The basic risk measures represent the plant risk frequency (core damage frequency) either as a static value - nominal risk frequency or reference risk frequency - or as a plant configuration dependent, dynamic measure - instantaneous risk frequency. Generated risk measures are used in applications. The most significant application, or rather an objective of PSA, is the risk contributor identification in which risk importance measures are practical. Other applications as well as generated risk measures can be seen as advanced forms of risk contributor identification and risk importance measures, respectively. In the living use of PSA, plant status knowledge is used to represent actual plant safety status in monitoring or follow-up perspective. The PSA model must be able to express the risk given a time and plant configuration. This requires a development of time-dependent basic event models. The basic events can be divided to evident and hidden events, and the hidden events can be further divided to time-independent and time-dependent events. One problem is to develop realistic common cause failure models in this context. The applicability of the measures should be tested in order to find a practical set of quantities for the result presentation. There might be a danger that too many measures and results are presented. On the other hand, while some measures do not seem to be applicable at core damage frequency level evaluations, they might be descriptive at system and component level evaluations. The risk measures do not only characterize the plant safety status, but they reflect the structure of the PSA model. A proper use of the measures requires that decision making criteria are established. Probability or frequency based criteria are not sufficient in complex decision making situations. They might, however, give guidance or first indication about the acceptability of the decision alternative. An absolute criterion or measure is perhaps the correct quantity from the rationality point of view, but a relative criterion can be justified due to uncertainties in PSA.

AB - Risk measures in various applications of probabilistic safety assessment (PSA) are defined and their use to support operational safety management of nuclear power plants is discussed. The basic risk measures represent the plant risk frequency (core damage frequency) either as a static value - nominal risk frequency or reference risk frequency - or as a plant configuration dependent, dynamic measure - instantaneous risk frequency. Generated risk measures are used in applications. The most significant application, or rather an objective of PSA, is the risk contributor identification in which risk importance measures are practical. Other applications as well as generated risk measures can be seen as advanced forms of risk contributor identification and risk importance measures, respectively. In the living use of PSA, plant status knowledge is used to represent actual plant safety status in monitoring or follow-up perspective. The PSA model must be able to express the risk given a time and plant configuration. This requires a development of time-dependent basic event models. The basic events can be divided to evident and hidden events, and the hidden events can be further divided to time-independent and time-dependent events. One problem is to develop realistic common cause failure models in this context. The applicability of the measures should be tested in order to find a practical set of quantities for the result presentation. There might be a danger that too many measures and results are presented. On the other hand, while some measures do not seem to be applicable at core damage frequency level evaluations, they might be descriptive at system and component level evaluations. The risk measures do not only characterize the plant safety status, but they reflect the structure of the PSA model. A proper use of the measures requires that decision making criteria are established. Probability or frequency based criteria are not sufficient in complex decision making situations. They might, however, give guidance or first indication about the acceptability of the decision alternative. An absolute criterion or measure is perhaps the correct quantity from the rationality point of view, but a relative criterion can be justified due to uncertainties in PSA.

KW - risk analysis

KW - probability theory

KW - safety engineering

KW - assessments

KW - evaluation

KW - utilization

KW - PSA

KW - operations research

KW - nuclear power plants

KW - nuclear energy

KW - reactor cares

KW - damage

KW - failure

KW - frequencies

KW - models

KW - configurations (mathematics)

KW - decision making

KW - criteria

M3 - Report

SN - 951-38-4384-X

T3 - VTT Publications

BT - Risk measures in living probabilistic safety assessment

PB - VTT Technical Research Centre of Finland

CY - Espoo

ER -

Holmberg J, Johanson G, Niemelä I. Risk measures in living probabilistic safety assessment. Espoo: VTT Technical Research Centre of Finland, 1993. 76 p. (VTT Publications; No. 146).