Combination of FMEA and stochastic DEA for risk analysis

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27200%2F17%3A10236308" target="_blank" >RIV/61989100:27200/17:10236308 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Combination of FMEA and stochastic DEA for risk analysis

Popis výsledku v původním jazyce

In this paper, we are interested in stochastic version of the combination of Failure Mode and Effects Analysis (FMEA) and Data Envelopment Analysis (DEA). In comparison with classical stochastic DEA approach, we consider uncertainties in the input values represented by normally distributed noise. In the case of FMEA, these input values represent three criteria to prioritize the failure modes - occurrence, severity, and detectability. The output considered in DEA method is Risk Priority Number (RPN) and it is given by a product of provided three random input criteria values. In the paper, we deal with non-gaussian distribution of this output by using first-order Taylor expansion which produces the linear combination of normally distributed variables, therefore we are able to follow the theory of classical stochastic DEA in very similar way. Our new approach is compared with Monte-Carlo sampling of input values provided to deterministic FMEA-DEA method.

Název v anglickém jazyce

Combination of FMEA and stochastic DEA for risk analysis

Popis výsledku anglicky

In this paper, we are interested in stochastic version of the combination of Failure Mode and Effects Analysis (FMEA) and Data Envelopment Analysis (DEA). In comparison with classical stochastic DEA approach, we consider uncertainties in the input values represented by normally distributed noise. In the case of FMEA, these input values represent three criteria to prioritize the failure modes - occurrence, severity, and detectability. The output considered in DEA method is Risk Priority Number (RPN) and it is given by a product of provided three random input criteria values. In the paper, we deal with non-gaussian distribution of this output by using first-order Taylor expansion which produces the linear combination of normally distributed variables, therefore we are able to follow the theory of classical stochastic DEA in very similar way. Our new approach is compared with Monte-Carlo sampling of input values provided to deterministic FMEA-DEA method.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20101 - Civil engineering

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

ESREL 2017: proceedings of the 27th edition of the international conference European Safety and Reliability Conference : June 18-22, 2017, Portorož, Slovenia

ISBN

978-1-138-62937-0

ISSN

—

e-ISSN

neuvedeno

Počet stran výsledku

8

Strana od-do

3027-3034

Název nakladatele

CRC Press

Místo vydání

Boca Raton

Místo konání akce

Portorož

Datum konání akce

18. 6. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—