Details
Originalsprache | Englisch |
---|---|
Titel des Sammelwerks | Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021 |
Herausgeber/-innen | Bruno Castanier, Marko Cepin, David Bigaud, Christophe Berenguer |
Seiten | 3312-3318 |
Seitenumfang | 7 |
Publikationsstatus | Veröffentlicht - 2021 |
Veranstaltung | 31st European Safety and Reliability Conference, ESREL 2021 - Angers, Frankreich Dauer: 19 Sept. 2021 → 23 Sept. 2021 |
Publikationsreihe
Name | Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021 |
---|
Abstract
Most of the attempts aimed at substituting expert-driven human reliability assessment methods with empirical data-driven techniques have failed due to the high uncertainty of human reliability databases and limitations of traditional probabilistic tools to deal with it. Although recent research suggests Bayesian and credal networks could be a more suitable approach to model human reliability data, such analyses implies the need for the assessment of a conditional probability distribution for each variable – requiring a much larger amount of data than other traditional tools. Therefore, ‘the problem of sparse data’ continues to play a crucial role in hindering the feasibility and credibility of human reliability analysis. This has fuelled research aiming at tackling data scarcity through the use of expert elicitation and, more recently, of imprecise probability. In addition to issues inherent to the nature of the available data, some modelling procedures such as normalisation have the potential to implicitly affect the degree of knowledge carried by such data, resulting in loss of reliability. For instance, our confidence about the probability of an event that has been observed in only one of ten trials (1/10) is not the same as that of an event observed to occur ten times in one hundred trials (10/100). Hence, the output of such a procedure does not carry any information regarding the unevenness of sample sizes. In this paper, we propose to tackle these limitations by using confidence boxes (c-boxes) with credal networks, aiming at providing risk assessors with a rigorous framework for data uncertainty guiding towards more efficient and robust modelling solutions. The approach is tested with a simple model of the causes of fatigue in the work environment.
ASJC Scopus Sachgebiete
- Ingenieurwesen (insg.)
- Sicherheit, Risiko, Zuverlässigkeit und Qualität
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021. Hrsg. / Bruno Castanier; Marko Cepin; David Bigaud; Christophe Berenguer. 2021. S. 3312-3318 (Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Handling the uncertainty with confidence in human reliability analysis
AU - Morais, Caroline
AU - Ferson, Scott
AU - Moura, Raphael
AU - Tolo, Silvia
AU - Beer, Michael
AU - Patelli, Edoardo
N1 - Funding Information: This work has been partially supported by the Brazilian Oil & Gas Regulator and the Engineering and Physical Sciences Research Council (EPSRC) with the project entitled “A Resilience Modelling Framework for Improved Nuclear Safety (NuRes)”, Grant No. EP/R020588/2.
PY - 2021
Y1 - 2021
N2 - Most of the attempts aimed at substituting expert-driven human reliability assessment methods with empirical data-driven techniques have failed due to the high uncertainty of human reliability databases and limitations of traditional probabilistic tools to deal with it. Although recent research suggests Bayesian and credal networks could be a more suitable approach to model human reliability data, such analyses implies the need for the assessment of a conditional probability distribution for each variable – requiring a much larger amount of data than other traditional tools. Therefore, ‘the problem of sparse data’ continues to play a crucial role in hindering the feasibility and credibility of human reliability analysis. This has fuelled research aiming at tackling data scarcity through the use of expert elicitation and, more recently, of imprecise probability. In addition to issues inherent to the nature of the available data, some modelling procedures such as normalisation have the potential to implicitly affect the degree of knowledge carried by such data, resulting in loss of reliability. For instance, our confidence about the probability of an event that has been observed in only one of ten trials (1/10) is not the same as that of an event observed to occur ten times in one hundred trials (10/100). Hence, the output of such a procedure does not carry any information regarding the unevenness of sample sizes. In this paper, we propose to tackle these limitations by using confidence boxes (c-boxes) with credal networks, aiming at providing risk assessors with a rigorous framework for data uncertainty guiding towards more efficient and robust modelling solutions. The approach is tested with a simple model of the causes of fatigue in the work environment.
AB - Most of the attempts aimed at substituting expert-driven human reliability assessment methods with empirical data-driven techniques have failed due to the high uncertainty of human reliability databases and limitations of traditional probabilistic tools to deal with it. Although recent research suggests Bayesian and credal networks could be a more suitable approach to model human reliability data, such analyses implies the need for the assessment of a conditional probability distribution for each variable – requiring a much larger amount of data than other traditional tools. Therefore, ‘the problem of sparse data’ continues to play a crucial role in hindering the feasibility and credibility of human reliability analysis. This has fuelled research aiming at tackling data scarcity through the use of expert elicitation and, more recently, of imprecise probability. In addition to issues inherent to the nature of the available data, some modelling procedures such as normalisation have the potential to implicitly affect the degree of knowledge carried by such data, resulting in loss of reliability. For instance, our confidence about the probability of an event that has been observed in only one of ten trials (1/10) is not the same as that of an event observed to occur ten times in one hundred trials (10/100). Hence, the output of such a procedure does not carry any information regarding the unevenness of sample sizes. In this paper, we propose to tackle these limitations by using confidence boxes (c-boxes) with credal networks, aiming at providing risk assessors with a rigorous framework for data uncertainty guiding towards more efficient and robust modelling solutions. The approach is tested with a simple model of the causes of fatigue in the work environment.
KW - Bayesian networks
KW - C-boxes
KW - Credal networks
KW - Fatigue
KW - Human factors
KW - Human reliability analysis
UR - http://www.scopus.com/inward/record.url?scp=85135438623&partnerID=8YFLogxK
U2 - 10.3850/978-981-18-2016-8_575-cd
DO - 10.3850/978-981-18-2016-8_575-cd
M3 - Conference contribution
AN - SCOPUS:85135438623
SN - 9789811820168
T3 - Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021
SP - 3312
EP - 3318
BT - Proceedings of the 31st European Safety and Reliability Conference, ESREL 2021
A2 - Castanier, Bruno
A2 - Cepin, Marko
A2 - Bigaud, David
A2 - Berenguer, Christophe
T2 - 31st European Safety and Reliability Conference, ESREL 2021
Y2 - 19 September 2021 through 23 September 2021
ER -