Rare Event Simulation for Dynamic Fault Trees

Enno J.J. Ruijters; D.P. Reijsbergen; Pieter-Tjerk  de Boer; Mariëlle I.A. Stoelinga

doi:10.1007/978-3-319-66266-4_2

Rare Event Simulation for Dynamic Fault Trees

Enno J.J. Ruijters
, D.P. Reijsbergen
, Pieter-Tjerk de Boer
, Mariëlle I.A. Stoelinga

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

231 Downloads (Pure)

Abstract

Fault trees (FT) are a popular industrial method for reliability engineering, for which Monte Carlo simulation is an important technique to estimate common dependability metrics, such as the system reliability and availability. A severe drawback of Monte Carlo simulation is that the number of simulations required to obtain accurate estimations grows extremely large in the presence of rare events, i.e., events whose probability of occurrence is very low, which typically holds for failures in highly reliable systems.

This paper presents a novel method for rare event simulation of dynamic fault trees with complex repairs that requires only a modest number of simulations, while retaining statistically justified confidence intervals. Our method exploits the importance sampling technique for rare event simulation, together with a compositional state space generation method for dynamic fault trees.

We demonstrate our approach using three parameterized sets of case studies, showing that our method can handle fault trees that could not be evaluated with either existing analytical techniques, nor with standard simulation techniques.

Original language	English
Title of host publication	Computer Safety, Reliability, and Security
Subtitle of host publication	36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings
Editors	Stefano Tonetta, Erwin Schoitsch, Friedemann Bitsch
Publisher	Springer
Pages	20-35
Number of pages	15
ISBN (Electronic)	978-3-319-66266-4
ISBN (Print)	978-3-319-66265-7
DOIs	https://doi.org/10.1007/978-3-319-66266-4_2
Publication status	Published - 17 Aug 2017
Event	International Conference on Computer Safety, Reliability, and Security - Trento, Italy Duration: 12 Sept 2017 → 15 Sept 2017 http://safecomp17.fbk.eu/

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	10488
ISSN (Print)	0302-9743

Conference

Conference	International Conference on Computer Safety, Reliability, and Security
Abbreviated title	SAFECOMP 2017
Country/Territory	Italy
City	Trento
Period	12/09/17 → 15/09/17
Internet address	http://safecomp17.fbk.eu/

Access to Document

10.1007/978-3-319-66266-4_2

SAFECOMP2017Accepted author version, 427 KB

Rare event simulation for dynamic fault trees
Ruijters, E. J. J., Reijsbergen, D. P., de Boer, P.-T. & Stoelinga, M. I. A., 10 Mar 2017, Enschede: Centre for Telematics and Information Technology (CTIT). 14 p. (CTIT technical report series; no. TR-CTIT-17-02)
Research output: Book/Report › Report › Other research output

File

Cite this

Ruijters, E. J. J., Reijsbergen, D. P., de Boer, P.-T., & Stoelinga, M. I. A. (2017). Rare Event Simulation for Dynamic Fault Trees. In S. Tonetta, E. Schoitsch, & F. Bitsch (Eds.), Computer Safety, Reliability, and Security: 36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings (pp. 20-35). (Lecture Notes in Computer Science; Vol. 10488). Springer. https://doi.org/10.1007/978-3-319-66266-4_2

Ruijters, Enno J.J. ; Reijsbergen, D.P. ; de Boer, Pieter-Tjerk et al. / Rare Event Simulation for Dynamic Fault Trees. Computer Safety, Reliability, and Security: 36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings. editor / Stefano Tonetta ; Erwin Schoitsch ; Friedemann Bitsch. Springer, 2017. pp. 20-35 (Lecture Notes in Computer Science).

@inproceedings{2f6c9cf14203459ab8d349469372b77d,

title = "Rare Event Simulation for Dynamic Fault Trees",

abstract = "Fault trees (FT) are a popular industrial method for reliability engineering, for which Monte Carlo simulation is an important technique to estimate common dependability metrics, such as the system reliability and availability. A severe drawback of Monte Carlo simulation is that the number of simulations required to obtain accurate estimations grows extremely large in the presence of rare events, i.e., events whose probability of occurrence is very low, which typically holds for failures in highly reliable systems. This paper presents a novel method for rare event simulation of dynamic fault trees with complex repairs that requires only a modest number of simulations, while retaining statistically justified confidence intervals. Our method exploits the importance sampling technique for rare event simulation, together with a compositional state space generation method for dynamic fault trees. We demonstrate our approach using three parameterized sets of case studies, showing that our method can handle fault trees that could not be evaluated with either existing analytical techniques, nor with standard simulation techniques.",

author = "Ruijters, \{Enno J.J.\} and D.P. Reijsbergen and \{de Boer\}, Pieter-Tjerk and Stoelinga, \{Mari{\"e}lle I.A.\}",

year = "2017",

month = aug,

day = "17",

doi = "10.1007/978-3-319-66266-4\_2",

language = "English",

isbn = "978-3-319-66265-7",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "20--35",

editor = "Stefano Tonetta and Erwin Schoitsch and Friedemann Bitsch",

booktitle = "Computer Safety, Reliability, and Security",

address = "Germany",

note = "International Conference on Computer Safety, Reliability, and Security, SAFECOMP 2017 ; Conference date: 12-09-2017 Through 15-09-2017",

url = "http://safecomp17.fbk.eu/",

}

Ruijters, EJJ, Reijsbergen, DP, de Boer, P-T & Stoelinga, MIA 2017, Rare Event Simulation for Dynamic Fault Trees. in S Tonetta, E Schoitsch & F Bitsch (eds), Computer Safety, Reliability, and Security: 36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings. Lecture Notes in Computer Science, vol. 10488, Springer, pp. 20-35, International Conference on Computer Safety, Reliability, and Security, Trento, Italy, 12/09/17. https://doi.org/10.1007/978-3-319-66266-4_2

Rare Event Simulation for Dynamic Fault Trees. / Ruijters, Enno J.J.; Reijsbergen, D.P.; de Boer, Pieter-Tjerk et al.
Computer Safety, Reliability, and Security: 36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings. ed. / Stefano Tonetta; Erwin Schoitsch; Friedemann Bitsch. Springer, 2017. p. 20-35 (Lecture Notes in Computer Science; Vol. 10488).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Rare Event Simulation for Dynamic Fault Trees

AU - Ruijters, Enno J.J.

AU - Reijsbergen, D.P.

AU - de Boer, Pieter-Tjerk

AU - Stoelinga, Mariëlle I.A.

PY - 2017/8/17

Y1 - 2017/8/17

N2 - Fault trees (FT) are a popular industrial method for reliability engineering, for which Monte Carlo simulation is an important technique to estimate common dependability metrics, such as the system reliability and availability. A severe drawback of Monte Carlo simulation is that the number of simulations required to obtain accurate estimations grows extremely large in the presence of rare events, i.e., events whose probability of occurrence is very low, which typically holds for failures in highly reliable systems. This paper presents a novel method for rare event simulation of dynamic fault trees with complex repairs that requires only a modest number of simulations, while retaining statistically justified confidence intervals. Our method exploits the importance sampling technique for rare event simulation, together with a compositional state space generation method for dynamic fault trees. We demonstrate our approach using three parameterized sets of case studies, showing that our method can handle fault trees that could not be evaluated with either existing analytical techniques, nor with standard simulation techniques.

AB - Fault trees (FT) are a popular industrial method for reliability engineering, for which Monte Carlo simulation is an important technique to estimate common dependability metrics, such as the system reliability and availability. A severe drawback of Monte Carlo simulation is that the number of simulations required to obtain accurate estimations grows extremely large in the presence of rare events, i.e., events whose probability of occurrence is very low, which typically holds for failures in highly reliable systems. This paper presents a novel method for rare event simulation of dynamic fault trees with complex repairs that requires only a modest number of simulations, while retaining statistically justified confidence intervals. Our method exploits the importance sampling technique for rare event simulation, together with a compositional state space generation method for dynamic fault trees. We demonstrate our approach using three parameterized sets of case studies, showing that our method can handle fault trees that could not be evaluated with either existing analytical techniques, nor with standard simulation techniques.

U2 - 10.1007/978-3-319-66266-4_2

DO - 10.1007/978-3-319-66266-4_2

M3 - Conference contribution

SN - 978-3-319-66265-7

T3 - Lecture Notes in Computer Science

SP - 20

EP - 35

BT - Computer Safety, Reliability, and Security

A2 - Tonetta, Stefano

A2 - Schoitsch, Erwin

A2 - Bitsch, Friedemann

PB - Springer

T2 - International Conference on Computer Safety, Reliability, and Security

Y2 - 12 September 2017 through 15 September 2017

ER -

Ruijters EJJ, Reijsbergen DP, de Boer PT , Stoelinga MIA. Rare Event Simulation for Dynamic Fault Trees. In Tonetta S, Schoitsch E, Bitsch F, editors, Computer Safety, Reliability, and Security: 36th International Conference, SAFECOMP 2017, Trento, Italy, September 13-15, 2017, Proceedings. Springer. 2017. p. 20-35. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-66266-4_2

Rare Event Simulation for Dynamic Fault Trees

Abstract

Publication series

Conference

Access to Document

Fingerprint

Research output

Rare event simulation for dynamic fault trees

Cite this