Using simulation to evaluate dynamic systems with weibull or lognormal distributions

Edifor, Ernest; Gordon, Neil; Walker, Martin; Papadopoulos, Yiannis

Abstract

© Springer International Publishing Switzerland 2014. Most techniques for quantitatively analysing the temporal fault trees of safety-critical systems are used with the assumption that the systems under study have exponentially distributed component failures. However, it is not impossible for real world systems to have various components with different failure distributions. This paper presents a simulation approach – Monte Carlo – for modelling, simulating and estimating the total system failure of state-of-the-art dynamic systems featuring Weibull or lognormal distributions. The proposed techniques have been formulated using the time-to-failure of these distributions to model temporal behaviours; they can be extended to model systems with other distributions.

Citation

Edifor, E., Gordon, N., Walker, M., & Papadopoulos, Y. Using simulation to evaluate dynamic systems with weibull or lognormal distributions. Presented at Ninth International Conference on Dependability and Complex Systems DepCoS-RELCOMEX, Brunów, Poland

• Presentation Conference Type: Conference Paper (published)
• Conference Name: Ninth International Conference on Dependability and Complex Systems DepCoS-RELCOMEX
• Acceptance Date: Feb 18, 2014
• Publication Date: Jan 1, 2014
• Deposit Date: Apr 26, 2022
• Journal: Advances in Intelligent Systems and Computing
• Print ISSN: 2194-5357
• Electronic ISSN: 2194-5357
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 286
• Pages: 117-187
• ISBN: 9783319070124
• DOI: https://doi.org/10.1007/978-3-319-07013-1_17
• Keywords: Safety-Critical systems; Temporal Fault Trees; Exponential Distribution; Weibull Distribution; Lognormal Distribution; Monte Carlo Simulation
• Public URL: https://hull-repository.worktribe.com/output/519670