Please use this identifier to cite or link to this item: https://doi.org/10.7511/jslx201306005
Title: Effect of bivariate distribution construction methods on series system reliability
Authors: Li, D.-Q.
Jiang, S.-H.
Zhou, C.-B.
Phoon, K.-K. 
Keywords: Joint probability density function
Pearson correlation coefficient
Probability of failure
Series system
Spearman correlation coefficient
Issue Date: Dec-2013
Source: Li, D.-Q.,Jiang, S.-H.,Zhou, C.-B.,Phoon, K.-K. (2013-12). Effect of bivariate distribution construction methods on series system reliability. Jisuan Lixue Xuebao/Chinese Journal of Computational Mechanics 30 (6) : 777-782. ScholarBank@NUS Repository. https://doi.org/10.7511/jslx201306005
Abstract: The two approximate methods for constructing bivariate distributions, namely method P and method S, are briefly introduced first. Thereafter, the closed-form expressions for calculating the series system probability of failure using direct integration are derived. For two negatively correlated performance functions underlying a series system, a formula for calculating the upper bound of probability of failure for a series system is derived. Then, an illustrative example is presented to demonstrate the capability and validity of two approximate methods. The results indicate that the methods P and S are effectively the same from a numerical viewpoint. Both two approximate methods can produce sufficiently accurate probabilities of failure for series systems. The two approximate methods provide a tool for series system reliability analysis under incomplete probability information. The errors in series system probability of failure increase with decreasing system probability of failure when the two performance functions underlying two components are positively correlated. They will decrease with decreasing system probability of failure when the two performance functions underlying two components are negatively correlated. The maximum error in the series system probability of failure may not be associated with a large correlation. It can happen at an intermediate correlation.
Source Title: Jisuan Lixue Xuebao/Chinese Journal of Computational Mechanics
URI: http://scholarbank.nus.edu.sg/handle/10635/90944
ISSN: 10074708
DOI: 10.7511/jslx201306005
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

1
checked on Mar 8, 2018

Page view(s)

43
checked on Apr 20, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.