Introduction —
Failure Analysis —
The Fault Tree —
The Bayesian Belief Network —
Application to the QNX Neutrino Microkernel —
Statistical Analysis —
Bathtub Curve —
Theoretical Foundations —
Conclusion —
Appendix: Bayes' Theorem
[+]
Introduction
When implementing a safety- or mission-critical application, it is essential to be able to present the argument that it meets the system requirements of availability and reliability ...
Failure Analysis
Two techniques are commonly employed for assessing the risks associated with the use of systems in safety- or mission-critical applications ...
The Fault Tree
Fault trees encapsulate the concept that the failure of a (sub)system can be caused by the failure of lower-level (sub)systems. Typical types of combination are ...
The Bayesian Belief Network
In theory, given a Fault Tree, failure rates and failure distributions can be associated with each leaf in a BNN and a computer program can then ...
Application to the QNX Neutrino Microkernel
While the trivial example shown in Figures 2 and 3 is useful to illustrate the technique of using a BBN to create a Fault Tree, this example is ...
Statistical Analysis
The methodology described above rests on the applicability of handling software failures statistically. It has been argued that ...
Bathtub Curve
Software failure rates do in fact follow the conventional bathtub curve. Everyone is familiar with the high failure rate of software when it is first released, and unanticipated usage patterns ...
Theoretical Foundations
Criticisms of the theoretical underpinnings of the statistical model can be answered by arguing, as do Littlewood et al, that the random ...
Conclusion
Fault Tree Analysis is a technique especially applicable to a mature product, such as the QNX Neutrino microkernel, where field usage figures and problem ...
Appendix — Bayes' Theorem
The Rev. Thomas Bayes published his famous theorem in the 18th century. If belief can be identified with probability, then the theorem allows reasoning from effect to cause ...
