Using Bayesian Modelling to Predict Software Incidents
Software Engineering
2021-02-02 v1
Abstract
Traditionally, fault- or event-tree analyses or FMEAs have been used to estimate the probability of a safety-critical device creating a dangerous condition. However, these analysis techniques are less effective for systems primarily reliant on software, and are perhaps least effective in Safety of the Intended Functionality (SOTIF) environments, where the failure or dangerous situation occurs even though all components behaved as designed. This paper describes an approach we are considering at BlackBerry QNX: using Bayesian Belief Networks to predict defects in embedded software, and reports on early results from our research.
Cite
@article{arxiv.2102.00293,
title = {Using Bayesian Modelling to Predict Software Incidents},
author = {Chris Hobbs and Waqar Ahmed},
journal= {arXiv preprint arXiv:2102.00293},
year = {2021}
}
Comments
14 pages, 3 Figures, Proceedings of the 29th Safety-Critical Systems Symposium (SSS'21) (https://scsc.uk/e683)