Automatic inference of fault tree models via multi-objective evolutionary algorithms

[For the latest version of this repository go to: https://gitlab.utwente.nl/fmt/fault-trees/ft-moea.git]

We present a novel approach to automatically infer Fault Tree (FT) models from a failure data set via multi-objective evolutionary algorithms (MOEAs), where the main contributions are that (i) we show it is possible to achieve more consistent and efficient FT structures via MOEAs, which simultaneously minimize the fault tree size, the error computed based on the failure data set, and the error based on Minimal Cut Sets (ii) we propose a metric to compare FT structures via Minimal Cut Sets using the RV-coefficient, (iii) we carry out a parametric analysis that explains the performance of the algorithm under different assumptions.

Edition May 17, 2022: We add some results concerning the parametric analysis we carried out to evaluate the effects of noise on the FT inference process.