Automatic generation of Temporal Fault Trees from AADL models

The Architecture Analysis and Design Language (AADL) is gaining growing acceptance in the aerospace, automobile and avionics industries. These industries are increasingly concerned with systems exhibiting sequence-dependent failures. About dependability (i.e. safety, reliability, availability and maintainability) analysis of AADL models, there is still a lack of techniques that can take into account the sequencing of failure events and determine minimal failure scenarios, i.e. which are made up of the relevant events causing a system to fail as a whole. In this paper, we present how we address this problem through an intelligent transformation, which captures the significant temporal ordering of faults and failures expressed by the AADL error models, to synthesise system Temporal Fault Trees (TFTs). © 2014 Taylor & Francis Group, London