FaultFlow: a tool supporting an MDE approach for Timed Failure Logic Analysis

In the application of software intensive systems to business-critical processes, tool-supported connection between the perspectives of Software Engineering and Reliability Engineering may largely help conciliation of the contrasting needs of reliability and complexity. In this paper, we present FaultFlow, a newborn tool enabling a Model-Driven Engineering approach for supporting specification, analysis, and simulation of failure logic in component-based systems. The tool exploits a meta-model combining the representation of system configuration hierarchies together with a decoupled representation of its stochastically-timed failure logic. FaultFlow is available in a as-a-service mode through a Web Application deployed on the cloud as well as in a standalone mode, featuring a well-crafted Java API enabling effective specification of the system hierarchy as well as of intra-component fault-to-failure and inter-component failure-to-fault propagation models, respectively starting from SysML Block Definition Diagrams and stochastic Fault Trees representations. The automated transformation towards Stochastic Time Petri Nets enables integration with external reliability tools for quantitative evaluation through numerical solution or simulation. We illustrate FaultFlow key benefits by experimenting the approach on a case study contextualised in an Internet of Things scenario modelling a Pollution Monitor System within a Smart City.