Safety Modeling and Analysis

Contact

Dassault Systèmes : raphael.schoenig@3ds.com

Industry / Process

  •  Energy
  •  Aerospace
  •  Automotive

Short Description

The objective is to introduce concepts needed to build safety-oriented models allowing to perform risk analysis: events (failure, repair, reconfiguration, commands…), probability distributions, transitions, synchronizations (common cause failure). This approach is great improvement regarding the usual formalisms (fault trees, reliability block diagrams…) which are low level languages and suffer from a lack of expressiveness.

Zoom sur l'image

Long Description

Today, the risk assessment of a system is based on specific formalisms and techniques of analysis like Fault Trees, Reliability Bloc Diagrams, Markov Graphs, Monte-Carlo simulation… The objectives are to identify what can go wrong and what are the consequences on the system failure. These analysis are performed during the system design to improve performance and robustness to failure. However, the usual modeling language and techniques suffer from their lack of expressiveness, hierarchy and the models are built manually by the engineer from the system specifications which is a cause of modeling error. The objective of this workpackage is to propose to the safety engineer a high level formalism dedicated to the risk analysis. This formalism based on the grammar of Modelica language includes safety concepts like event (failure, repair…) and their features (probability laws), synchronizations and is convenient to handle with looped systems.

The benefits of this approach are:

  •   A complete integration in a System Engineering design process
  •   The capability to reuse and to share knowledge of a system
  •   Traceability between the system specifications and the safety models
  •   The capability to verify and validate the modeling with event-driven simulation and observation of failure propagation
  •   The possible compilation into a classical fault tree to deduce the minimal cutsets and the probabilities
  •   Make accessible to the non-specialists the safety-oriented modeling and analysis
  • Release History

    Release Date

    Release ID

    Compliant
    with Modelica

    Dependencies

    Tested with

     

     

     

     

     

    Links

    URL

    Information or News

    NONE

    NONE