Title :
Improving reliability studies with SysML
Author :
David, Pierre ; Idasiak, Vincent ; Kratz, Frédéric
Author_Institution :
Inst. PRISME, Ecole Nat. Super. d´´lngenieurs de Bourges, Bourges
Abstract :
In this paper, we introduce a methodology for the reliability analysis of complex systems during their design. The challenge of such a method is to be efficient and accessible for system engineers. We gear this method towards the reuse of functional models expressed in the widely used languages of OMG (Object Management Group). We discuss the improvements brought by SysML for reliability studies in comparison with previous works using UML. Namely, we underline its utilization for the automatic synthesis of FMEA and indicate how it may be expanded for other reliability analyses. We describe a new approach for the analysis of system models that, thanks to its focus on flows, enables the revelation of the failure mode propagation deducted from the functional model. The technique presented here can be implemented on commercial tools, because it strictly respects the SysML metamodel. Regarding our experience in using SysML, we can conclude that this new modeling language is well-suited to support at least the first step of reliability analysis (i.e. FMEA-like analysis) and shows potential for integrating other analyses concurrently with system design (e.g. Fault Tree analysis, failure scenario studies).
Keywords :
Unified Modeling Language; object-oriented programming; software reliability; software reusability; systems analysis; FMEA; OMG; Object Management Group; SysML; UML; failure mode propagation; metamodel; software reliability; software reuse; systems design; Failure analysis; Fault trees; Gears; Object oriented modeling; Process design; Reliability engineering; Software engineering; System analysis and design; Systems engineering and theory; Unified modeling language; FMEA; Reliability in design; SysML; UML;
Conference_Titel :
Reliability and Maintainability Symposium, 2009. RAMS 2009. Annual
Conference_Location :
Fort Worth, TX
Print_ISBN :
978-1-4244-2508-2
Electronic_ISBN :
0149-144X
DOI :
10.1109/RAMS.2009.4914731