5.4.1 Simulation and Verification of [Dys]functional Behavior Models: Model Checking for SE

Abstract

Verification is a key process in the dependability engineering of complex systems. As we have shown in earlier works, formal verification techniques such as model checking can be efficiently used in a Systems Engineering (SE) context, despite their inherent complexity. Considering the widely used Enhanced Function Flow Block Diagrams (EFFBDs), we have indeed developed a formal simulation and verification tool for these functional behavior models. Moreover, great care has been taken to conceal the processing complexity from the tool end-user. In this paper, we present our latest developments as well as an extension of both method and tool to the case of dysfunctional models, to take into account failures affecting the model elements. By addressing both fault removal and fault forecasting problems with formal methods, we thus hope to improve the dependability analysis practice in SE.

1.3.3 Extending eFFBD formalism to task model

“…The main concern consists to evaluate how our extensions contribute to solve the notional gap between HF and SE disciplines. As another prolongation of our work, on the basis of previous works (Seidner et al 2010), we plan to define translation mechanism of xFFBD to other formalisms (eg. Timed Petri nets) in order to support model checking.…”

Managing the use of simulation in systems engineering: An industrial state of practice and a prioritization method

“…For supporters of MBSE, the superposition of modeling and simulation in SE provides a prospect that is an integral part of a cost-effective process to meet user requirements and needs (Kossiakoff & Sweet 2003). Simulation provides the system designer with a fair insight upon its behavior, which is particularly needed in the early design phases of a complex system (Seidner et al 2010). For example, analyzing simulation model output results are one way a systems engineer can early verify and/or validate a system requirement (MacCalman at al.…”

“…It increases the confidence of the developers in the design and fosters the correction of intrinsic conceptual design failures that usually cause high rework costs if found during system testing (Hoppe et al 2007). The combination of the simulation and verification tools help the system designer to understand the trace results given by model checkers, ultimately enhancing the use of the formal verification method (Seidner et al 2010). A significant amount of work has been devoted to develop the modeling infrastructure so that multidomain knowledge can be centrally managed and shared among stakeholders.…”

1.6.1 xFFBD: towards a formal yet functional modeling language for system designers

“…Moreover, a formal semantics has been established in (Seidner et al 2008). As a result, a simulation tool described in (Seidner 2009, Seidner et al 2010) has been developed. So, eFFBD formalism is here considered expressive enough to permit a direct execution of the model i.e.…”

“…To overcome this limitation, the designer may use a formal method such as model checking, where the identified properties are first formally expressed, then confronted to a formal model of the system, using efficient algorithms and data structures. Previous works have shown that the inherent complexity of model checking can be overcome and efficiently used in a SE context (Seidner et al 2010). Indeed, it is possible to:…”