Essay on Semantics Definition in MDE - An Instrumented Approach for Model Verification

Abstract

<p class="MsoNormal" style="margin: 0cm 0cm 0pt; layout-grid-mode: char;"><span style="font-family: Times New Roman; font-size: x-small;">In the context of MDE (Model-Driven Engineering), our objective is to define the semantics for a given DSL (Domain Specific Language) either to simulate its models or to check properties on them using model-checking techniques. In both cases, the purpose is to formalize the DSL semantics as it is known by the DSL designer but often in an informal way. After several experiments to define operational semantics on the one hand, and translational semantics on the other hand, we discuss both approaches and we specify in which cases these semantics seem to be judicious. As a second step, we introduce a pragmatic and instrumented approach to define a translational semantics and to validate it against a reference operational semantics expressed by the DSL designer. We apply this approach to the XSPEM process description language in order to verify process models.</span></p

A CPN/B method transformation framework for railway safety rules formal validation

“…The method proposed in this paper is based on informed transformation [17], which is in accordance with the mapping approach in the scope of Model Driven Engineering [18]. Informed transformation stipulates that a mapping of the source formal description of a formalism to the target one exists.…”

A Generic Tool for Tracing Executions Back to a DSML’s Operational Semantics

“…However, in the case that R was proved to be such a relation (typically, in a theorem prover, as in, e.g., [5]) , our algorithm nicely complements the proof, by obtaining the parameters R and n that it needs, and by explicitly finding which dsml executions match a given target-language execution. Specifically, -our algorithm requires the parameters R and n as inputs, and one can reasonably assume that these parameters characterise the bisimulation relation against which the model transformations was verified; hence, our algorithm benefits from that verification by obtaining two of its crucial inputs; -our algorithm provides information that model transformation verification does not: dsml executions that correspond to given target-language ones.…”

Contracts for Model Execution Verification

“…As already stated by previous works (such as [7,9,18]), a model execution requires that its meta-model defines several kinds of element specification, such as dynamic ones. Here, we propose our own meta-model part classification for an execution specification.…”

“…For instance, [4] verifies invariants or temporal constraints through the Maude framework and LTL properties. [7] proposes prioritized timed Petri nets while [19] stresses the expression of transformations towards colored Petri nets. [6] allows the specification of a behavioral semantics through abstract state machines (ASM).…”