Assessing the Quality of B Models

This paper proposes to define and assess the notion of quality of B models aiming at providing an automated feedback on a model by performing systematic checks on its content. We define and classify classes of automatic verification steps that help the modeller in knowing whether his model is well-written or not. This technique is defined in the context of ``behavioral models´´ that describe the behavior of a system using the generalized substitutions mechanism. From these models, verification conditions are automatically computed and discharged using a dedicated tool. This technique has been adapted to the B notation, especially on B abstract machines, and implemented within a tool interfaced with a constraint solver that is able to find counter-examples to unvalid verification conditions.