Model-based Verification and Validation of Systems

At ESI, we are actively developing model-based verification and validation methodologies to empower engineers and architects to efficiently and effectively design, implement, and qualify systems with high confidence and guarantees on quality, while reducing cost, effort, and lead times.

The strategy is to capture the implicit knowledge of domain experts and the informal descriptions in natural language that describe system requirements and designs at different levels of abstraction, and translate them into precise and unambiguous models—such as ComMA, SysML, and BPMN.

As a consequence, specification models become analyzable and enable many opportunities to automate a range of design and testing activities (with a focus on functional aspects) across the systems engineering lifecycle. This improves efficiency and effectiveness, while providing measurable insights and guarantees on product quality.

Solving 5 key industry challenges

Sound & aligned specifications

Ensure soundness of requirements and design specifications at different levels of abstraction and conformance between these levels.

Collaborative, evolvable modeling

Support collaborative modeling and evolution triggered by updates and upgrades

Efficient design & testing

Improve efficiency and effectiveness of design and testing processes.

Reduced modeling effort

Make adoption easier by reducing modeling effort

Continuous knowledge integration

Enable continuous integration of knowledge into models from field data and reported defects and prioritize testing efforts.

How it works?

The methodology addresses these challenges by incorporating and extending state-of-the-art research on:

Modeling Frameworks

Formal semantics of modeling languages, capturing product-line variability, model checking, correct-by-construction design methods, change impact analysis, model learning and model management strategies.

Test Automation

Model-based testing & run-time monitoring techniques, regression test selection and semi-automated root cause analysis of failing tests.

Code Synthesis

Executable mock/stub generation to support testing activities, or to de-risk choice of hardware and middleware platforms during design-time.

Engineering Artefacts

Documentation generation from specification models.

Successful stories