IVVES: Industrial-grade Verification and Validation of Evolving Systems

IVVES will develop the approaches in three directions to cover three main aspects in quality development for Evolving Systems (ES):

• V&V approaches dedicated to ML-enabled applications in ES, including innovation in ML-based testing, transparent and robust assessment of ML models and trained networks, and metrics for quality and trustworthiness of data used for ML training
• AI and data-driven V&V techniques dedicated to ES to cover areas that currently cannot be covered by state-of-the-art specification-based testing. This includes ML-based test case development and test automation, and continuous and runtime-enabled testing to detect quality degradation of ES during runtime.
• Smart engineering approaches that use data analytics to establish efficient and high-quality engineering processes for ES to improve coverage of testing in simulated environments prior to initial end-user exposure.

All IVVES methods, techniques and tools will be driven and evaluated by the case studies representing relevant industrial domains: Automotive & Transportation, Banking & Finance, Healthcare, Telecom, Industrial Automation, Agriculture & Forestry, Cybersecurity. To conclude, IVVES will develop cross-domain solutions with broad applicability that will be the foundation for standardization and certification. Thus, IVVES will shape a breakthrough in innovation power for European industry in AI-based systems and applications.