Skip to main content

MegaModelling at Runtime - scalable model-based framework for continuous development and runtime validation of complex systems.

Periodic Reporting for period 3 - MegaMaRt2 (MegaModelling at Runtime - scalable model-based framework for continuous development and runtime validation of complex systems.)

Reporting period: 2019-04-01 to 2020-03-31

European industry faces stiff competition on the global arena. Electronic Components and Systems become more and more complex, thus calling for modern engineering practices to be applied in order to better tackle both productivity and quality. Model-based technologies promise significant productivity gains, which have already been proven in several studies and applications. However, these technologies still need more enhancements to scale up for real-life industrial projects and to provide more benefits in different contexts.

The ultimate objective of improving productivity, while reducing costs and ensuring quality in development, integration and maintenance, can be achieved by using techniques integrating seamlessly design time and runtime aspects. Industrial scale system models, which are usually multi-disciplinary, multi-teams and serving several product lines have to be exploited at runtime, e.g. by advanced tracing and monitoring, thus boosting the overall quality of the final system and providing lessons-learnt for future product generations. MegaM@Rt brings model-based engineering to the next level in order to help European industry to reduce development and maintenance costs while reinforcing both productivity and quality.

To achieve that, MegaM@Rt provides a framework which incorporates methods and tools for continuous development and runtime validation to significantly improve productivity, quality and predictability of large and complex industrial systems. MegaM@Rt addresses the scalability challenges with advanced megamodelling and traceability approaches, while runtime aspects will be tackled via runtime trace analysis, monitoring, validation and verification. MegaM@Rt is based on a strong international consortium involving experts from France, Spain, Italy, Sweden, Czech Republic, and Finland. The partners cover the whole value chain from research organizations to tool providers, including 9 end-users with large industrial case studies for results validation.

Several improvement objectives have been defined in the beginning of the project and a set of KPIs have been used to measure the impact of the MegaM@Rt2 framework on the industrial case studies. In the best cases, the results showed an improvement up to 50% and 140% in different areas based on the 13 KPIs measured.

The work done by the project's team led to exploitable results on other complex projects. The solutions validated in the last part of the project have proven effective for European industry reducing development and maintenance costs while reinforcing both productivity and quality.
We have conducted the requirements elicitation and analysis for all the case studies to be fully supported by the MegaM@Rt2 Framework. Different types of requirements have been proposed following state-of-the-art software and systems architecture methods: conceptual (use case) requirements, framework requirements and toolset (technology) requirements. Gap analysis has been carried out in order to dice which requirements (features) are really needed and at which release of the Framework. These requirements have mapped to the definition of the baseline architecture of the toolsets brought by the partners, a preliminary high-level vision of the MegaM@Rt2 Framework. A first version of the Framework architecture roadmap has been proposed.

The tools and technology providers in the project have commenced the work on the proposed architecture initial and intermediate versions of the framework have been released to case study providers for evaluation. During the evaluation of the framework, case study providers also measured predefined key performance indicators against the baseline. The results of the evaluation and the measurements have been provided back to tool and technology providers for improving and developing further the framework.

In the second part of the project, further development of the framework has been done based on the feedback received in the first evaluation from case study providers. The developed tools have been integrated in the final version of the framework, re-evaluated on the case studies, and the new round of KPI evaluations was collected. In the same time, detailed description of the provided tools and methodologies have been made available to the community via a set of public project deliverables and via a public tool catalogue published on the project website.

In total 28 tools have been developed, of which a large share are provided as open source tools The tools are divided into three tool sets, for system engineering, runtime validation, and, respectively, for traceability and megamodeling.

The methodologies behind the tools and their scientific aspects have been published in 128 scientific publications of which in important share are journal articles. In addition, the results of the project have disseminated in several european and via bilateral collaboration with other European research projects. The results of the project have been disseminated not only via social media (LinkedIn, Twitter, Instagram, etc) and via the project web site, but also via public hackathons and workshops organized in conjunction academic events in Europe.
The project created a framework incorporating methods and tools for continuous system engineering and validation leveraging the advantages in scalable model-based methods to provide benefits in significantly improved productivity, quality and predictability of large and complex industrial systems. The combination of different methods and tool sets produces an important feedback from run time to design time level that improves the design. Preserved information from design time improves the traceability and quality of the end product. The Management and Traceability tool set addresses the scalability issues and the feedback loop of the model-based engineering methods.

More specifically, the project provides advancements of the state of the art on system modeling (e.g better integration of existing practices of model-based engineering, requirements coverage, executable modeling and simulation, HW/SW codesign, performance design and optimisation, static analysis), on runtime analysis (advanced traceability of requirements and scalability of model-based V&V activities, improved log and trace analysis) and megamodeling and traceability (traceability via common metamodel, model traceability and federation, and improved scalability for manipulation of large models)

The project has an indirect impact on society through accelerating the development and use of CPS scenarios, in transport and healthcare, where the impact is visible, and in other fields, notably energy and telecommunications where success is invisible to the consumer, but failures are particularly acute (that is to say the consumer benefit is in less outages or more reliable, cheaper or sustainable energy provision).

The megam@rt framework has been used for international projects. The framework optimized the development process and had a significant economic impact on all phases of the project where the framework was used.
The MegaM@Rt2 booth at EFECS 2018 in Lisbon
Producing a MegaM@Rt2 paper on Hackathons, to disseminate the project