Skip to main content

Contributions

Openflexo is both a software infrastructure and the vehicle of a research programme on model federation. This page groups the contributions by theme; each links to the publications where it is presented. The full list is on the publications page.

Model federation​

The starting point is the observation that complex systems are described by many models, in different technical spaces, formalisms and paradigms, and that ensuring global consistency, traceability and cross-concern views is hard. Conceptual interoperability through Models Federation (2013, with a French companion article, De la nécessité de fédérer des modèles dans une chaîne d'outils) introduces the approach: models stay autonomous in their original technological spaces, and correspondences between their elements are made explicit and given a behaviour, instead of gathering everything in one space through transformations.

Later work applies and examines it:

  • modularity of heterogeneous multi-model systems, as an alternative to transformations between models (2016);
  • continuous requirements engineering, keeping a dynamic link between requirement models and their sources (2016);
  • traceability between informal requirements and formal specifications (2018);
  • the consistency of multi-model systems that evolve, with a focus on security requirements (2023, by other members of the team).

In 2024, 10 years of Model Federation with Openflexo: Challenges and Lessons Learned reports over a decade of practice, in close collaboration between a small software engineering company and academia: five use cases stemming from real industrial and academic needs, the lessons learned, the challenges of developing and maintaining a model-driven federation tool, and a road map. It received the Best Paper Award of the Practice Track of MODELS 2024. The same year, A Survey of Federative Approaches for Model Management in MBSE reviewed 23 papers selected from 59 contributions and classified them.

FML, a language for model federation​

FML (the Flexo Modeling Language) is the language in which federations are written. It makes it possible to conceptualise and to reify the interpretation of a source of information, and it has a designation mechanism to establish federation links to heterogeneous data sources. It can react to the behaviour and evolution of the federated models and program behaviours acting on them. It is formalised and validated on four use cases in the thesis FML: un langage de fédération de modèles pour l'interopérabilité sémantique de sources d'information hétérogènes (2023). Multi-Level Modeling with Openflexo/FML (2022) applies it to a multi-level process modeling challenge and discusses the advantages and limits of federation for multi-level modeling: see the MULTI Process Challenge.

Free modeling​

Free modeling lets modelers build models and their modeling language together, instead of choosing a language first. The poster Free The Modeling! (MiSE 2013) presents an agile way of building conceptual views expanding upon existing models and tools. Two papers on modeling situations (elementary actions on models and metamodels) present the idea and Openflexo as a free modeling tool (2014, 2015), and Using free modeling as an Agile method for developing domain specific modeling languages (2016) applies it to a local government project, Brest Métropole.

PAMELA and model-oriented programming​

PAMELA is the annotation-based Java modeling framework on which Openflexo's model objects are built. It blends programming and modeling: Java developers work at the conceptual level and at the source-code level, without code transformation or generation and without round-trip issues (PAMELA: an annotation-based Java Modeling Framework, 2021). Monitoring Association Constraints in Model-Oriented Programming (2023) uses it to reify complex association constraints so that they can be specified at development time and monitored at runtime.

Security​

The team also applies its modeling work to security:

  • Pimca, a domain-specific modeling language that captures the attacker's point of view of a system and its attack surface (2020, see the Cyber Threat Application);
  • the specification of security patterns with Design by Contract (2020) and with Security Contracts, which monitor them at runtime (2024);
  • the automatic synchronisation of a PAMELA security pattern with its formal specification in Alloy, through model federation with Openflexo, so that a correctness check runs whenever a deployed pattern evolves (2026, by other members of the team);
  • an approach to analyse a cyber-physical system while taking the attacker's interests into account (2022);
  • a framework for the secure systems engineering of space missions, developed in response to a European Space Agency tender (2023).

Design evolution​

The Design Multiverse (2026) calls for a scientific model of design evolution, with its branching, rework and context sensitivity, and proposes a research agenda to validate it experimentally.