The AMASS baseline deliverables have been published
The AMASS deliverables that present the baseline for the project have already been published. These deliverables review the state of the art and the state of the practice related to the four scientific and technical objectives of the project: Architecture-Driven Assurance, Multi-Concern Assurance, Seamless Interoperability, and Cross- and Intra-Domain Reuse. In addition, the deliverables propose a way forward for the work in AMASS to fill the gaps identified in existing solutions.
Baseline on Architecture-Driven Assurance: AMASS aims to provide a modelling language (metamodel), tools, and techniques to support architecture-driven assurance, i.e., an assurance that exploits and is linked to the system architecture in order to show system dependability. For system architecture modelling for assurance, it appears that there is currently a trend towards extending modelling languages (e.g. SysML) to better and explicitly support the concepts and needs from assurance standards. Concerning assurance patterns library management, further investigation needs to be carried out to develop enhanced libraries that cover not only safety argumentation patterns but also some other concerns, e.g. as security. Assurance of novel technologies might require the adaptation of standards’ requirements to address technology-specific needs. Finally, regarding contract-based assurance composition approaches, standard architectures (e.g., AUTOSAR in the automotive industry) require some safety/security architectural patterns definition and application, and auto-generation of platform models and configurations based on these patterns.
Baseline on Multi-Concern Assurance: The review of the state of the art on multi-concern assurance has covered co-design, co-analysis, co-V&V, and co-assurance of multiple dependability aspects. Especial attention has been paid to safety and security co-engineering and to the integration of these two concerns. The state of the practice has also been analysed for the application domains of the industrial case studies. The way forward on multi-concern assurance will focus on safety and security, dealing with challenges such as security-informed safety engineering and the creation of assurance cases that jointly justify security and safety. Nonetheless, other concerns (availability, performance, robustness, reliability…) must also be considered for multi-concern assurance, as evidenced by the current industrial practices and standards.
Baseline on Seamless Interoperability: AMASS has analysed the state of the art and the state of the practice related to seamless interoperability, especially on technologies for safety engineering and for safety-critical systems engineering. It is the same issue that application lifecycle management tools in software engineering have tried to solve for years, with varying success. Seamless interoperability is a cross-cutting concern across multiple architectural layers, therefore some technologies are orthogonal while others are mutually exclusive. Tool support for systems engineering is usually limited to point-to-point data exchange in some specific data format using import and export functionality. Nonetheless, recent technologies such as OSLC and ModelBus seem to be promising solutions to several challenges, e.g. “live” automated collaboration and data exchange. Using modern web technologies, it could be possible to close the gaps between tools and allow for a seamless integration
Baseline on Cross- and Intra-Domain Reuse: Process-based, product-based, and cross-concern aspects have been considered when analysing the current means for assurance reuse. Different methodological solutions exist to enable reuse of engineering or assurance artefacts, based on patterns, families, components, and models. The proposed way forward consists in a consolidation of existing results from OPENCOSS, SafeCer and other projects, and of some available technology on the market and state of practice. Based on the three dimensions (process, product, assurance case), methods and technologies that could enable systematic reuse include: an EPF Composer-based solution for processes; concerning products, contract-based reasoning, patterns, model-based principles applied to engineering, and variability management; for assurance cases patterns, variability management, contract-based, and module-based argumentation approaches, as well as model-based argumentation.