To main content

Qualities of long lived systems-of-systems

Qualities of long lived systems-of-systems

ICT solutions for smart cities are large and long-lived systems-of-systems. These are complex solutions with many actors and many heterogeneous systems involved. The systems involved must adapt both to different situations and to a totality that will emerge over time. To cope with these challenges, the systems must be flexible, i.e. they must be designed for system qualities such as:
  • Evolvability
    The large and long-lived systems-of-systems will evolve in an unpredictable way. The challenge is to design systems that can work in dynamic environments. We do research within domain architectures and languages, meta design, model driven development and architecture patterns and styles that address this quality.
  • Interoperability
    Today, interoperability between proprietary systems is solved on peer-to-peer basis. In the future, systems must interact in a dynamic and ad hoc way. We do research within model driven development and semantic technologies that will facilitate seamless exchange of information.
  • Configurability
    The system components will have to adapt to different environments and different user needs. In our work in the area of adaptive systems, we exploit dynamic adaptation by reconfiguring at the component level based on a runtime representation of the architecture model.
  • Scalability
    The systems may get many users, and the amount of data to be handled and exchange may grow considerably. We do research on optimal provisioning and use of information systems by means of model based analyses and prediction.
  • Traceability
    In the development of the systems, it is important to trace the backgrounds for the different system design decisions. In our work on model driven development we address creating and following of such traces.
  • Manageability
    In our work on domain architectures and languages and on model driven development, we aim for improved manageability of complexity through abstraction.
  • Reusability
    In our work on model driven development, we aim for higher level of reusability as the architectures, modelling languages and transformations that are generic for the domain.
  • Security
    Information security must be ensured. This is about confidentiality, privacy, authenticity and trustworthiness, integrity, availability, non-repudiation, accountability and auditability.
Senior Scientist