Modeling and Verification for System of Systems
As software systems are used widely in every corner of our society, each software system to be developed and maintained gets larger and more complex than ever. Developing such software systems in a timely and cost-effective manner requires more careful analysis and modeling effort in order to reduce the total software life cycle cost. In addition to developing and maintaining such software systems, those large and complex systems developed separately need to be connected and work together to satisfy a common goal. We call such a system as System of Systems(SoS). An SoS consists of constituent systems, each of which is developed separately, but connected together to become a larger and more complex system. Some SoS examples are unmanned vehicles, military defense systems, health care systems, etc. In this talk, we would like to share with the audience, concept of the SoS, its characteristics, and modeling and verification issues for it. Depending on the type of the SoS, we try to model behavior of each and verify whether each can achieve the SoS goal quantitatively. We also introduce an attempt to model selfishness of each constituent system which abstracts behaviors of systems and verify the SoS level goal achievement.
Lecture at NEMO2017
Date/Time: Tuesday, July 18, 2017 at 16:00