1st International ICST Conference on Systems Engineering and Modeling

Research Article

Alleviating Complexity of Modeling in System Engineering Tools

  • @INPROCEEDINGS{10.1109/ICSEM.2007.373340,
        author={A. Soffer and S.  Cohen},
        title={Alleviating Complexity of Modeling in System Engineering Tools},
        proceedings={1st International ICST Conference on Systems Engineering and Modeling},
        publisher={IEEE},
        proceedings_a={ICSEM},
        year={2007},
        month={6},
        keywords={},
        doi={10.1109/ICSEM.2007.373340}
    }
    
  • A. Soffer
    S. Cohen
    Year: 2007
    Alleviating Complexity of Modeling in System Engineering Tools
    ICSEM
    IEEE
    DOI: 10.1109/ICSEM.2007.373340
A. Soffer1, S. Cohen1
  • 1: Technion-Israel Inst. of Technol., Haifa

Abstract

We present a strategy for balancing various aspects of modeling in order to control and manage complexity in model-based system engineering. Modern systems engineering methods need to support development and maintenance of increasingly complex systems. The system's complexity affects the engineering process as well as the tools that are used to support them. Consequently, system engineers who handle these processes face the dual challenge of coping with the complexity of the development and maintenance process and tools, in addition to the complexity of the system itself. Since the inherent svstem complexity cannot be reduced the system engineering methods and tools should perferably include a capability to manage and control complexity. How can this be done? In our study we analyzed and characterized the difficulties that system engineers encounter when engaging in model-driven development of a large intricate avionic system. We propose a set of measures of model complexity from the modeler's perspective - by applying general system complexity measures to the modeling environment and artifacts. Finally, a number of product- and system-independent methods were developed for alleviating complexity-related difficulties in a modeling environment. Our study shows that incorporating such techniques into the modeling process resulted in improved understanding of models, shorter modeling times and adequate consistency throughout the model's various levels of abstractions. We thus conclude that the addition of such mechanisms by vendors of system modeling environments may help improve the efficiency and. effectiveness of the system modeling procedure.