Integration of Constraint Programming and Model-Based Approach for System Synthesis

Abstract

Most of the work in the field of Model-Based System Engineering for the design of technical systems consists of implementing solution-oriented approaches. Several system modeling languages are available to represent fully defined systems from several points of view. It is also possible to link these descriptions with simulation or analysis tools to evaluate the solutions thus described. After having studied the limits of this way of designing system, we propose in this paper an approach oriented to the description of the design problem to be solved, through an adapted formalism called DEPS. This formalism allows a model-based approach for architecture and system synthesis. DEPS (Design Problem Specification) addresses problems of sizing, configuration, resource allocation and more generally of architecture generation or synthesis encountered in system design. The systems considered can be physical systems, software-intensive systems or mixed systems (embedded, mechatronical, cyber-physical). This language combines structural modeling features specific to object-oriented languages with problem specification features from constraint programming. We also present an integrated approach through the DEPS Studio environment, allowing DEPS modeling, model compilation and solving using an integrated constraint programming solver. This integration allows, among other things, the development and the debugging of models directly in DEPS rather than in the language of an external solver. The approach is illustrated on a simple case of electrical system synthesis.