Rule-based generation of model structures in multifaceted modeling and system design

A framework for aiding the construction of simulation models in system design problems is presented. The framework employs concepts of artificial intelligence and simulation modeling. A knowledge representation scheme called system entity structure expresses information about the structure of the system to be designed and its corresponding models. More specifically, the entity structure represents objects and their attributes, decompositions, and taxonomies. A knowledge base of production rules is defined for a given design domain and is incorporated in an expert system shell which recommends a feasible configuration of design objects from the system entity structure, based on specific design objectives, constraints, and requirements. This configuration is a basis for constructing a model of the system. The methodology for constructing the system entity structure and its corresponding rule base is presented. A case study based on a high level robot design problem is discussed to illustrate the conceptual framework.