Meta-modeling Design Expertise

  • Marcelo Bernal, META-MODELING DESIGN EXPERTISE, +8 authors Chris Paredis
  • Published 2016


ion The proposed methodology for meta-modeling provides an integrated repository for distributed design knowledge of a well-defined domain. This design knowledge from the case studies shows a taxonomy of objects that range from detailed specifications of physical components to very abstract conceptuality. To define the level of detail of the specifications of the meta-model, distinguishing the semantics, or meaning of the objects, from the complexity of their representations through any kind of computational tool. While a design concept is a generalization based on semantics, the representations can differ markedly, depending on the type of object and the means of representation. In this regard, the meta-model is defined in three layers of abstraction (Figure 5.2): object specification, geometric representation, and the mapping between the two. Every layer hides information from the next one (Parnas, 1972). The first layer captures the design concept, the properties, the attributes, and the relationships in a non-system specific language (Schaefer, 2006), avoiding any reference to the final means of representation such as parametric models or any kind of computational implementation in order to preserve the generality of the definition specified in the meta-model. The bottom layer corresponds to the actual representation of the objects through parametric tri-dimensional models or function. The final geometric model does not deal with any conceptuality or specification of objects with parametric modeling issues. This abstraction allows multiple mappings between a design specification and computational tools.

Cite this paper

@inproceedings{Bernal2016MetamodelingDE, title={Meta-modeling Design Expertise}, author={Marcelo Bernal and META-MODELING DESIGN EXPERTISE and John Haymaker and Russell T. Gentry and Ashok Kumar Goel and Dirk Shaefer and Dirk Schaefer and Axel Reichwein and George W. Woodruff and Chris Paredis}, year={2016} }