FOOBAR : an artificial intelligence based finite element system

This paper presents altematives to the traditional programming methodologies found in current finite element analysis systems. Artificial intelligence based representations are discussed, as are their incorporation in FOOBAR, a finite element system under development by the authors. Particularly useful architectures include constraint satisfaction paradigms, object-oriented techniques, and declarative knowledge representation. The development of FOOBAR is based on these ideas, which functionally decompose "what" to compute from "how" to compute it. We expect the resulting system to provide an intelligent, flexible environment for exploring research issues in finite element analysis. University Libraries Carnegie Mellon University Pittsburgh, Pennsylvania 15213 FOOBAR: An Artificial Intelligence Based Finite Element System John W. Baugh Jr. and Daniel R. Rehak Department of Civil Engineering, Carnegie Mellon University, Pittsburgh, PA 15213-3890, USA