EIS Lab homeResearch Snapshots, Accolades, ...Faculty, Staff, Students, and AlumniPapers, Reports, Theses, ...Analysis Theory & Methodology, X-Analysis Integration (XAI), Change Management, Engineering Information Technology, ... Projects, Sponsors, Toolkits, ...Conferences, Workshops, Thesis Presentations, ... Georgia Tech search engineCourses, Tools, Related Organizations, Directions & Locale Guides, ...

Knowledge Composition for Efficient Analysis Problem Formulation - Part 2: Approach and Analysis Meta-Model

Citation

Bajaj M, Peak RS, Paredis CJJ (2007) Knowledge Composition for Efficient Analysis Problem Formulation - Part 2: Approach and Analysis Meta-Model. Paper DETC2007-35050, Proc ASME CIE Intl Conf, Las Vegas.

Keywords

simulation-based design (SBD), problem formulation, CAD-CAE interoperability, design-analysis integration (DAI), knowledge composition methodology (KCM), variable topology multi-body (VTMB) problem, model transformation, NIST Core Product Model (CPM2), SysML

Abstract

In Part 1 we presented technical background and a gap analysis leading to the identification of five requirements for a methodology for efficient formulation of analysis problems for VTMB design alternatives. These requirements are founded on (a) abstraction of analysis knowledge as modular, reusable, computer-interpretable, analyst-intelligible building blocks, and (b) automated creation, reconfiguration, and verification of analysis models.

In this paper (Part 2), we present an example scenario to overview the Knowledge Composition Methodology (KCM) that is aimed at satisfying these requirements. The methodology is founded on analysis knowledge building blocks and a model transformation process based on graph transformations. With KCM an analyst may automatically compose an analysis model from a design model and these building blocks.

In this paper, we focus on the analysis knowledge component of this methodology (illustrated for structural and thermal disciplines), and describe four dimensions of analysis knowledge. Using these dimensions, we develop a decision template for analysts to create specifications for analysis models. Analysis models can be automatically created from a given specification using model transformation techniques (not described in this paper). We leverage the notion of choices and decisions to (a) define primitive and complex building blocks of analysis knowledge, and (b) formalize an analysis meta-model that represents the structure of analysis models. We also relate this analysis meta-model to the NIST Core Product Model (CPM2).

The envisioned methodology impact is a formal and systems-oriented foundational approach for analysis problem formulation that is time- and cost-effective.

Related Resources