Designing and developing highly engineered products requires direct (and more dynamic) associations between customers’ specifications and product characteristics (or its behaviors). In order to meet the specified customer performance, cost, and integrity goals, a multitude of specialized analyses, heuristics, shortcuts, look-up tables, equations, algorithms, finite elements, and material substitution at multiple levels (system, subsystems, components and parts) are ought to be performed. The product geometries of such engineered product are often complex and many parts are designed interactively from scratch using a 3D commercial computer-aided design (CAD) — lately often referred as Product Life-cycle Management (PLM) system. Today, this very “PLM-based” engineered product-design process is often “static”, very “feature or geometry-dependent,” “knowledge-intensive,” and therefore, engineers often takes considerable time (months) to complete this manual process. Today, more and more companies want to quickly reengineer a product from a multitude of family solutions (corresponding to various design trade-off studies). They are interested in some dynamic form of a decision-based system that could automatically filter through a multitude of historical product solutions and quickly reconfigure one that meets the customer requirements with the least cost, weight, and time investment. Such decision-based product automation is not an easy task by any means. Product definitions without knowing specific geometry are hard to conceive, capture generically, and reuse widely (via any generative tool). A typical product development process —by its nature—is highly dynamic, nonlinear, discrete, feature-dependent, and part-dependent. The solution is not easy, since problem formulation is time-bound, has numerous discrete inputs, topologies, and several mathematical discontinuities. This paper discusses the system architecture of the Knowledge-driven Automation (KDA) program — established at Parker in 2002. It addresses many of the above product development issues and problems. In particular, authors describe a Knowledge-based System Engineering Process for Obtaining Engineering Design Solutions in a Commercial PLM Setting. The architecture and solutions use a number of innovative knowledge-based engineering (KBE) concepts and procedures. Through strategic use of generative modeling, spreadsheet tables, part and assembly templates, system engineering concepts, and our proprietary “smart part concepts,” authors were able to engineer-to-configure a family of hydraulic actuators automatically from their customer specifications using a set of PLM (CATIA V5 and its underlying knowledgeware) tools.
Modeling Language for Systems Engineering in Defense Industry