On the Development of a Computing Infrastructure that Facilitates IPPD from a Decision-Based Perspective

† ‡ § This is white Integrated Product and Process Development (IPPD) embodies the simultaneous application of both system and quality engineering methods throughout iterative design processes. The use of IPPD results in the timeconscious, cost-saving development of engineering systems. A computing infrastructure called IMAGE is designed to implement IPPD from a decision-based perspective. IMAGE has four components: designer activities, available assets, agent collaboration, and a computing architecture. IMAGE captures a designer's activities through a timeline partitioning scheme, problem formulation and solution, and comprehensive information management. To support these activities, IMAGE incorporates design resources through the use of agents. Agents are a critical computational enabling technology that provide accountable mechanisms for resource collaboration in an integrated computing environment. Background and Motivation Considerable time and effort has been invested in the development of new computing technologies and their associated methods for Integrated Product and Process Development (IPPD). These technologies have been applied in systems that emphasize modularity, interdisciplinary program utilization, resource collaboration, and distributed processing. These systems include integrated design frameworks, conceptual design systems, and quasi-procedural systems.1-9 Though these systems have marked improvements in information processing, their applicability to aiding a designer in making decisions based on new design knowledge remains questionable.10 Furthermore, the applicability of these systems to * Graduate Researcher, Student Member AIAA, Aerospace Systems Design Laboratory, Corresponding Author [email protected] http://www.cad.gatech.edu/image † Professor, Member AIAA, Aerospace Systems Design Laboratory ‡ Professor, Member AIAA, Systems Realization Laboratory § Professor, Member AIAA, Aerospace Systems Design Laboratory Copyright © 1995 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. continuous, iterative design processes has not been proven and is uncertain, at best. The characteristics of a computing environment that enable a designer to perform design activities through the use of state-of-the-art computing technologies are summarized in this paper. The development is founded upon several underlying principles concerning the means by which the computing architecture embodies design activities. These principles are as follows: • A designer would like to do more with less throughout all design processes. A designer would like to have more robustness, greater openness, and better efficacy while expending fewer resources, paying less cost, and reducing the number of design cycles. • The principal role of a designer is that of a decision-maker. Decision-making activities can be marked by discrete milestones in a design process. This designer-centered approach forms the basis of Decision Based Design (DBD) paradigm for IPPD. • The Decision Support Problem (DSP) Technique is one embodiment of DBD with which a designer can manage design activities. The DSP Technique has two phases. First, a designer partitions a design timeline into manageable sub-problems, called Support Problems. Second, the resulting Support Problems are solved. • Support Problems are the basic design elements that represent design processes. • A computing infrastructure will support design activities throughout an entire design timeline. This is true for both product and process design. • A product model must support both form and function representations as well as the process by which this is done. This model is inherently object-oriented and complex inter-relationships and hierarchies exist. • Computer design environments should be designed to assist a designer in performing design activities; therefore, a designer is "in the loop". Automation should exist only when an appropriate context can be added to design information produced by automated processes.