Ph.D. Proposal Presentation by Marco Gero Fernández
Monday, September 27, 2004
(Dr. Farrokh Mistree, Co-Chair; Dr. Janet K. Allen, Co-Chair)
"A Framework for Agile Goal-Oriented Collaboration in Engineering Design"
The main objective in developing a Framework for Agile Goal-Oriented Collaboration in Engineering Design is that of devising a consistent means of supporting the collaboration of interacting domain experts, acting as stakeholders in a common product realization endeavor and sharing control over coupled design parameters, in light of evolving information along a design timeline. Specifically, the focus is on (1) formulating stakeholder design sub-problems in a consistent, flexible manner, (2) communicating decision critical information, required by interacting stakeholders effectively, and (3) modeling the ensuing stakeholder interactions in an openly adaptable manner. The underlying goal is that of providing guidance for the non-biased resolution of system level tradeoffs. To achieve this objective, the fundamental question addressed is -- How can collaboration in engineering design be modeled so as to facilitate the reflection of evolving stakeholder aspirations, while promoting the unbiased achievement of system level objectives?
It is in response to this question that an effective means of reconciling stakeholder objectives in light of ensuring system level performance is proposed. Specifically, design sub-problems, pertaining to individual decision-makers, are modeled using domain independent, modular, reusable decision templates based on Decision Support Problems. Similar templates, based on extensive form game theoretic protocols, are developed to model stakeholder interactions. The resulting constructs may be captured, archived, analyzed and manipulated on a computer. The proposed research draws on elements of object-oriented programming, port-based modeling, systems engineering, product architecture design, utility theory, game theory, open systems, and the effective synthesis of these constituents. Expected results of this research include an effective alternative to existing collaboration models that is focused on tactical iteration and strategic resolution of system level objectives. Additionally, (1) the development of modular, extensible, and computer interpretable decision models, (2) the development of flexible stakeholder interfaces, (3) the determination of stakeholder precedence in collaborative decision-making, and (4) the investigation of system level objective sensitivity to sub-problem design variables are required for developing a concise and consistent means of facilitating stakeholder interactions throughout a design process.