Ph.D. Thesis Presentation by Angran Xiao
Tuesday, July 8, 2003

(Dr. Farrokh Mistree, advisor)

"Collaborative Multidisciplinary Decision Making in Distributed Environment"

Abstract

In distributed product realization, new paradigms and accompanying software systems are necessary to support the collaborative work of multidisciplinary engineering teams. Current researches focus mostly on the development of online tools and software frameworks that coordinate these tools. However, a gap exists between coordination technologies that integrate and manage these tools and methodologies that facilitate the teams’ decision making using these tools. On one side, each team holds its own perspective towards the product realization problem, and each controls a subset of design variables and seeks to maximize its own payoff function. On the other side, the interdisciplinary interactions tend to present additional challenges for teams to reach a consensus which is superior from the overall perspective. Therefore, the principal research goal in this dissertation is to develop the Collaborative Multidisciplinary Decision-making Methodology (CMDM) that bridges the gap between coordination technologies and decision making methodologies.

In the dissertation, four principal challenges are identified in the development of the CMDM. We construct clean digital interfaces between the product realization activities, which resolve the challenge of information communication in distributed environment. We postulate game theoretical principles to handle coupled problems, which resolves the challenge of collaborative decision making without incurring iteration. We then use design capability indices to help teams make ranged sets of decisions, which resolves the challenge of maintaining design freedom so the downstream teams can make superior decisions from the overall perspective. Finally we use utility theory to evaluate the weights of design goals, which solves the challenges of evaluating team’s preferences in the condition of uncertainties. Consequently, a product realization process is accomplished in a sequential manner without or with little iterations, wherein the final decision’s superiority can be guaranteed. Additionally, the CMDM helps fill the gap between coordination technologies and decision making methodologies due to the independence and autonomy of the individual activities.

Three product realization scenarios are presented to demonstrate the efficacy of the CMDM. An electronic chip package design and analysis scenario is used for explanation purpose. The Rapid Tooling TestBed (RTTB) system is then introduced, in which a light switch design and rapid prototyping scenario, and a robot arm design and injection molding scenarios are accomplished to further validate this methodology. It is proved the CMDM is efficacious at facilitating collaborative multidisciplinary decision making in distributed environment, with shorter time, lower cost and higher product quality.