Comparison of Information Passing Strategies in System-Level Modeling

摘要

Research on complex system optimization has focused on areas including algorithms, coordination strategies, and communication tools. This paper considers optimization from the perspective of information coordination during the solution process. This work aims to determine the potential impact of factors including the role of a system facilitator in managing system-level tradeoffs, the influence of an ideal information scenario, the choice of system variables, and the amount and form of passed information. The goal is not to develop the "best" communication structure or optimization algorithm, but to understand the significance of these factors. A traditional multidisciplinary design optimization coordination framework, a game theoretic approach, and a modified game theory approach are used to mimic design team behaviors. Results suggest these factors can influence the coordination process outcomes: The choice of system variable plays a significant role in the optimality of the final design; perfect information does not necessarily reduce the number of iterations or improve optimality; a system facilitator aids in making better design tradeoffs, though it is possible to achieve optimality without one; and the stability of the design cycle depends on the amount of passed information. This suggests a design team should carefully analyze the initial design space to determine an appropriate design formulation before applying optimization.