0
Research Papers

Managing Design-Process Complexity: A Value-of-Information Based Approach for Scale and Decision Decoupling

[+] Author and Article Information
Jitesh H. Panchal

School of Mechanical and Materials Engineering, Washington State University, P.O. Box 642920, Pullman, WA 99164-2920panchal@wsu.edu

Christiaan J. J. Paredis

Systems Realization Laboratory, G.W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332chris.paredis@me.gatech.edu

Janet K. Allen

Systems Realization Laboratory, George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, 210 Technology Circle, Savannah, GA 31407janet.allen@me.gatech.edu

Farrokh Mistree

Systems Realization Laboratory, George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, 210 Technology Circle, Savannah, GA 31407farrokh.mistree@me.gatech.edu

In this paper, we assume that the decision maker uses utility functions to quantify the payoffs. Hence, the word “payoff” is used synonymously with “utility” in the rest of the paper.

J. Comput. Inf. Sci. Eng 9(2), 021005 (May 28, 2009) (12 pages) doi:10.1115/1.3130791 History: Received March 08, 2008; Revised July 25, 2008; Published May 28, 2009

Design-processes for multiscale, multifunctional systems are inherently complex due to the interactions between scales, functional requirements, and the resulting design decisions. While complex design-processes that consider all interactions lead to better designs, simpler design-processes where some interactions are ignored are faster and resource efficient. In order to determine the right level of simplification of design-processes, designers are faced with the following questions: (a) How should complex design-processes be simplified without affecting the resulting product performance? (b) How can designers quantify and evaluate the appropriateness of different design-process alternatives? In this paper, the first question is addressed by introducing a method for determining the appropriate level of simplification of design-processes—specifically through decoupling of scales and decisions in a multiscale problem. The method is based on three constructs: interaction patterns to model design-processes, intervals to model uncertainty resulting from decoupling of scales and decisions, and value-of-information based metrics to measure the impact of simplification on the final design outcome. The second question is addressed by introducing a value-of-information based metric called the improvement potential for quantifying the appropriateness of design-process alternatives from the standpoint of product design requirements. The metric embodies quantitatively the potential for improvement in the achievement of product requirements by adding more information for design decision-making. The method is illustrated via a datacenter cooling system design example.

FIGURES IN THIS ARTICLE
<>
Copyright © 2009 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Interaction patterns in multidisciplinary design (28)

Grahic Jump Location
Figure 2

Simplification of a system using intervals

Grahic Jump Location
Figure 3

Decision made using bounds on payoff. (UH=utility using Hurwicz criterion; H∗=utility using Hurwicz criterion at decision point; (Umax)∗=upper-bound expected utility at decision point, (Umin)∗=lower-bound expected utility at decision point; max(Umax)=maximum of upper-bound on utility throughout the design space.)

Grahic Jump Location
Figure 4

Steps for decision and scale decoupling

Grahic Jump Location
Figure 5

Multiple scales in the datacenter cooling system design

Grahic Jump Location
Figure 6

Illustration of datacenter design variables and parameters—cabinet level model

Grahic Jump Location
Figure 7

Illustration of datacenter design variables and parameters—computer level analysis

Grahic Jump Location
Figure 8

Coupling between the cabinet and computer level simulation-models

Grahic Jump Location
Figure 9

Results from decision-making using Patterns P1, P2, and P3 for different preference scenarios for the datacenter

Grahic Jump Location
Figure 10

Variation of the improvement potential for decision Patterns P4, P5, and P6 with weight for cost goal

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In