Research Papers

Unlocking Organizational Potential: A Computational Platform for Investigating Structural Interdependence in Design

[+] Author and Article Information
Jesse Olson

 Northrop Grumman Corporation, Johnstown, PA 15901jesse.olson@alumni.cmu.edu

Jonathan Cagan

Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA 15213cagan@cmu.edu

Kenneth Kotovsky

Department of Psychology, Carnegie Mellon University, Pittsburgh, PA 15213kotovsky@cmu.edu

The algorithm was implemented in JAVA in just over 100,000 lines of code. There are 1120 domain methods and nearly 1000 domain variables (on the order of 10,000 when unpacking the arrays). It takes 1–2 min on a 2.80 GHz Intel Celeron to run a complete design proposal.

It may be worthwhile noting here that the process is not guaranteed to converge. In the more extreme cases when a system driver, such as ship mass or cost, grows too large the process will diverge. To avoid this, these drivers are capped at a prespecified maximum value.

J. Mech. Des 131(3), 031001 (Jan 26, 2009) (13 pages) doi:10.1115/1.3066501 History: Received November 28, 2007; Revised December 02, 2008; Published January 26, 2009

A team’s design—the structuring of its resources and flows of knowledge—is an important element determining its effectiveness. An essential element in achieving a team’s problem-solving potential is the role that interdependence, in both the task and the organization, plays in determining the dynamic and emergent system-level properties of the organization. In this paper, we present a computational platform for experimentally investigating the influence of informational dependencies found in the design of a complex system for exploring their role in determining system behaviors and performance. The approach presented in this paper is a multiagent simulation of the conceptual design of space mission plans by Team X, an advanced project design group at NASA’s Jet Propulsion Laboratory. The algorithm is composed of rich descriptive models of both the team-types and timing of interactions, collaborative methods, sequencing, rates of convergence- and the task-primary variables, their behaviors and relations, and the approaches used to resolve them. The objective is to create an environment of interaction representative of that found in actual design sessions. Better understanding how the dynamics arising from organizational and domain interdependencies impact an organization’s ability to effectively resolve its task should lead to the development of guidelines for better coping with task complexities, suggest ways to better design organizations, as well as suggest ways for improving the search for innovative solutions.

Copyright © 2009 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Relative rates of progression of individual agents described in team interviews

Grahic Jump Location
Figure 2

Schematic overview of the algorithm’s design loop

Grahic Jump Location
Figure 3

Algorithm base-case mission scope

Grahic Jump Location
Figure 4

Schematic overview of the direct heuristic solver method

Grahic Jump Location
Figure 5

Schematic overview of the iterative negotiation method

Grahic Jump Location
Figure 6

The observed and algorithm social network ties for all sessions in each proposal. Note that the data in these matrices are weighted (the weight indicates the amount of time individuals interacted, 1=10 min) and are nondirectional (this is because the exchange of information in these interactions was dominantly collaborative (two way)).

Grahic Jump Location
Figure 7

Magnitudes of the primary system drivers for Cassini and the algorithm

Grahic Jump Location
Figure 8

Percent mass distribution for 24 space missions (light line) taken from Wertz and Larson (22) and the algorithm (dark line)




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