0
Research Papers: Team Dynamics in D3

Concept Clustering in Design Teams: A Comparison of Human and Machine Clustering

[+] Author and Article Information
Chengwei Zhang

Department of Mechanical Engineering,
Tsinghua University,
A1003-1 Lizhaoji,
Beijing 100084, China;
Department of Mechanical Engineering,
University of California, Berkeley,
Berkeley, CA 94720
e-mail: zhangcw13@mails.tsinghua.edu.cn

Youngwook Paul Kwon

Department of Mechanical Engineering,
University of California, Berkeley,
2114 Etcheverry,
Berkeley, CA 94720
e-mail: young@berkeley.edu

Julia Kramer

Department of Mechanical Engineering,
University of California, Berkeley,
354/360 Hearst Memorial Mining Building,
Berkeley, CA 94720
e-mail: j.kramer@berkeley.edu

Euiyoung Kim

Mem. ASME
Jacobs Institute for Design Innovation,
University of California, Berkeley,
2530 Ridge Road,
Berkeley, CA 94720
e-mail: euiyoungkim@berkeley.edu

Alice M. Agogino

Fellow ASME
Department of Mechanical Engineering,
University of California, Berkeley,
415 Sutardja Dai Hall,
Berkeley, CA 94720
e-mail: agogino@berkeley.edu

1Corresponding author.

Contributed by the Design Theory and Methodology Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received February 20, 2017; final manuscript received July 18, 2017; published online October 2, 2017. Assoc. Editor: Harrison M. Kim.

J. Mech. Des 139(11), 111414 (Oct 02, 2017) (9 pages) Paper No: MD-17-1159; doi: 10.1115/1.4037478 History: Received February 20, 2017; Revised July 18, 2017

Concept clustering is an important element of the product development process. The process of reviewing multiple concepts provides a means of communicating concepts developed by individual team members and by the team as a whole. Clustering, however, can also require arduous iterations and the resulting clusters may not always be useful to the team. In this paper, we present a machine learning approach on natural language descriptions of concepts that enables an automatic means of clustering. Using data from over 1000 concepts generated by student teams in a graduate new product development class, we provide a comparison between the concept clustering performed manually by the student teams and the work automated by a machine learning algorithm. The goal of our machine learning tool is to support design teams in identifying possible areas of “over-clustering” and/or “under-clustering” in order to enhance divergent concept generation processes.

FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Processing flow of our proposed method

Grahic Jump Location
Fig. 2

An example half-sheet from team 1

Grahic Jump Location
Fig. 3

Heat map of concepts generated by team 1

Grahic Jump Location
Fig. 4

A depiction of an HM plot. The x-axis shows the index of human clustering, the y-axis shows the index of machine clustering, and the size of the circles shows the number of concepts in each cluster.

Grahic Jump Location
Fig. 5

A depiction of the under- and over-clustering pattern of part of the clusters on an HM plot

Grahic Jump Location
Fig. 6

Overview of 11 teams' HM plots. In the subplot, the x-axis shows the human clusters, the y-axis shows the machine clusters, and the size of the circles shows the number of concepts in each cluster. More description of the HM plots are available under the “Supplemental Materials” tab for this paper on the ASME Digital Collection.

Grahic Jump Location
Fig. 7

Sample HM plot of team 1

Grahic Jump Location
Fig. 8

Overview of under-clustering shown in teams' HM plots. Example: team 4 generated 12 concepts in cluster “children friendly designs” which the algorithm broke into six smaller clusters.

Grahic Jump Location
Fig. 9

Overview of over-clustering shown in teams' HM plots. Example: team 3 created two clusters “outside exercise” and “cycling” but the algorithm combined these into one larger cluster.

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