Clustering design structure matrices: A comparison of methods using minimum description length

Amol Kulkarni, Connor Jennings, Michael Hoffman, Erica Blanco, Janis P. Terpenny, Timothy William Simpson

Research output: Contribution to conferencePaper

Abstract

Understanding interactions between components is fundamental in the design of products. Design Structure Matrices (DSMs) are often used to represent the relationships between every component or subsystem in a product. The complex network of interactions can then be clustered into subassemblies and other hierarchies, aiding designers in making critical decisions that will impact assembly, maintenance, and end-of-life disposal. This paper explores three methods for clustering components in a DSM to create a modular product architecture: (1) genetic algorithm, (2) hierarchical clustering, and (3) divisive clustering using a graph. A discussion on each algorithm is followed by an industrial example. This paper leads to the conclusion that genetic algorithm is better at identifying complex structures like bus module, 3D structure and overlapping cluster whereas hierarchical and divisive clustering are computationally inexpensive and are able to find optimal DSMs faster than the genetic algorithm.

Original languageEnglish (US)
Pages1114-1120
Number of pages7
StatePublished - Jan 1 2018
Event2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018 - Orlando, United States
Duration: May 19 2018May 22 2018

Other

Other2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018
CountryUnited States
CityOrlando
Period5/19/185/22/18

Fingerprint

Genetic algorithms
Complex networks
Optimal design

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Cite this

Kulkarni, A., Jennings, C., Hoffman, M., Blanco, E., Terpenny, J. P., & Simpson, T. W. (2018). Clustering design structure matrices: A comparison of methods using minimum description length. 1114-1120. Paper presented at 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018, Orlando, United States.
Kulkarni, Amol ; Jennings, Connor ; Hoffman, Michael ; Blanco, Erica ; Terpenny, Janis P. ; Simpson, Timothy William. / Clustering design structure matrices : A comparison of methods using minimum description length. Paper presented at 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018, Orlando, United States.7 p.
@conference{ba740aa8c86646279e54db1b926b7f8c,
title = "Clustering design structure matrices: A comparison of methods using minimum description length",
abstract = "Understanding interactions between components is fundamental in the design of products. Design Structure Matrices (DSMs) are often used to represent the relationships between every component or subsystem in a product. The complex network of interactions can then be clustered into subassemblies and other hierarchies, aiding designers in making critical decisions that will impact assembly, maintenance, and end-of-life disposal. This paper explores three methods for clustering components in a DSM to create a modular product architecture: (1) genetic algorithm, (2) hierarchical clustering, and (3) divisive clustering using a graph. A discussion on each algorithm is followed by an industrial example. This paper leads to the conclusion that genetic algorithm is better at identifying complex structures like bus module, 3D structure and overlapping cluster whereas hierarchical and divisive clustering are computationally inexpensive and are able to find optimal DSMs faster than the genetic algorithm.",
author = "Amol Kulkarni and Connor Jennings and Michael Hoffman and Erica Blanco and Terpenny, {Janis P.} and Simpson, {Timothy William}",
year = "2018",
month = "1",
day = "1",
language = "English (US)",
pages = "1114--1120",
note = "2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018 ; Conference date: 19-05-2018 Through 22-05-2018",

}

Kulkarni, A, Jennings, C, Hoffman, M, Blanco, E, Terpenny, JP & Simpson, TW 2018, 'Clustering design structure matrices: A comparison of methods using minimum description length' Paper presented at 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018, Orlando, United States, 5/19/18 - 5/22/18, pp. 1114-1120.

Clustering design structure matrices : A comparison of methods using minimum description length. / Kulkarni, Amol; Jennings, Connor; Hoffman, Michael; Blanco, Erica; Terpenny, Janis P.; Simpson, Timothy William.

2018. 1114-1120 Paper presented at 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018, Orlando, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Clustering design structure matrices

T2 - A comparison of methods using minimum description length

AU - Kulkarni, Amol

AU - Jennings, Connor

AU - Hoffman, Michael

AU - Blanco, Erica

AU - Terpenny, Janis P.

AU - Simpson, Timothy William

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Understanding interactions between components is fundamental in the design of products. Design Structure Matrices (DSMs) are often used to represent the relationships between every component or subsystem in a product. The complex network of interactions can then be clustered into subassemblies and other hierarchies, aiding designers in making critical decisions that will impact assembly, maintenance, and end-of-life disposal. This paper explores three methods for clustering components in a DSM to create a modular product architecture: (1) genetic algorithm, (2) hierarchical clustering, and (3) divisive clustering using a graph. A discussion on each algorithm is followed by an industrial example. This paper leads to the conclusion that genetic algorithm is better at identifying complex structures like bus module, 3D structure and overlapping cluster whereas hierarchical and divisive clustering are computationally inexpensive and are able to find optimal DSMs faster than the genetic algorithm.

AB - Understanding interactions between components is fundamental in the design of products. Design Structure Matrices (DSMs) are often used to represent the relationships between every component or subsystem in a product. The complex network of interactions can then be clustered into subassemblies and other hierarchies, aiding designers in making critical decisions that will impact assembly, maintenance, and end-of-life disposal. This paper explores three methods for clustering components in a DSM to create a modular product architecture: (1) genetic algorithm, (2) hierarchical clustering, and (3) divisive clustering using a graph. A discussion on each algorithm is followed by an industrial example. This paper leads to the conclusion that genetic algorithm is better at identifying complex structures like bus module, 3D structure and overlapping cluster whereas hierarchical and divisive clustering are computationally inexpensive and are able to find optimal DSMs faster than the genetic algorithm.

UR - http://www.scopus.com/inward/record.url?scp=85054018756&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054018756&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85054018756

SP - 1114

EP - 1120

ER -

Kulkarni A, Jennings C, Hoffman M, Blanco E, Terpenny JP, Simpson TW. Clustering design structure matrices: A comparison of methods using minimum description length. 2018. Paper presented at 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018, Orlando, United States.