Clustering Technique for DSMs

DSM 14 Proceedings of the 16th International DSM conference: Risk and Change management in complex systems

Year: 2014
Editor: Marle, F.; Jankovic, M.; Maurer, M.; Schmidt, D. M.; Lindemann, U.
Author: Behncke, F. G. H.; Maurer, D.; Schrenk, L.; Schmidt, D. M.; Lindemann, U.
Series: DSM
Section: Clustering and Optimization
Page(s): 177-186

Abstract

This paper provides a clustering technique (CT) for Design-Structure- Matrices (DSMs) that explore the entire solution space of cluster configurations (CCs) for a given system. Therefore the paper gives an overview of established CTs for exclusive clusters as basis for the development of an alternative CT that generates all possible CCs. These configurations are assessed against a set of performance metrics, which are selected by the decision makers to determine the quality of the cluster. Through a representation of the CCs in a portfolio according to the values of the performance metrics, decision makers are provided with a ranking of CCs as s support for their decision. Thereby, it is observed that established CTs do not capture the entire solution space of CCs and therefore miss comparable configurations. As a result, decision makers are not necessarily equipped with ideal CCs by established CTs.

Keywords: Clustering, Cluster Configuration, Solution Space Exploration