Finding parameter constraint networks in a product system

DSM 2012 Proceedings of the 14th International DSM Conference, Kyoto, Japan

Year: 2012
Editor: Onishi, M.; Maurer, M.; Kirner, K.; Lindemann, U.
Author: Hirao, A.; Oizumi, K.; Aoyama, K.
Series: DSM
Institution: The University of Tokyo, Japan
Section: Modelling Approaches and Information Acquisition
Page(s): 101-110
ISBN: 978-3-446-43354-0

Abstract

Several research papers on DSM have argued that, in order to realize the advanced management of product development, it is very useful to have a dependency network of parameters. In order to benefit from a dependency matrix, we require accurate and in-depth product information. Many modeling languages, such as SysML, have been developed to accurately represent information about
increasingly complex product systems. However, methodologies to simplify the representation of product systems have not been extensively discussed. This paper proposes a new support method for finding a parameter dependency network for a product system. This method is based on the assumption that product information can be classified into four elements, namely system functions,
functional roles, structural members, and parameters, and that possible dependencies between parameters can be deduced using the relationships between these elements (e.g., as a form of functionstructure diagram). This method is performed in three stages: 1) Suggestion of dependencies between parameters, 2) Extraction of constraints among parameters and generation of a parameter–constraint network, and 3) Generation of a feasible and better design process.

Keywords: Dependency network, constraint network, function structure model, product modeling, design process