Nowadays, to achieve competitive advantage, the industrial companies are considering that success is sustained to great product development. That is to manage the product throughout its entire lifecycle. Achieving this goal requires a tight collaboration between actors from a wide variety of domains, using different software tools producing various product data types and formats. The actors' collaboration is mainly based on the exchange /share product information. The representation of the actors' viewpoints is the underlying requirement of the collaborative product development. The multiple viewpoints approach was designed to provide an organizational framework following the actors' perspectives in the collaboration, and their relationships. The approach acknowledges the inevitability of multiple integration of product information as different views, promotes gathering of actors' interest, and encourages retrieved adequate information while providing support for integration through PLM and/or SCM collaboration. In this paper, a multiple viewpoints representation is proposed. The product, process, organization information models are discussed. A series of issues referring to the viewpoints representation are discussed in detail. Based on XML standard, taking electrical connector as an example, an application case of part of product information modeling is stated.
OLLABORATIVE product development constitutes the core activity of many industrial companies. It includes actually a variety of business processes associated with the activities appearing in the product lifecycle. This is where PLM (Product Lifecycle Management) comes in. PLM system is the solution to support management of product information definition throughout its complete lifecycle. However, this Manuscript received January 30, 2007. H. Geryville, Y. Ouzrout and A. Bouras are with the CERRAL/LIESP laboratory, Lumiere University of Lyon, 160 Boulevard de l'Université, 69676 Bron Cedex, Lyon, FRANCE (phone: +33478772670; fax: +33478006328; email:
hichem.geryville@univ-lyon2.fr, yacine.ouzrout@univ-lyon2.fr, abdelaziz.bouras@univ-lyon2.fr).
N. Sapidis is with Department of Product and Systems Design Engineering, University of The Aegean, 84100 Syros, GREECE (e-mail: sapidis@aegean.gr).
system has not good filter methods for pertinent information extraction, especially to exchange and share information between actors. Also, Several different interpretations of the same information often exist, and many metadata contain information that is related or complementary to other metadata, but these different interpretations are rarely integrated, and often are only used together in an ad hoc way; so that an actor must keep switching back and forth between different applications and interfaces in the course of a single seek. In this context, the multiple viewpoints system is key element to integrate more than one interpretation approach across a body of information, gaining leverage in retrieval from their diversity and simplifying actor interactions by uniting them under a common interface. Where the multiple viewpoints approach permits to capture/retrieve the adequate information following the actors’ interests/objectives on the collaboration, based on his own experiences/knowledge.
The remainder of this paper is organized as follows: section 2 gives an overview of the main viewpoints approach. Section 3 is dedicated to the description of our multiple viewpoints system. In section 4, an example of product is studied and used for instantiating the viewpoints approach. The viewpoint instantiation is then used to show how an effective information extraction is achieved. The paper is ended with some concluding remarks and important perspectives in section 5.
We use the term viewpoint to mean a scheme for representing a collection of information objects, along with a mechanism for accessing this information. The term has also been used, with some frequency and great inconsistency, in the areas of information visualization and Human Computer Interfaces (HCI). Researchers such as Teraoka and Maruyama [1] are generally interested in representing a user’s interests and purpose. They use “multiple viewpoints” to parameter an information visualization system that indicates how to present information based on a particular interest profile. While a user’s interests might serve as the basis for a viewpoint, the viewpoints are not limited to different visualizations of the same information relationships; the relationships themselves may differ as well.
Ribière [2] considers “Viewpoint” as a polysemous word, Powell and French [3] introduce the concept of multiple viewpoints by using both multiple relevance judgments and multiple representations together. Collection selection [4] may also prove an important element in the design of multiple viewpoint algorithms that attempt to gain efficiency by consulting only a subset of the viewpoints in a system.
In the collaborative development, a lot of studies focusing on knowledge representation and product modelling during collaborative development process have been investigated [5], [6], [7]. However, most of them emphatically address only a part of the entire issue of product modelling. And, up until now, few papers have synthetically investigated product information modelling based on the users’ points of view on a large-system, from the design of product information modelling until there use in different application.
Geryville and all. [8] describe the concept of the multiple viewpoints as a composition of four components: the actor, the viewpoint’s domain, its related objectives (objectives of the actors in the collaboration), and its relationships with the other viewpoints. This definition is characterized by a context, which allows the restitution of the information that the actor wants use/retrieve, and a degree of importance of each viewpoint.
The scope of the presented works mainly deals with the integration of an information visualization interface, where the same information has different meaning following its relationships. However their limitation is that none of them integrate the real interests of the actors within collaborative product context, such as the extraction/retrieve of adequate information. In multidisciplinary collaboration, the actors need to integra
This content is AI-processed based on open access ArXiv data.
