Re-Use of Semantic Product Knowledge in New Product Design Processes (RESPOND)

Funding: BMBF IngenieurNachwuchs 2009 (Informatik)
Project Runtime: June 2009 – December 2012

What is RESPOND?

RESPOND – Re-Use of Semantic Product Knowledge in New Product Design Processes is a project partly funded by the German Ministry for Education and Research (BMBF). It's total duration is set to three and a half years (June 2009 - December 2012). In cooperation with organizations of different branches, the research team investigates the communication between customer, vendor, and salespersons in early stages of industrial design-projects. On basis of these insights, we'll develop Distributed Intelligent Design Systems (DIDS) that support the conceptualization of customized design solutions. In this context, we utilize cloud computing approaches, semantic technologies and learn and inference mechanisms from the field of Artificial Intelligence to realize intelligent information services that specifically support salespersons in the development of new design projects.

Following the Situational Design Methodology for Information Systems (SiDIS) we conducted a structured analysis of typical sales situations by using Pre-Artifact representations. This review of current design knowledge reuse was validated by our industrial partners and formed the basis for conceptualization and development of adequate services.

RESPOND Bild

At the beginning of a new product development process it is a crucial target of the project development phase to investigate the customer's needs and to frame these into project goals. In the following planning phase people will draw upon this information to formulate certain requirements and to develop focused design solutions. These first conceptual steps in the design process are conducted by salespersons. In this manner, salespersons lay the groundwork for downstream solution development. However, our analysis of sales situations showed that access to the required design knowledge can be limited.

RESPOND's approach is to facilitate knowledge transfer through a formalized representation of design knowledge which can be reused by inference mechanisms to support design tasks (Eichhoff & Maass, 2011). This representation not only allows for describing uncertainties in design knowledge and integrating multiple information sources. It can also be distributively managed and used by different organizations. Probabilistic inference mechanisms on basis of Bayesian networks use this knowledge representation to generate an assessment of the current design situation and to offer intelligent information services for design support.

Currently, we are implementing this approach in two prototype applications which will be used for empirical evaluation. For being successful, the project relies on the support of a consortium formed by members of research and industry:

oxid-esales.png vitra.png universitaet-siegen.png universitaet-koeln.png
hfu.png

Students of different bachelor's and master's courses such as Computer Science, Computer Science in Media, Online Media, Advanced Computer Science and Business Information Systems are free to join the research team and may contribute their services to the project. This is in line with the aims of the subsidy program IngenieurNachwuchs 2009 (Informatik) and helps the students to enhance their scientific skills.

Contact

Group leader: Andreas Filler


Related publications

Eichhoff, J. R. & Maass, W.

Functional Design Space Representations for Lead Qualification Situations, Fifth International Conference on Design Computing And Cognition, 2012.
PDF (1.19 MB) - BibTeX

Eichhoff, J. R. & Maass, W.

Representation and Reuse of Design Knowledge: An Application for Sales Call Support, 15th Int. Conf. on Knowledge-Based and Intelligent Inf. & Eng. Systems (KES2011), Kaiserslautern, 2011, pp. 387–396.
PDF (379 KB) - BibTeX

Eichhoff, J. R. & Maass, W.

Distributed Imprecise Design Knowledge on the Semantic Web, 7th Int. Workshop on Uncertainty Reasoning for the Semantic Web (URSW 2011), 2011, pp. 101-104.
PDF (187 KB) - BibTeX