Call for Participation
Description
Topics
Attendance and Submission Information
Important Dates
Web Pages
Organizing Committee
Program Committee
Representing and solving configuration problems have always been subjects of interest for applying and developing AI techniques because powerful knowledge-representation models are necessary to capture the great variety and complexity of configurable product models, and efficient reasoning methods are required to provide intelligent interactive behavior in configurator software, such as solution search, satisfaction of user preferences, optimization, diagnosis, etc.
Today, the number, the diversity and the complexity of configurable products available on the market is growing, expanding from conventional equipment configuration to service configuration, such as loans, insurance, travel packages, and so forth. Configuration is more than ever a challenging area for applying novel AI techniques since more and more sophisticated reasoning tasks are delegated to the configurator software; the software must thus integrate product-assembly knowledge along with customer classification, adaptive sales strategies, and customer assistance. This integration becomes particularly critical for e-business applications where customers directly configure products through the Web with no human assistance and without a deep knowledge of the products they are buying.
The workshop gives AI researchers from different areas (e.g. description logic, constraint satisfaction, nonmonotonic reasoning, logic programs, case-based reasoning, ontologies, planning), real-application developers, and industrial configurator vendors the opportunity to exchange needs, ideas, work, methods, experiments, user-cases, and benchmarks related to the various problems met by configuration applications: theory, knowledge representation, reasoning, optimization, diagnosis, solution repair, interactivity, ontology, cooperative processes, etc. Additionally, the workshop hopes to attract case studies of applying configurators in organizations, including results from operation management researchers and others who may take a “black-box” perspective.
The workshop continues the series of Configuration workshops started at the AAAI 1996 Fall Symposium and continued at AAAI 1999, ECAI 2000, IJCAI 2001, ECAI 2002, and IJCAI 2003 and the special issues of AI_EDAM 2003, AI-EDAM 98 and IEEE Expert 98 journals on configuration. The four previous workshops were particularly successful since they had each more than 40 participants representing academia, end users (such as Siemens, HP or Daimler Benz), and the major configurator vendors (such as Oracle, SAP, or BAAN).
The goal of the workshop is to promote high quality research in configuration and to strengthen the interaction between industry and research. More generally, the workshop is intended for researchers and product developers interested in this area and in the application of AI techniques to real problems and the research fostered by it.
The workshop will be a one and half day event. Accepted submissions are organized as panels according to topics with short introductory presentations by panel members, allowing ample time for discussion to stimulate a workshop-like event.
The workshop wishes to attract submissions from many researchers and to spread the awareness of configuration problems among those working with the wide range of applicable AI technologies that complement each other. As an additional feature of the workshop, we propose consideration of new areas of application of constraint programming, particularly those which may drive development of the theory and/or technology. This could be approached via an invited lecture, or a round-table discussion. Suggestions are welcome.
We invite submissions describing novel and previously unpublished research (possibly in progress) or experiences with AI in configuration-related areas, including but not limited to:
[1] Configuration problems and models
Configuration problems typology. Knowledge representation & acquisition.
Fuzzy and incomplete knowledge. Knowledge base validation, diagnosis. Standardization
of catalog exchange format.
Configuration problem including discrete/continuous/mixed constraints.
Product and process configuration. Service configuration (travel , insurance...).
As well as: Product Design and Configuration.
[2] Reasoning methods
Constraint Satisfaction Problems and its extensions. Preference based
reasoning. Descriptive logic, Rules, Case-based reasoning. Local search,
Genetic algorithms, Neural networks. Problem decomposition. Optimization.
Multi-criteria optimization. Symmetry breaking. Cooperative configuration
processes. Re-configuration of existing systems. Explanations. Benchmark
proposals.
[3] Interactivity & e-business
Personalization. Ontology. Intelligent man machine interaction. Machine
learning. Client/Server architecture, Configuration server. Configuration
Web service. Distributed configurators.
[4] Integration with other modules
Product Data Management. CAD. Pricer. ERP. Product Configuration and
CRM Interfaces. Interaction with Product Data Management systems. Process
configuration.
[5] Applications & Tools
Application reports. Case studies. Designing and Configuration Tools.
Real world challenges.
We hope to attract a balance of
[ATTENDANCE AND SUBMISSION INFORMATION]
All workshop participants must register to the ECAI-2004 conference, which also handles the practical arrangements such as workshop registration, location etc. Workshop participation will be by invitation only, and will be limited to 40 participants.
If you wish to participate, submit either a full paper of no more than 6 pages (or 6000 words), or a position statement, a short paper, or a problem instance (at most 3 pages or 3000 words). Short papers may address an important problem for further research or describe a practical problem or an interesting lesson learned. In addition, we solicit proposals for short demonstrations (at most 3 pages or 3000 words, and software demonstrations taking at most 15 minutes), emphasizing the original contribution, functionality or conceptual foundation of the system.
If a short paper describes a problem instance, it can do this using natural language or any suitable formalism, in addition to giving details such as the source and domain (e.g. computer, car or telecommunications industry, possibly name of the company and product) of the problem, the possible system(s) used for solving it, performance data for the system(s), general characteristics such as the potential search space vs. the number of correct configurations or suitable configurations with respect to some requirements, and modeling method specific characteristics such as number of variables and the sizes of their domains for a CSP representation. However, the authors are strongly encouraged to formulate their descriptions based on the generally used concepts: components, attributes, ports and connections, resources, and functions (features). If necessary to preserve confidential information, you may rename the elements of the problem description to meaningless symbols to hide their origin and omit the names of the company and product. The problem instances should also be accompanied with a longer, possibly more detailed description file of the problem. The longer description should contain the same information as the shorter version in addition to giving a more detailed model and comments on the meaning and relation between different parts of model to make it accessible to other researchers willing to try their system on it.
The submissions should follow the ECAI-2004 style guide (soon available) and guidelines pertaining to blind reviewing. Electronic submission to Claire.Bagley@oracle.com in both postscript and PDF format is preferred, i.e. submit two distinct pieces, one containing the paper without author identification, the other containing the title page with author identification, and both in PS and PDF formats, altogether four files. No tracking numbers or declarations on submissions to other forums are needed.
Each submission is blind refereed by at least two members of the program committee. Refereeing criteria are relevance to workshop topics, significance and novelty of the research, technical content, discussion on relation to previous work and clarity of presentation. A contribution submitted as a long paper may be accepted as a short paper, if the program committee considers it to be inadequate for a long paper but to present an important issue.
At least one author of each accepted paper is required to attend the conference to present the paper.
Submission deadline: April 1st
2004
Notification of acceptance: May 7th
2004
Final paper due: May 24th 2004
Camera-ready version deadline: June 1st
2004
Camera-ready workshop notes due: June 15th
2004
Workshops at ECAI-2004: Aug 22nd-24th
2004
ECAI 2004: http://www.dsic.upv.es/ecai2004/index2.html
ECAI 2004 Configuration Workshop: http://www.ifi.uni-klu.ac.at/Conferences/ECAI04-Configuration-Workshop
Chair and contact person: Claire Bagley.
Oracle Corporation
10 Van de Graaff Drive, Burlington - MA 01803, USA
Tel: +1-781-744-0500
Fax: +1-781-744-0001
E-mail: Claire.Bagley@oracle.com
Prof. Michel Aldanondo.
Ecole des Mines d'Albi-Carmaux, France.
Prof. Eugene Freuder.
University College Cork, Ireland.
Associate Prof. Dietmar Jannach.
University Klagenfurt, Austria.
Daniel Mailharro.
ILOG S.A., France.
Prof. Timo Soininen.
Helsinki University of Technology, Finland.
Prof. Markus Stumptner.
University of South Australia, Australia.
Michel Aldanondo, Centre de Genie Industriel, Ecole des Mines d'Albi,
France.
Claire Bagley, Oracle Corporation, USA.
Boi Faltings, Swiss Federal Institute of Technology, Switzerland.
Alexander Felfernig, University Klagenfurt, Austria.
David Franke, Trilogy, USA.
Felix Frayman, Felix Frayman Consulting, USA.
Gerhard Friedrich, University Klagenfurt, Austria.
Esther Gelle, ABB Switzerland, Corporate Research, Switzerland.
Albert Haag, SAP, Germany.
Ulrich Junker, ILOG S.A., France.
Diego Magro, Universita di Torino, Italy.
Daniel Mailharro, ILOG S.A., France.
Barry O'Sullivan, University College Cork, Ireland.
Klas Orsvarn, Tacton System AB, Sweden.
Frank Piller, TUM Research Center Mass Customization & Customer
Integration, Germany.
Marty Plotkin, Oracle Corporation, USA.
Mihaela Sabin, Rivier College, USA.
Timo Soininen, Helsinki University of Technology, Finland.
Markus Stumptner, Advanced Computing Research Center, Australia.