Skip to Main content Skip to Navigation
Book sections

PROFIL: A decision support tool for metallic sections design using a CBR approach

Abstract : This paper presents the PROFIL system, a Decision Support Tool for the Design of metallic sections production. The PROFIL system uses a Case-Based Reasoning approach for the selection of pertinent former designs. Production of metallic sections is done by a succession of forming operations. The design task is composed of two parts: selection and scheduling of operations and assessment of forming tools geometry. The PROFIL system also provides support to production technicians, to input process parameters and control measurements as part of cases, and to adjust settings during preproduction trials. Selection of cases is made by following two orthogonal points of view, on similarity and quality of produced parts. Similarity is based on selection of relevant dimensions of the section, while quality is based on selection of potential defects. Each dimension or defect is associated to an evaluation function which uses a fuzzy logic approach. To improve ergonomy, relevant cases are displayed in a two-dimension space: similitude/quality. Implementation of the system has been done on a computer workstation, using a commercial DBMS, in order to cope with industrial requirements. The PROFIL system is operational in the industrial site and provides support to designers and process technicians.
Document type :
Book sections
Complete list of metadata
Contributor : Magalie Prudon Connect in order to contact the contributor
Submitted on : Monday, December 21, 2015 - 11:28:55 AM
Last modification on : Wednesday, November 17, 2021 - 12:28:19 PM

Links full text



Jean-Luc Wybo, Frédéric Geffraye, Aline Russeil. PROFIL: A decision support tool for metallic sections design using a CBR approach. Manuela Veloso, Agnar Aamodt Case-Based Reasoning Research and Development. First International Conference, ICCBR-95 Sesimbra, Portugal, October 23–26, 1995 Proceedings, 1010, pp.33-42, 1995, Lecture Notes in Computer Science, 978-3-540-60598-0. ⟨10.1007/3-540-60598-3_4⟩. ⟨hal-01247088⟩



Record views