Towards a knowledge based framework for numerical design of experiment optimization and management. Issue 6 (1st November 2016)

• Record Type: Journal Article
• Authors: Blondet, Gaëtan; Belkadi, Farouk; Le Duigou, Julien; Bernard, Alain; Boudaoud, Nassim
• Is Part Of: Computer-aided design and applications. Volume 13:Issue 6(2016)
• Page Start: 872
• Page End: 884
• Publication Date: 2016-11-01
• Subjects: Knowledge management -- adaptive design of experiment -- simulation data management -- knowledge-based system; Computer-aided design -- Congresses; Computer-aided design -- Periodicals; Engineering design -- Data processing -- Congresses; Engineering design -- Periodicals; 620.00420285
• DOI: 10.1080/16864360.2016.1168236
• Languages: English
• ISSNs: 1686-4360

Abstract

ABSTRACT: Numerical Design of Experiment (DoE) is a powerful tool for product development, used to improve product quality and robustness. However, the simulation process can be highly extended by the DoE process. While methods have been developed to shorten the execution of numerical DoE, the time needed to set up the numerical DoE process is longer and longer. This paper presents a description of the objectives and first results of the SDM4DOE project (Simulation Data Management for Design of Experiment). This project aims to define a set of tools and methods to improve the simulation process involving DoE: data management, data robustness improvement and process shortening. A knowledge-based approach is proposed to solve this main issue, based on a specific knowledge representation. GRAPHICAL ABSTRACT: