Knowledge Graph of Design Rules for a Context-Aware Cognitive Design Assistant - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Knowledge Graph of Design Rules for a Context-Aware Cognitive Design Assistant

Résumé

[Context] The design of a system shall comply with many design rules that help industrial designers to create high quality design in an efficient way. Nowadays, design rules try to consider all product lifecycle’s phases leading to an ever-increasing growth. This context makes the management of design rules a difficult but essential task. This is why many research and industrial works try to automate this task [1, 3, 4]. [Problem] The processing of design rules, which are natural language sentences stored in unstructured documents, requires expert software. Moreover, existing tools interrupt the design workflow and slow down the design process. [Proposition] We propose a Context-Aware Cognitive Design Assistant (CACDA) to support designers who have to satisfy some design rules among “Big Data”. First, we describe the CACDA from the user’s perspective. Second, we detail the process for modelling unstructured design rules into a com-putable knowledge graph that will feed the cognitive design assistant. [Future Work] Once our knowledge graph of design rules will be operational, we will concentrate on its processing to retrieve, recommend, and verify design rules. Experiments will also help to determine pros and cons of the design assistant.
Fichier principal
Vignette du fichier
LCPI_PLM_2020_HUET.pdf (1.07 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03027946 , version 1 (27-11-2020)

Licence

Paternité

Identifiants

Citer

Armand Huet, Romain Pinquié, Philippe Veron, Frédéric Segonds, Victor Fau. Knowledge Graph of Design Rules for a Context-Aware Cognitive Design Assistant. 17th IFIP International Conference on Product Lifecycle Management (PLM), Jul 2020, Rapperswil, Switzerland. pp.334-344, ⟨10.1007/978-3-030-62807-9_27⟩. ⟨hal-03027946⟩
101 Consultations
105 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More