Fuzzy engineering design semantics elaboration and application

• Published in: Soft computing letters 2021-12, Vol.3, p.100025, Article 100025
• Main Authors: Fougères, Alain-Jérôme, Ostrosi, Egon
• Format: Article
• Language: English
• Subjects: Fuzzy collaborative design; Fuzzy collaborative system; Fuzzy engineering design platform; Fuzzy requirement engineering; Natural language processing
• ISSN: 2666-2221; 2666-2221
• DOI: 10.1016/j.socl.2021.100025

•Five principles are used to set out a computing procedure for fuzzy intelligent requirements engineering from natural language to Computer-Aided Design.•The isomorphism between fuzzy conceptual graphs, and fuzzy predicate logic enables a fuzzy formal language of requirement to be produced.•From Z language the requirements are chained into Computer Aided Three-Dimensional Interactive Application (CATIA) models.•The formalisation increases requirements’ reliability and relevance in both CAD models and PLM systems more generally. Product design activities are predicated on fuzzy modelling, given that verbalising and interpreting engineering requirements are inherently fuzzy processes. The aim of this paper is to present a method for fuzzy intelligent requirement engineering from natural language to Computer-Aided Design (CAD) models. The field exploring the dynamics of computational processes from fuzzy linguistic modelling to fuzzy design modelling is complex and remains under-explored. No existing research has been identified which focuses specifically on fuzzy requirements engineering from natural language to CAD modelling. This paper seeks to address this by providing a design formalisation system based on five key principles. These principles are used to set out a computing procedure which follows a method broken up into six phases. The results of these six phases are fuzzy semantic graphs, which provide engineering requirements according to reliable design information. The approach is put into practice using the fuzzy agent-based tool developed by the authors, called F-EGEON (Fuzzy Engineering desiGn sEmantics elabOration and applicatioN). The proposed method is illustrated through an application from the automotive industry.