Linking objective and subjective modeling in engineering design through arc-elastic dominance - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Expert Systems with Applications Année : 2012

Linking objective and subjective modeling in engineering design through arc-elastic dominance

Résumé

Engineering design in mechanics is a complex activity taking into account both objective modeling processes derived from physical analysis and designers' subjective reasoning. This paper introduces arc-elastic dominance as a suitable concept for ranking design solutions according to a combination of objective and subjective models. Objective models lead to the aggregation of information derived from physics, economics or eco-environmental analysis into a performance indicator. Subjective models result in a confidence indicator for the solutions' feasibility. Arc-elastic dominant design solutions achieve an optimal compromise between gain in performance and degradation in confidence. Due to the definition of arc-elasticity, this compromise value is expressive and easy for designers to interpret despite the difference in the nature of the objective and subjective models. From the investigation of arc-elasticity mathematical properties, a filtering algorithm of Pareto-efficient solutions is proposed and illustrated through a design knowledge modeling framework. This framework notably takes into account Harrington's desirability functions and Derringer's aggregation method. It is carried out through the re-design of a geothermal air conditioning system.
Fichier principal
Vignette du fichier
IMC_ESA_Pailhes_2012.pdf (1.47 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01020737 , version 1 (08-07-2014)

Identifiants

Citer

Patrick Sebastian, Yann Ledoux, Arnaud Collignan, Jérôme Pailhès. Linking objective and subjective modeling in engineering design through arc-elastic dominance. Expert Systems with Applications, 2012, 39 (9), pp.7743-7756. ⟨10.1016/j.eswa.2012.01.079⟩. ⟨hal-01020737⟩
165 Consultations
155 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More