New belief functions based methods for multicriteria decision making

Abstract : Any decision is closely linked to the quality and availability of information. Innovative methodologies are proposed to help decisions based on imperfect information provided by more or less reliable and conflicting sources in uncertain context. These methods combine new uncertainty theories and multicriteria decision making. The ER-MCDA methodology associates the Analytic Hierarchy Process (AHP), used as a conceptual analytical framework, fuzzy sets, possibility and belief function theories. DSmT-AHP replaces the initial AHP aggregation principles by a fusion process and introduces specific discounting factors able to make a difference between importance and reliability of criteria. Finally, COWA-ER is the newest method that allows multicriteria decision making under uncertainty. All these methods are fitting perfectly to the domain of expert assessment in the context of natural hazards in mountains. This paper proposes a synthesis of the principles of the methods developped and described in details in other specific papers.
Type de document :
Communication dans un congrès
Euro Working Group – Decision Support Systems Workshop, Nov 2011, Paris, France. 14 p., 2011
Liste complète des métadonnées

Littérature citée [22 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00655752
Contributeur : Import Ws Irstea <>
Soumis le : lundi 2 janvier 2012 - 11:11:02
Dernière modification le : jeudi 7 février 2019 - 17:10:23
Document(s) archivé(s) le : mardi 3 avril 2012 - 02:21:54

Fichier

gr2011-pub00034295.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00655752, version 1
  • IRSTEA : PUB00034295

Collections

Citation

J.M. Tacnet, J. Dezert. New belief functions based methods for multicriteria decision making. Euro Working Group – Decision Support Systems Workshop, Nov 2011, Paris, France. 14 p., 2011. 〈hal-00655752〉

Partager

Métriques

Consultations de la notice

191

Téléchargements de fichiers

192