DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
(titres de publication, de périodique et noms de colloque inclus)
2015-07-15 - Colloque/Présentation - communication orale - Anglais - 1 page(s)

Rodrigues Teresa, BANA e Costa Carlos A., BANA e Costa João, De Corte Jean-Marie , Vansnick Jean-Claude , "M-MACBETH for Multicriteria Resource Allocation" in EURO 2015 - 27th European Conference on Operational Research, Glasgow, Ecosse, 2015

  • Codes CREF : Modèles mathématiques d'aide à la décision (DI1151)
  • Unités de recherche UMONS : Mathématique et analyse de la décision (W738)
  • Instituts UMONS : Institut de Recherche en Développement Humain et des Organisations (HumanOrg)

Abstract(s) :

(Anglais) The M-MACBETH DSS (www.m-macbeth.com) implements the MACBETH approach to evaluate projects on multiple criteria base only on qualitative pairwise comparison judgements about difference of attractiveness. This multicriteria decision aid tool supports the selection of a good/best project. However, in a context of scarce resources, choosing a portfolio of projects is a more demanding problem, as it requires not only to balance benefits against costs and the risks of realising the benefits, but also to evaluate several projects together. There are several DSS for multicriteria portfolio analysis, that differ on the resource allocation procedure used: prioritizing projects by decreasing values of benefit-to-cost ratios or identifying the optimal portfolio by mathematical programming. It is well-known that the portfolios arising from the approaches do not always coincide, therefore it would be useful to combine both approaches, but few DSS do so. Within this framework, a new resource allocation component of the M-MACBETH DSS was developed, which implements the two approaches interactively. One distinctive feature is the ability to explicitly address the baseline problem, by sensitivity analysis of the stability of priority ranking and of the optimal portfolio. Besides, it is possible to deal with other constraints than the budget limitation, such as to force the inclusion or exclusion of projects from the portfolio or to model the mutually exclusion between projects.