Weighted sum model with partial preference information: application to Multi-Objective Optimization - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue European Journal of Operational Research Année : 2017

Weighted sum model with partial preference information: application to Multi-Objective Optimization

Résumé

Multi-objective optimization problems often lead to large nondominated sets, as the size of the problem or the number of objectives increases. Generating the whole nondominated set requires significant computation time, while most of the corresponding solutions are irrelevant to the decision maker (DM). Optimizing an aggregation function reduces the computation time and produces one or a very limited number of more focused solutions. This requires, however, the elicitation of precise preference parameters, which is often difficult and partly arbitrary, and might discard solutions of interest. An intermediate approach consists in using partial preference information with an aggregation function. In this work, we present a preference relation based on the weighted sum aggregation, where weights are not precisely defined. We give some properties of this preference relation and define the set of preferred points as the set of nondominated points with respect to this relation. We provide an efficient and generic way of generating this preferred set using any standard multi-objective optimization algorithm. This approach shows competitive performances both on computation time and quality of the generated preferred set.
Fichier non déposé

Dates et versions

hal-01494263 , version 1 (23-03-2017)

Identifiants

Citer

Sami Kaddani, Daniel Vanderpooten, Jean-Michel Vanpeperstraete, Hassan Aissi. Weighted sum model with partial preference information: application to Multi-Objective Optimization. European Journal of Operational Research, 2017, 260 (2), ⟨10.1016/j.ejor.2017.01.003⟩. ⟨hal-01494263⟩
101 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More