Hierarchical majority-rule sorting models for temporal multi-criteria decision aiding

19 Feb 2019
multi-criteria decision aiding, sorting, outranking relation, mixed-integer programming, times series


In certain applications, evaluations of decision alternatives on multiple criteria may vary in order to reflect its performance, for instance, at various time steps in the past and/or in the future. Taking all of these factors into account at once may render the problem too difficult from the perspective of the decision-maker (DM). For this reason, we propose to divide the decision problem into a hierarchy of sub-problems. We also consider the case where a majority-rule sorting model, or MR-Sort, has been selected to model the preferences of the DM. The motivation behind this work is mainly based on a practical context linked to reacting to cyber-attacks of naval systems. Following a cyber-attack, a DM (the ship captain, for example) may wish to take an action in order to restore some or all of the ship’s functions. The time when these functions will be restored may also play a role based on the mission requirements at each time step. To our knowledge, multi-criteria decision aiding (MCDA) has only been poorly applied to the cyber-defence context (see, e.g. [2]).


Arthur Valko, Patrick Meyer, Alexandru Liviu Olteanu. Hierarchical majority-rule sorting models for temporal multi-criteria decision aiding. ROADEF 2019 : 20ème congrès annuel de la société Française de Recherche Opérationnelle et d’Aide à la Décision, Feb 2019, Le Havre, France. ⟨hal-02146688⟩

Lien vers l’article en PDF