2193. Assessing the ranking quality of the PROMETHEE II method
Invited abstract in session MD-8: Methodological developments in PROMETHEE methods, stream Multiple Criteria Decision Aiding.
Monday, 14:30-16:00Room: Clarendon SR 2.08
Authors (first author is the speaker)
| 1. | Boris Coquelet
|
| Service Mathématiques de la Gestion, Université libre de Bruxelles | |
| 2. | Gilles Dejaegere
|
| Ecole Polytechnique, Université libre de Bruxelles | |
| 3. | Yves De Smet
|
| SMG - CoDE, Université libre de Bruxelles |
Abstract
PROMETHEE methods belong to the family of multicriteria outranking methods. They have been applied for the last forty years in many applications including finance, environmental management, health care, etc. Their successful application is based on their simplicity and the existence of user-friendly software such as PROMCAL, Decision Lab 2000, Visual PROMETHEE or D-SIGHT. These methods are based on the computation flow scores to generate partial or complete rankings of the alternatives.
Researchers have recently investigated the net flow scores' properties to characterise or justify the obtained rankings. In addition, a special focus has been given to the influence of third alternatives in the case of PROMETHEE II.
In this contribution, following the idea of the consistency ratio developed in the Analytical Hierarchy Process, we propose an indicator to evaluate the quality of the PROMETHEE II algorithm's ranking. This will be derived from an objective function proven to be optimised while computing net flow scores. The indicator will provide a tool for the decision-maker to assess his confidence in the algorithm's ranking. This is illustrated on real data sets.
Keywords
- Multi-Objective Decision Making
- Reliability
Status: accepted
Back to the list of papers