Introducing new uncertainty theories into decision-aid methods: application to expert assessment of natural risks in mountains and hydraulic dams
Résumé
Management and control of natural and technological risks is an important societal issue. This paper deals more specifically with natural hazards in mountains (snow avalanches, torrential floods) and technological risks induced by hydraulic dams. In those two contexts, it is essential to provide methods and tools to make risk management the most safe and efficient as possible. These tools are used either by stakeholders or infrastructures managers. Decision are often based on imperfect information (uncertain, imprecise, incomplete, conflicting) provided by multiple and heterogeneous sources (numerical models, expert assessments, G.I.S). Information imperfection is classically represented by probabilistic approaches. In our application cases, knowledge is often lacking (epistemic uncertainty) and expert assessment can remain vague and qualitative. In those cases, probabilities remain quite subjective and other frameworks can be more efficient to represent imperfections. Two application cases using possibility, belief function theories and multicriteria decision making are described.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...