Hierarchical outranking methods for multi-criteria decision aiding

Identificador:  TDX:1724

Handle: https://hdl.handle.net/20.500.11797/TDX1724

Autores:  Del Vasto Terrientes, Luis Miguel

Resumen:
Multi-Criteria Decision Aiding (MCDA) methods support complex decision making involving multiple and conflictive criteria. MCDA distinguishes two main approaches to deal with this type of problems: utility-based and outranking methods, each with its own strengths and weaknesses. Outranking methods are based on social choice models combined with Artificial Intelligence techniques (such as the management of categorical data or uncertainty). They are recognized as providing tools for a realistic assessment and comparison of a set of alternatives, based on the decision maker’s knowledge and needs. One of the main weaknesses of the outranking methods is the lack of consideration of hierarchies of criteria, which enables the decision maker to naturally organize the problem, distinguishing different levels of generality that model the implicit taxonomical relations between the criteria. In this thesis we focus on developing hierarchical outranking tools and their application to real-world case studies for ranking and sorting problems.