A fuzzy approach to the Weighted Overlap Dominance model

TY - GEN

T1 - A fuzzy approach to the Weighted Overlap Dominance model

AU - Franco de los Rios, Camilo Andres

AU - Hougaard, Jens Leth

AU - Nielsen, Kurt

N1 - Proceedings of the 15th Conference of the Spanish Association for Artificial Intelligence (CAEPIA'13), Madrid, September 17-20, 2013

PY - 2013

Y1 - 2013

N2 - Decision support models are required to handle the various aspects of multi-criteria decision problems in order to help the individual understand its possible solutions. In this sense, such models have to be capable of aggregating and exploiting different types of measurements and evaluations in an interactive way, where input data can take the form of uniquely-graded or interval-valued information. Here we explore the Weighted Overlap Dominance (WOD) model from a fuzzy perspective and its outranking approach to decision support and multidimensional interval analysis. Firstly, imprecision measures are introduced for characterizing the type of uncertainty being expressed by intervals, examining at the same time how the WOD model handles both non-interval as well as interval data, and secondly, relevance degrees are proposed for obtaining a ranking over the alternatives. Hence, a complete methodology is presented for ordering and identifying the best alternatives under an interactive procedure that takes into account the natural imprecision and relevance of information.

AB - Decision support models are required to handle the various aspects of multi-criteria decision problems in order to help the individual understand its possible solutions. In this sense, such models have to be capable of aggregating and exploiting different types of measurements and evaluations in an interactive way, where input data can take the form of uniquely-graded or interval-valued information. Here we explore the Weighted Overlap Dominance (WOD) model from a fuzzy perspective and its outranking approach to decision support and multidimensional interval analysis. Firstly, imprecision measures are introduced for characterizing the type of uncertainty being expressed by intervals, examining at the same time how the WOD model handles both non-interval as well as interval data, and secondly, relevance degrees are proposed for obtaining a ranking over the alternatives. Hence, a complete methodology is presented for ordering and identifying the best alternatives under an interactive procedure that takes into account the natural imprecision and relevance of information.

M3 - Article in proceedings

SP - 1240

EP - 1249

BT - Multiconferencia CAEPIA'13

T2 - Multiconferencia CAEPIA'13

Y2 - 17 September 2013 through 20 September 2013

ER -