Extending the multi-criteria decision making method DEX with numeric attributes, value distributions and relational models

摘要: DEX is a qualitative multi-criteria decision analysis method. The method supports makers in making complex decisions based on multiple, possibly conflicting, attributes. attributes have value scales and are structured hierarchically. hierarchical topology allows for decomposition of the problem into simpler sub-problems. In DEX, alternatives described with values, taken from corresponding input hierarchy. evaluation performed bottom-up way, utilizing aggregation functions, which defined every aggregated attribute form rules. has been used numerous practical applications—from everyday problems to solving financial ecological domains. Based experience, we identified need three major methodological extensions DEX: introducing numeric attributes, probabilistic fuzzy values relational models. These were proposed by users existing new demands problems, require advanced approaches. this paper, introduce these describing formally, justifying their contributions process illustrating them didactic example, followed throughout paper.