Ordinal data in multicriteria decision making. A stochastic dominance approach to siting nuclear power plants

This paper addresses decision problems with ordinal data (weights, criterion scores). A random sampling approach is proposed to generate quantitative values which are consistent with the underlying ordinal information. An attractive feature of the approach is that it is applicable with mixed (quantitative/ordinal) data. Another feature is that the approach can be extended to rankings with degrees of difference. The outcome of the approach is a distribution of performance scores. Stochastic dominance concepts are proposed to arrive at a final ranking of alternatives. An application of these procedures is given for a location study of nuclear power plants.