Dominance Measuring Approach using Stochastic Weights

Abstract

In this paper we propose an approach to obtain a ranking of alternatives in multicriteria decision-making problems when there is imprecision concerning the alternative performances, component utility functions and weights. We assume decision maker's preferences are represented by an additive multi-attribute utility function, in which weights are modeled by independent normal variables, the performance in each attribute for each alternative is an interval value and classes of utility functions are available for each attribute. The approach we propose is based on dominance measures, which are computed in a similar way that when the imprecision concerning weights is modeled by uniform distributions or by an ordinal relation. In this paper we will show how the approach can be applied when the imprecision concerning weights are represented by normal distributions. Extensions to other distributions, such as truncated normal or beta, can be feasible using Monte Carlo simulation techniques.