On Surrogate Weighting Methods: Some Missing Comparison Analyses

Surrogate weighting methods (SWMs) are weighting methods based on ordinal information, in which a weight vector is selected from a set of admissible weights to represent the set, and then used to evaluate the alternatives on the basis of multi-attribute utility/value theory. In this paper, a missing comparison analysis of novel SWMs against those having traditionally been identified as performing better is carried out on the basis of Monte Carlo simulation techniques. For this, several scenarios accounting for a different number of criteria and alternatives and probability distributions are considered, whereas the hit ratio, the rank-order correlation, the value loss and the proportion of maximum value range achieved are used as quality measures.