Using data visualization for ranking alternatives with partial information and interactive tradeoff elicitation

Abstract

This paper puts forward the use of data visualization in a new method for solving multiple criteria decision-making problems for ranking of alternatives, based on Flexible and Interactive Tradeoff (FITradeoff) elicitation. This approach uses partial information about the decision maker’s preferences, based on a structured process for eliciting scale constants (or weights) within the scope of multi-attribute value theory. Different from most of the partial information methods present in the literature, our approach is based on the traditional Tradeoff, which is the most axiomatically founded procedure for elicitation of criteria weights. Pairwise dominance concept is incorporated into the mathematical model of FITradeoff in such way that, at each interaction with the decision maker, a partial—or complete—order of the alternatives can be achieved, based on a two-step algorithm proposed. The method is operated by means of a decision support system, which provides graphical visualization of the ranking at each interaction, in order to support the decision-making process with a simpler visualization of dominance relations. The proposed method is applied for solving a practical case of supplier selection in a food industry.