DSS for Multicriteria Preference Modeling with Partial Information and Its Modulation with Behavioral Studies

This paper discusses the trends for building DSS (Decision Support Systems) for Multicriteria Preference Modeling by using partial information to be obtained from DMs (Decision-makers). Also, it discusses the use of results from behavioral studies, including those that take a Decision Neuroscience approach, in order to modulate changes in the decision process and in the design of a DSS. The preference modeling is considered from two different perspectives: elicitation by decomposition and elicitation by holistic evaluations. This chapter focuses on a DSS that deals with Multicriteria Preference Modeling in the scope of MAVT (Multiattribute Value Theory) and describes the evolution of these DSSs in recent years. Finally, the trends in the decision aiding process using this kind of DSS for Preference Modeling with partial information is illustrated with the DSS for the FITradeoff method. The trends in the flexibility of this DSS is one of the features explored. It is shown how to combine two different paradigms for preference modeling: decomposition and holistic evaluations. Also, this chapter demonstrates how results from neuroscience experiments can be used to prompt the analyst to have insights when talking with and advising decision-makers (DMs) and how to improve the design of the DSS, both for the choice and the ranking problematic.