When conflict induces the expression of incomplete preferences

Abstract

a b s t r a c tMulticriteria conflict arises in pairwise comparisons, where each alternative outperforms the other one on some criterion, which imposes a trade-off. Comparing two alternatives can be difficult if their respective advantages are of high magnitude (the attribute spread is large). In this paper, we investigate to which extent conflict in a comparison situation can lead decision makers to express incomplete preferences, that is, to refuse to compare the two alternatives, or to be unable to compare them with confidence. We report on an experiment in which subjects expressed preferences on pairs of alternatives involving varying conflicts. Results show that depending on whether the participants are allowed to express incomplete preferences or not, attribute spread has a different effect: a large attribute spread increases the frequency of incomparability statements, when available, while it increases the use of indifference statements when only indifference and preference answers are permitted. These results lead us to derive some implications for preference elicitation methods involving comparison tasks.

A consolidated approach to the axiomatization of outranking relations: a survey and new results

“…Such a definition automatically yields a complete reflexive relation (in view of the properties of the co-dual operator that were recalled in Section 2). This is problematic since reflexive preference relations cannot always be assumed to be complete (see Deparis et al, 2012, for an experimental investigation of incomparability in preferences). In particular, in the context of outranking methods, pairs of alternatives may be incomparable, due for instance to veto effects (see Roy, 1996).…”

“…It would be interesting to test whether the resulting relations can be interpreted as a concordance & nondiscordance relation. Regarding the incompleteness issue, recent experiments (Deparis et al, 2012) show that preferences may be incomplete when the decision maker is granted the possibility not to choose. Our standpoint, in this work, is that concordance & non-discordance relations could actually be observed in such situations.…”

Managing incomplete preference relations in decision making: A review and future trends

“…Indeed, a study by Deparis et al [22] corroborates empirically the following hypothesis: "increasing the intensity of conflict in a multicriteria comparison increases the likelihood that DMs consider two alternatives as incomparable," and therefore leading to the expression of incomplete preferences. Their results indicate that a large attribute spread increases the frequency of incomparability statements when allowed, otherwise an increase of indifference statements happens.…”

“…The main advantage of preference relations, which are built by pairwise comparisons, is that of focusing exclusively on two options at a time, which facilitates experts when expressing their preferences. However, the drawback is that some experts might not been able to discriminate the degree up to which some of the options are better than others, and as a consequence incomplete preferences are provided [22].…”

Stakeholder interviews with two MAVT preference elicitation philosophies in a Swiss water infrastructure decision: Aggregation using SWING-weighting and disaggregation using UTAGMS

“…Perhaps a different design of these reference alternatives would lead to a different perception of the pairwise comparisons. For example, Deparis et al (2012) report that larger differences on each objective in the comparison of two alternatives increases the frequency of incomparability statements, when available. On the other hand, a larger magnitude of differences increases the use of indifference statements when only indifference and preference answers are permitted.…”