Stochastic group preference modelling in the multiplicative AHP

This paper proposes a model for the representation of a consensus-seeking group decision making process under the multiplicative variant of the analytic hierarchy process (AHP). In particular, it proposes a group preference model which expresses the group's preference intensity judgements as random variables with associated probability distributions, and the structure of the multiplicative AHP is exploited to derive interval judgements for the alternatives' final impact scores as perceived by the group. These interval judgements can be used to determine the probability that a rank reversal may occur amongst alternatives due to the group preference uncertainty, i.e. to assess the stability of the final impact score vector, which in turn offers information about the degree of consensus within the group about the decision.