Several approaches have been proposed to derive the weights for the Analytic Hierarchy Process (AHP). These methods can be used for any problem in which we need to determine weights for an aggregation operator and we interview an expert to obtain them. For example, they can be used for weighted means, ordered weighted averaging (OWA) operators, and quasi-arithmetic means.

Most of these methods interview an expert to elicit a preference matrix, and use this matrix to obtain the vector.

In this paper we consider the problem in which each of the cells in the preference matrix does not contain a single value but a set of them. We propose a method following the logarithmic least square approach.