In this paper, we deal with the problem of modelling preferences in multiple criteria evaluation situations. When the number of objects to be evaluated is small, then it is possible to make a detailed analysis of the decision-maker’s preferences to find out a “value” or a “score” for each object. For example, in the Analytic Hierarchy Process, preference analysis is based on pairwise comparisons. We consider the statistical analysis of pairwise comparisons, and show that several issues of measurement scale must be clearly understood, before one can reliably apply the methods in practice. Our approach is based on the use of regression analysis rather than the eigenvalue technique of the AHP, to find the value scores for alternatives.
For further details log on website :