On Sensitivity Analysis in the Analytic Hierarchy Process

The analytic hierarchy process (AHP) is a popular multiobjective tool of decision analysis. We describe several ways of enhancing this decision-making process through the use of sensitivity analysis, an extension to AHP which is relatively unstudied. Sensitivity analysis can be useful in eliminating alternatives, enhancing a group decision process, or in providing information as to the robustness of a decision. Concentrating at the first level of the decision hierarchy, we create a weight space which represents all possible combinations of weights for the first-level objectives. This weight space is then partitioned into subsets, and spatial information is generated from it. We use a small example to demonstrate our ideas.