An estimation procedure to detect and remove unintentional judgemental bias (UJB) in the analytic hierarchy process methodology

This paper provides an estimation procedure to detect and to remove unintentional judgmental bias (UJB) in the analytical hierarchy process (AHP) methodology for multicriteria decision making (MCDM). UJB reflects a “cyclical hierarchy” because of the existence of a structural dependence of alternatives on criteria. This violates the principle of hierarchic composition embodied in the AHP method. The judgment data for comparing the different sources of risk are susceptible to the “framing effect” bias. The reference AHP and the supermatrix approach can be used to detect for the presence of UJB in the original AHP data. The estimation and removal of UJB can be achieved by an L2 norm estimation procedure using the Stein estimator. This involves use of the generalized inverse theory and methodology. The feedback data matrices provide a “dual” formulation for the original AHP hierarchy. The supermatrix approach is based on a stationary Markov system involving the feedback matrix and the original AHP matrix local priorities at each level of the hierarchy. Once the original AHP judgment data has been revised to remove the UJB influence, it can then be used to provide the correct AHP sensitivity analysis