Understanding local ignorance and non-specificity within the DS/AHP method of multi-criteria decision making

DS/AHP is a nascent technique for multi-criteria decision-making, inspired by the Analytic Hierarchy Process and whose mathematical foundation is based on the Dempster-Shafer theory of evidence (DST). One aspect, a consequence of the utilisation of DST in DS/AHP is the allowance of a decision maker to make preference judgements on groups of decision alternatives (DAs) over different criteria. In this paper, developments are undertaken on DS/AHP with respect to the measurement of the level of local ignorance in the judgements made, together with the elucidation of an associated measure of non-specificity. Through the evaluation of the limits on these measures, subsequent index values are constructed. These index values are used to introduce the concept of a vagueness plot within DS/AHP. The vagueness plot is shown to aid in gauging the results of interactive changes in the preference judgements made. The index values and the vagueness plot presented enable the notion of the anchoring and adjustment aspect to decision-making in DS/AHP to be investigated.