Decisionmaking Among Multiple-Attribute Alternatives

A description and evaluation of several existing methods designed to help decisionmakers deal with the multiple-decision problem. In all types of decision situations, the alternatives from which a choice must be made are characterized by multiple attributes (or properties). For example, a weapon system may be characterized by performance, cost, availability date, etc. The attributes of these characteristics may be considered at various levels of aggregation, i.e., performance may be considered relative to range, delivery, time, yield, vulnerability, and accuracy. As the number of relevant attributes and alternatives increases, the ability of the decisionmaker to handle the problem decreases, and the information-processing requirements may rapidly exceed the decisionmaker's processing capacity. Methods to deal with this problem include Dominance, Satisficing, Maximin, Minimax, Lexicography, Additive Weighting, Effectiveness Index, Utility Theory, Tradeoffs, and Nonmetric Scaling. Similarities and differences in the various approaches are demonstrated by a simplified weapon system-selection problem. It is concluded that using a combination of methods frequently may be more feasible than using any one method separately.