Value Clarification: An Engineering Approach

In multi-attribute and multi-criterion situations, decision makers often appear to experience considerable difficulties when asked to express explicit preferences. The engineering approach suggests that it may be useful to sacrifice some rigor, to make client uncertainty about preferences explicit by encoding it in the form of random variables, and to concentrate on the problem of value clarification as opposed to value explication. The reasonable tolerance of value uncertainty, and the amount of effort which may be devoted to value clarification are examined. The problem of finding methods with which to assist clients in becoming clearer about their preferences is discussed.