Solving MCDM problems: Process concepts

In this paper we try to answer the question: what is a good decision-making process? Specifically, we endeavour to answer two questions: (1) what are the features of a good process for making decisions when confronted by multiple criteria and (2) what should one expect from applying an MCDM solution method? Our description of the process rests on a structure comprising two mappings: alternatives to attributes and attributes to criteria. Solving an MCDM problem amounts to an objective investigation into the impact of alternatives on attributes and to a subjective evaluation of the decision maker's preference system. It is actually an art to model the problem, to distinguish between the objective and the subjective and treat them appropriately. In our view such a process will force the decision maker to understand his or her preferences and allow the set of alternatives to be expanded. We believe that the alternative chosen by such a process will indeed be the best (based on the decision maker's preferences) given the limited resources available to be invested in the process. In recent years many solution methods have been proposed to solve MCDM problems, most of them with little emphasis on decision-making methodology. We think that often this technical approach is the least important part of the decision-making process. MCDM solution methods should become less accommodating and more ‘confrontational’ by forcing decision makers to explicitly consider their preferences. We discuss upgrading solution methods in this context.