Multiobjective and Multicriteria Decision Processes and Methods

An appropriate decision-making process is relevant for the strategic success of any organization. Most of the decision problems in these organizations have multiple objectives that have to be dealt with simultaneously. This chapter gives a brief description of a few multicriteria (MCDM; MCDA; MCDM/A) methods, including deterministic additive methods (MAVT), Multi-Attribute Utility Theory (MAUT), connected with Decision Theory, and outranking methods (ELECTRE and PROMETHEE). Additionally, Group Decision and Negotiation process is considered. A procedure for building an MCDM/A decision model is presented, which enables several factors to be incorporated such as: the DM’s preference structure and experts’ prior knowledge regarding the state of nature. The choice of the method is considered. Some concerns related to choosing an appropriate MCDM/A method are presented, including preference modeling with the evaluation of the DM’s compensatory and non-compensatory rationality. This procedure enables an MCDM/A problem to be solved. Several issues concerning the implementation of this procedure are presented, such as: setting scales and normalizing criteria, time management in the scheduling of the decision process (including the procrastination process), and incorporating the intelligence stage of Simon’s model into the procedure. Although this procedure may be applied to any context, some particular considerations are given to those of Risk, Reliability and Maintenance. For instance, a multidimensional risk analysis allows a broader view and may include the DM’s behavior regarding risk (prone, neutral or averse). In the reliability and maintenance contexts, the models may include availability, maintainability, dependability, quality of repair and other aspects besides cost.