(An) enhanced multiple criteria decision support system with graphic user interface and knowledge-based approach

Abstract

Decision making is the process of determining a possible course of action under the given circumstances. In almost of such problems, the multiplicity of criteria for evaluating the possible alternative actions is pervasive. And, this increasing recognition of the decision maker``s conflicting portfolio with multiple criteria has resulted much research on Multiple criteria Decision Making (MCDM) over the past two decades. MCDM has the ultimate goal of helping a decision maker (DM) to find the most preferred solution with given circumstances. In the 1970``s, research focused on the theoretical foundations of multiple objective mathematical programming. However, in the 1980``s, the emphasis has clearly shifted toward multiple criteria decision support. This shift reflects the fact that the decision maker``s involvement with the solution process is not only mandatory but also of focus in MCDM. In one phrase, the primary aim of this research is ``to help a DM to find the properly satisfied solution, by minimizing his cognitive bias and burden and by providing integrated decision support which is usable even for the naive users``. Since the DM``s preference structure is initially often vague and not exactly known, correct guidance of a DM through proper decision support is of most importance. Stepwise contouring of efficient criterion vectors for large scale MOLP is introduced to present a DM with the insight over the solution set. The iterative procedures of presenting rough map to DM and determining more detailed structure on certain subset guided by DM can be a good decision aid in MCDM. It could reduce the DM``s cognitive bias over the solution set and stepwise focused contouring could reduce the computational complexity and enhance the convergency to optimum without much burden of the DM. And, this stepwise focused contouring has been combined with the graphic-aided interactive method, so called the visual interactive method. To contract the criterion space w...