Interactive preference elicitation incorporating a priori and a posteriori methods

In a typical decision-making process, preference elicitation methods require a priori knowledge about the desired outcomes. It is expected that the decision maker (DM) has a rough idea of the alternatives or the production possibilities, and has a relatively clear idea of his/her preferences regarding the decision. For those DMs who do not have this general understanding, the determination of preferences (and/or targets for specific criteria) can be difficult, uncertain and may lead towards a suboptimal solution for the particular DM. In a typical planning process, a limited set of alternatives are generated by a professional. These are then evaluated using preference information. A more useful approach would be to use that preference information towards developing a more acceptable alternative. The method proposed in this article is an interactive system which first requires the DM to indicate their initial preference value for each criterion. Then a series of pre-determined goal programming functions generate a solution. The solutions are then compared simultaneously, so that the DM can understand how feasible the preferences are when used in the creation of alternative solutions. The DM is then asked to provide a minimum preference level guided by a comparison of the alternative solutions. By proceeding in an iterative fashion, the DM can adjust his/her preferences until the DM is satisfied with one of the generated solutions. A real forest planning situation is described as a case study of this method. Copyright Springer Science+Business Media New York 2015