A linguistic intelligent user guide for method selection in multi-objective decision support systems

Some multi-objective decision-making (MODM) methods are more effective than others for particular decision problems and/or particular decision makers. It is therefore necessary to provide a set of MODM methods in a multi-objective decision support system (MODSS) to support a wide range of problem solving. However, it is always difficult for decision makers to select the most suitable method for individual cases because MODM methods involve a deep knowledge of mathematics. To handle this difficulty, this study develops a MODM method selection guide supported by a fuzzy matching optimization method. In this paper, we first present the modelling process for the knowledge of characteristics of the main MODM methods. We then present related matching techniques between the characteristics of a real-world decision-making situation and a set of predefined situation descriptions (characteristics of a MODM method) where the elements of the two sets may be expressed by linguistic terms. Based on this process, a fuzzy matching optimization-based MODM method selection approach is proposed. The approach applies general fuzzy numbers, fuzzy distance, fuzzy multi-criteria decision-making concepts, and rule-based inference techniques to recommend the most suitable method from a MODM method-base. The approach is adopted in a linguistic intelligent user guide within a MODSS. Experiments have shown that the development of the linguistic intelligent user guide can increase the ability of the MODSS to support decision makers in arriving at a satisfactory solution in a most effective way.