NEW DECISION MAKER MODEL FOR MULTIOBJECTIVE OPTIMIZATION INTERACTIVE METHODS

Decision Maker Model instead of human Decision Maker can be used for testing/comparing multiobjective optimization interactive methods. New Decision Maker Model (called ZuMo) was defined as two criteria multiobjective optimization problem which is solved by using multiobjective evolutionary algorithm NSGA-II Model can significantly reduce time and workload of testing/comparing experiment design and implementation and easy can be used for ad-hoc methods. Evolutionary algorithms are effective in parallel computing that can reduce experiment time also. The designed model was tested comparing STEM and GUESS interactive methods solving three 2D criteria and three 3D criteria testing problems. Different metrics are used: iteration count, stopping reason, general distance, error ratio, spacing and maximal Pareto error. The GUESS method was more effective in obtaining goal solution then STEM method. The new ZuMo model is universal and can be updated for testing/comparing different multiobjective optimization interactive methods and ad-hoc methods also. Model framework if necessary can be extended by defining additional criteria.