An Algorithm for Complex Multi-criterion Decision-making Problem Solution

Multi-objective optimization is a very competitive issue that emerges naturally in most real-world problems. Evolutionary multi-objective optimization (EMO) methodologies have been amply applied to find a representative set of trade-off optimal solutions — known as Pareto-optimal (PO) solution — in the past decade and beyond. This research contributes to the existing set of knowledge in the field, as we present combination of evolutionary algorithm R-NSGA-II and penalty boundary intersection (PBI) approach that allows to get a part of PO points instead of a single point at each iteration. Such procedures will provide the decision-makers with a powerful tool to gain more reliable results. The suggested model can be effectively used to solve various multi- and many-objective optimization problems, achieving excellent results. We also provide a comparative analysis with other existing solutions. The results emphasize the reached advantages of our solution, which ensures good convergence and diversity in the area of interest with sufficient computational time reduction.