The decomposition-based multi-objective evolutionary algorithm (MOEA/D) transforms a optimization problem (MOP) into set of single-objective subproblems for collaborative optimization. Mismatches between and solutions can lead to severe performance degradation MOEA/D. Most existing mismatch coping strategies only work when the L∞ scalarization is used. A strategy that use any Lp scalarization, ...

In this paper, we propose a new combined scalarization method of multi-objective optimization problems by using the surplus variables and generalized Tchebycheff norm then use it to obtain some equivalent characterizations (weakly, strictly, properly) efficient solutions adjusting range parameters. These results do not need any convexity assumption conditions object...