A decomposition-based multiobjective evolutionary algorithm with weights updated adaptively

• A new trigger is developed to control when the weight should be updated. adaptive weighting method proposed. decomposition algorithm with weights updated adaptively The proposed applied solve various multiobjective optimization problems . Recently, decomposition-based evolutionary algorithms (DMEAs) have become more prevalent than other patterns (e.g., Pareto-based and indicator-based algorithms) for solving (MOPs). They utilize a scalarizing decompose an MOP into several subproblems based on provided, resulting in performances of being highly dependent uniformity between problem’s optimal Pareto front distribution specified weights. However, generation generally simplex lattice design, which suitable “regular” fronts (i.e., simplex-like fronts) but not “irregular” fronts. To improve efficiency this type algorithm, we develop DMEA (named DMEA-WUA) regarding Specifically,the DMEA-WUA introduces novel exploration versus exploitation model environmental selection.The process finds appropriate given problem four steps: generation, deletion, addition replacement. Exploitation means using these from step guide evolution population. Moreover, carried out stagnant; different existing periodically updating Experimental results show that our fronts, including those shapes.