Academic Updates: Jing Wang from School of Software and Internet of Things Engineering Published his Findings on Soft Computing Journal

Recently, a paper titled “A decomposition-based multi-objective evolutionary algorithm using infinitesimal method” , completed by Associate Professor Jing Wang from the School of Software and Internet of Things Engineering ,was published on the SSCI Q1 Soft Computing journal ,volume 167,2024.

The paper introduces an improved decomposition-based multi-objective evolutionary algorithm (MOEA/D-DKS), which effectively addresses the issue of inappropriate initial weight vector distribution in traditional MOEA/D algorithms when dealing with multi-objective optimization problems with complex Pareto front structures.

By optimizing the characteristic information of the Pareto front and adjusting the distribution of weight vectors, the algorithm achieves greater uniformity and adaptability, thereby enhancing resource allocation efficiency and the algorithm's global search capability. The MOEA/D-DKS demonstrated superior performance in experiments on 28 standard test suites, including ZDT, DTLZ, and WFG, especially showing higher solution accuracy and stronger robustness in complex and irregular Pareto front problems. The Wilcoxon rank-sum test and Friedman test further proved the significant advantages of MOEA/D-DKS in various performance indicators, reflecting the algorithm's innovation and practicality in the field of multi-objective optimization.