.png){kind=icon}
.png){kind=icon}
Return
A shift density and θ-dominance based fitness evaluation mechanism for large-scale many-objective optimization
.png){kind=icon}
PRE
AI
C
A
DOI：10.1007/s12293-026-00500-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The fitness evaluation mechanisms (FEMs) methods used in the multi and many-objective optimization plays a crucial role in generating the well converged and diverse approximation of the Pareto front. Recently, various FEMs have been proposed and incorporated in the traditional multi and many-objective optimization algorithms to address the different forms of multi and many-objective optimization problems. Even there have been developed various effective FEMs, still these algorithms often face serious scalability issues when applied to large-scale many-objective optimization problems (LSMaOPs). In this paper a many-objective optimization algorithm (MaOA) based on new theta-Dominance and Shift Density FEM, i.e., theta shift density evolutionary algorithm theta-SDEA for LSMaOPs is proposed. The proposed theta-SDEA aims to enhance the performance of NSGA-III by exploiting the (theta-Dominance and Shift Density FEM). To validate the performance of the proposed theta-SDEA, it is tested on several LSMaOPs benchmark problems with 3-10 objectives and 100-500 decision variables. The obtained results compared with some existing approaches. The comparative study demonstrates that the proposed theta-SDEA approach consistently highlight the effectiveness and superiority over the existing approaches in solving the LSMaOPs. Furthermore, we address multi-objective knapsack problems using theta-SDEA.
Keywords:
Large-Scale Many-Objective Optimization
Shift Density estimation
Large-Scale Many-Objective Optimization
Shift Density estimation
theta-dominance
Journal
IF:
2.3
Papers:
447
Citations:
718
Organization
Cited Papers