Engineering with Computers | 2019
A new VPS-based algorithm for multi-objective optimization problems
Abstract
In this paper, a multi-objective variant of the vibrating particles system (MOVPS) is introduced. The new algorithm uses an external archive to keep the non-dominated solutions. Besides, the maximin strategy and crowding distance concept are employed for the selection of candidates to achieve the diversity and convergence of the solutions in the objective function space. MOVPS is evaluated using seven mathematical optimization problems and three structural design problems with continuous and discrete variables. The results are compared with competitive multi-objective optimization algorithms and show that the MOVPS can effectively find acceptable approximations of Pareto fronts for the multi-objective optimization problems.