An Electromagnetism-Like Meta-Heuristic for Multi-Objective Optimization

This paper presents an algorithm based on the electromagnetism-like mechanism (EM) to solve multi-objective optimization problems (MOP). EM is based on the electromagnetism theory in physics; it simulates attraction and repulsion of charged particles in order to move towards an optimal solution. A framework, so called multi-objective EM (MOEM), for generating the non-dominated front of MOP has been developed heuristically. Computational results show that the non-dominated solutions generated by the MOEM are closer to the Pareto front than those by another well-known multi-objective evolutionary algorithm, the strength Pareto evolutionary algorithm (SPEA). For the distribution and spread of non-dominated solutions, MOEM is also competitive with SPEA. Finally, a number of directions for extension and application of the algorithm is discussed.