This paper proposes a parallel particle swarm optimization (PPSO) to solve the multiobjective optimization problems (MOP). PPSO makes the use of the parallel characteristic of the PSO algorithm to deal with the multiple objectives issue of the MOP. PPSO uses as many swarms as the number of the objectives in the MOP and lets each swarm optimize only one of the objectives. These swarms work in parallel and each swarm can use a standard PSO or any other improved PSO variants to solve a single objective problem. PPSO has advantages on the following two aspects. First, as each swarm focus on optimizing only one objective, PPSO can avoid the difficulty of fitness assignment because the particles can be evaluated like in the single objective optimization problem. Second, as different swarms optimize different objectives, PPSO can maintain the population diversity to make a throughout search along the whole Pareto front to obtain nondominated solutions as many as possible. The performance of PPSO is tested on a set of benchmark problems complicated Pareto sets in CEC2009. The experimental results compared with those obtained by the state-of-the-art algorithms demonstrate the effectiveness and efficiency of PPSO, showing the good performance of PPSO in solving the MOP with complicated Pareto sets.
[1]
Ponnuthurai Nagaratnam Suganthan,et al.
Comprehensive learning particle swarm optimizer for solving multiobjective optimization problems: Research Articles
,
2006
.
[2]
Jing J. Liang,et al.
Comprehensive learning particle swarm optimizer for solving multiobjective optimization problems
,
2006,
Int. J. Intell. Syst..
[3]
Jun Zhang,et al.
Orthogonal Learning Particle Swarm Optimization
,
2009,
IEEE Transactions on Evolutionary Computation.
[4]
DebK.,et al.
A fast and elitist multiobjective genetic algorithm
,
2002
.
[5]
Kalyanmoy Deb,et al.
A fast and elitist multiobjective genetic algorithm: NSGA-II
,
2002,
IEEE Trans. Evol. Comput..
[6]
Qingfu Zhang,et al.
Multiobjective optimization Test Instances for the CEC 2009 Special Session and Competition
,
2009
.
[7]
Jun Zhang,et al.
Adaptive Particle Swarm Optimization
,
2008,
ANTS Conference.