An Improved Artificial Fish Swarm Algorithm and Its Application to Packing and Layout Problems

Packing and layout problems belong to NP-Complete problems theoretically and they occur extensively in many engineering fields in practice. Artificial fish swarm algorithm (AFSA) is a newly proposed promising swarm intelligent optimization algorithm. Therefore we try to apply this novel intelligent algorithm to solving packing and layout problems. But there still exist some defects of this algorithm itself, such as low convergence rate and precision, premature as well as poor ability of balancing exploitation and exploration. To overcome them, an improved parallel adaptive hybrid artificial fish swarm algorithm (PAHAFSA) is proposed. This algorithm divides the whole population into two subpopulations (groups) with the same size, and different adaptive strategies are applied to the two groups respectively to make one group focus on global search while the other on local search. The two subpopulations evolve independently and individual migrations are conducted regularly to achieve information communication, increase the population diversity and improve convergence rate of proposed algorithm. When the information on the bulletin board does not change for a certain times, the hybrid strategy based on simulated annealing method will be introduced to help the algorithm escape from local optima and accelerate convergence rate. An example of packing and layout problems shows that PAHAFSA is feasible and effective.