Performance Comparison of Genetic Algorithm and Particle Swarm Optimization in Solving Product Storage Optimization

Product storage provides considerable influence in obtaining profit for traders in selling products. However, the existing product storage is inefficient because products with high selling prices are stored in large quantities even though this does not necessarily give a high profit because it can also provide high losses. Traders must be able to determine the number of products stored at high selling prices and smaller losses. We propose Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) as an optimization method. We use GA with real code representation, one cut point crossover, insertion mutation and elitism selection. We also use PSO to solve the same problem. Both GA and PSO have been proved that can solve optimization problem. We compare which performance of them is better based on profit gained, fitness value and computational time. The experiment result shows that with the same problem and data set, PSO is better in gaining profit than GA but it needs longer computational time than GA.