A comparison of continuous genetic algorithm and particle swarm optimization in parameter estimation of Gompertz growth model

Genetic algorithm and Particle Swarm Optimization are heuristic optimization methods inspired by genetic principles and swarm behavior phenomena, respectively. Those two methods are initiated by random generation of initial populations (initial solutions), fitness evaluation of every solution, solution updating until a termination condition are met. It is well known that those two methods are not always converge to an optimal solution. Those methods sometimes converge to suboptimal solutions, solution near the optimal solution. In this paper, continuous genetic algorithm and particle swarm optimization were implemented to estimate parameters in the Gompertz growth model from rooster weight data cited from literature. Although the best results of the two models were not significantly differs, we found that the particle swarm optimization method was more robust than the continuous genetic algorithm. Hence, the particle swarm optimization method is more recommended than the continuous genetic algorithm.