Dynamic optimisation of chemical engineering processes using the bees algorithm

An improved version of the Bees Algorithm is proposed for solving dynamic optimisation problems. This new formulation of the Bees Algorithm includes new search operators, and a new selection procedure that enhances the survival probability of newly formed individuals. The proposed algorithm is tested on six benchmark dynamic optimisation problems. The benchmark problems include minimisation and maximisation tasks of different levels of complexity. For all the problems, the Bees Algorithm finds very satisfactory optima. The very small standard deviation of the results proves the high reliability of the proposed technique. Experimental tests show that the Bees Algorithm outperforms the state-of-the-art Ant Colony Optimisation algorithm. The Bees Algorithm improves also the best results published in the literature for the six benchmark problems.