الفهرس 
Computational optimization
Trends for future workOnly 14 pages are availabe for public view
 |  | from | 223 |  |  |
| | | from | 223 | | |
Abstract
A single-objective, multivariate, unconstrained, continuous optimization problem is investigated. Instead of imitating the behavior of biological organisms like birds, fish, ants, and bees, as in previously published procedures, our approach adopts as metaphor the social behavior of human beings in tackling life problems through counseling within a group. Counseling is a well-known discipline in psychology and sociology. This is the first time a link is established between group counseling, in particular, and population-based computational optimization. We identify twenty striking items of significant analogy. Based on these metaphoric items, a novel optimization algorithm is developed. We call it a Group Counseling Optimizer and abbreviate it as GCO. The GCO algorithmic iterations are visualized as counseling sessions, with counselees and counselors. Solutions are progressively improved over such’ iterations (sessions) by means of two strategies: other-members counseling and self-counseling. The algorithm is tested in minimization problems using several unrotated and rotated benchmark functions. The global minimum of all functions is reached without being trapped in a local minimum. To demonstrate its efficacy, the GCO is applied to specific models of two spacecraft missions, Cassinil and GTOC1. The objective function is to be minimized in the first mission and maximized in the second. The best known values of the objective function in the two cases are obtained with acceptably small error. In this respect, the GCO is compared with eight other optimizers: Particle Swarm Optimizer (PSO), Genetic Algorithm, Simulated Annealing.