| Abstract [eng] |
The purpose of this paper was to create an improved bee population computer model. After researching current bee models, two current models (BEEHAVE and Bumble-BEEHAVE) were combined. This joint model was expanded by adding a swarming process and an additional bee drifting process implementation was described. An improved quick swarming artificial bee colony algorithm was introduced and compared to currently existing artificial bee colony algorithms. 25 different goal functions for fitness evaluation and Wilcoxon signed-rank test was used to determine any statistical significance between results. Another variation of said algorithm was adopted and used in the new improved bee model to implement a swarming process. |
