AI Summary & Key Takeaways
AI InsightThe result shows that HFO performs better than the ABC.
📌 3 Key Takeaways
Abstract
In this paper, a new optimization framework, namely Honey Formation Optimization (HFO), is proposed. In contrary to the Artificial Bee Colony Optimization (ABC) variants in literature, the HFO considers food sources consisting of many components and model the honey formation inside bees as a process of mixing the components with their special enzymes during chewing up the food source. We believe that bees analyze the amounts of components inside the food source and attempt more to collect weaker (less amount) components to improve the honey formation process. Thus, each time a worker exploits a food source it selects a component in such a way that weaker components are more frequently selected. The approach requires decomposing the solution into components where each component is evaluated by a component fitness function. The honey formula maps the component fitness to honey amount and considered as the equivalence of the fitness function. The worker bee uses the fitness of the selected component to evaluate the food source and does local search only around the selected component. The HFO and ABC Frameworks are compared on the basis of 9 benchmark functions. The result shows that HFO performs better than the ABC.