The Artificial Bee Colony Algorithm And Its Applied Research In Combinatorial Optimization

Swarm Intelligence Algorithm has become the focus of research in the optimization field, which based on gregarious creature. Marco Dorigo, the Italian scholars, proposed Ant Colony Optimization(ACO) algorithm in 1991. In 1995, Kennedy and Russell Eberhar proposed Particle Swarm Optimization(PSO) Algorithm, by simulating bird flocking and fish schooling. In 2005, D. Karaboga proposed a systematic Artificial Bee Colony Algorithm(ABC), which successfully applied the behavior of foraging to function optimization problem. Above algorithms had admitted rapidly in the international optimization field, and widely used in production and life, because they are simple, convenient, robust and strong.Artificial Bee Colony Algorithm is a new heuristic bionic algorithm and also a typical kind of swarm intelligence algorithm, based on foraging behavior of honeybee swarms. In the algorithm, each bee takes as an agent, they collaborate with each other to achieve the effect of swarm intelligence by the division of work, the role conversion and memory.This paper analyzes and summarizes the existing ABC algorithm, which proposes the framework of the ABC to solve combinatorial optimization problem, it takes a typical combinatorial optimization problem--Knapsack Problem to research. At the beginning, we use the method of combining transition probability and random selection to select the status, which not only ensure the better, but also the randomness. Then, the role conversion is an unique mechanism in bee colony algorithm. By studying the different role of honeybee and parameter analysis, we propose a new algorithm to improve the optimization performance. Finally, we introduce the tabu search to improve the ability of algorithm to jump out of local optimum.We solve the Knapsack Problem using the improved algorithm, the simulation results show that the improved ABC algorithm is correct and effective for solving combinatorial optimization.