Improvement Of Artificial Bee Colony Algorithm And Its Application In Economic Order Model

Economic order quantity(EOQ)is the quantity that balanced various cost accounting to make the total cost become minimum.In order to calculate the quantity,it is necessary to predict the amount of orders for the accuracy.Support vector machine(SVM)can be used to calculate the amount of the previous order,and predict the amount of the future,then obtain the result of EOQ.Therefore,in order to make the learning effect of SVM more accurate,the method of optimizing support vector machine has become one of the hot research problems.The artificial bee colony algorithm is a group intelligent optimization algorithm which simulates the behavior of honey bees.Because it has the advantages of less control parameters,easy to implement,simple calculation and strong robustness,it has a good performance in optimization problem such as optimize support vector machines,and more and more researchers paid attention to it.Artificial colony algorithm has two main disadvantages: the algorithm is easy to fall into local optimal and premature convergence especially when dealing with complex optimization problems;The algorithm has better exploration ability,but the exploitation ability is insufficient and the convergence veracity is slow.This paper attempts to improve the artificial bee colony algorithm and improve its performance in dealing with complex optimization problems.Then,use the algorithm into optimize support vector machine to predict the order prediction problem in the economic order quantity model.The main contents of this paper include the following two aspects:On the one hand,in order to improve the accuracy of algorithm optimization and local search ability,based on the existing bare-bones artificial bee colony(BBABC)algorithm and hybrid bee colony algorithm,a hybrid bare-bones artificial bee colony algorithm(HBABC)has been presented.The algorithm mainly improves the following two aspects: For the algorithm is easy to fall into the local optimal,The HBC algorithm is introduced to improve the model of the honey source update from the characteristics of the simulated annealing algorithm;for the algorithm is lack of convergence,the property of tend to optimal individual in BBABC is imported to improve the function of onlooker bees choosing employed bees.The algorithm improves the convergence accuracy and the optimization speed through the above two improvements.Through using 10 benchmark functions to comparison test,the effectiveness of the improved algorithm has been proved.On the other hand,the HBABC algorithm was used to optimize the two parameters of the support vector machine,and the optimization results are used to solve the existing practical problems – the problem to fitting and predicting of the order amount and price based on the EOQ model.The experimental results show that the support vector machine with HBABC algorithm is more accurate than the support vector machine optimized by ABC and BBABC.