Research And Application Of Artificial Bee Colony Algorithm For Parallel Machine Scheduling Problem

With the profound change of the world economic development mode,logistics,as the third-party profit source,has attracted more and more attention of the international community.Generalized logistics includes transportation,storage,processing,distribution,information processing and other processes.Parallel machine scheduling is an important problem in the field of logistics production and processing,which widely exists in the production workshop,and the research on this kind of problem has important practical significance.Artificial bee colony(ABC)is a relatively new swarm intelligence optimization algorithm inspired by the biological phenomenon of bee population foraging in nature.It has the characteristics of strong robustness,high accuracy,few parameters,simple principle and easy to realize.It is widely used in the fields of artificial neural network training,function optimization,production scheduling,power system optimization,etc.In this paper,an improved discrete artificial bee colony algorithm(IDABC)is proposed to solve two kinds of parallel machine scheduling problems whose objective function is to minimize the maximum processing time and the minimum processing cost.The improvement of the algorithm includes the following aspects,firstly,using Logistic chaotic map to design the initialization strategy of population,to obtain the initial population with uniform distribution.secondly,an adaptive control method of parameters to be optimized is proposed to accelerate the convergence of the algorithm.thirdly,using the mutation operator of the differential progressive algorithm for reference,a new local search method is proposed to balance the global search and the local search.Then,in the following bee stage,we use the idea of accepting the bad solution with a certain probability in the simulated annealing algorithm to design the updating process of the food source,so as to prevent the algorithm from falling into the local optimum.Finally,in the bee detection stage,we make full use of the information of the high-quality solution to update the food source and improve the searching ability.Through a large number of comparative tests,this paper shows that the IDABC algorithm is better than other methods in solving accuracy,convergence speed and solution stability,which shows that this algorithm can effectively solve the above two kinds of parallel machine scheduling problems.