| Honey-bee Mating Optimization （HBMO） was originally introduced by abass in2001.Because of its basic principle simple and easy combining with other algorithm.It has beenattracting more and more attention.It is used in many fields such as TSP、function optimization、cluster and other applied fields.Arranging course structure is essential in academic administration of university, likewise itis a complicated management task. However the traditional curriculum arrangement way is lowefficiency and unfair. Therefore,it has great theotetical value and application value to researchthe stable high intelligent curriculum arrangement method. The paper is aimed at improvingoriginal honey-bee mating optimization and using the improved algorithm to solve coursetimetabling problem. The main research works can conclude as follows:First,due to the drone of HBMO hasn’t change in the iteration process and the worker’sexploration ability is infficient,the colony’s diversity is disappear and the algorithm has a poorperformance of global.An improved HBMO is proposed where use the brood to replace thepeakest drone and apply double neighborhood local search strategy.the improved algorithmeffectivelly avoid the occurence of premature phnomenon and improved the precision ofsolution.Secondly,gives the way of real number coding for crossover operator and mutationoperator.the new crossover operator mixture the parents advantage gene, produce a moreexcellent progeny individuals,accelerate the algorithm convergence. In the mutation operatorintroduces kempt chain,reduce the running timeThirdly, analyzed the main factors which influence the curriculum arrangement quality.Otherwise,constructed course timetable model and applied the improved bees mating algorithm to solvecourse timetabling problem.Experiments upon socha benchmark show that,the improved HBMOis more efficient to solve the timetabling problem with higher global successive rate andaccuracy.At last,design and develop curriculum arrangement system. |
PDF Full Text Request