Self-adaptation Genetic Genetic Operators To Solve Curriculum Scheduling Probelm

There are still many issues to be tackled with in modern scientific theory research and practice,with regard to Combination&Optimization and self-adaptation etc. Routine methods are quitehelpful resolving simple optimization and self-adaptation problems but helpless for complicatedlarge-scale systems. Genetic Algorithms,based on the biological mechanism of natural selection&heredity and leveraging colony searching technology,is particularly applicable for the resolution of complicated&non-linear problems intractable with traditional searching methods. For nearly 40 years'development,Genetic Algorithms has made great achievements in both theory research andpractical applications.However,its mathematical foundation is still incomplete compared with thedistinctive and sound biologic foundation.Curriculum scheduling problem is a typical problem aboutmulti-restraints and combination optimization,and has been proved to be aNondeterministic Polynomial Completed (NPC) problem in the 1970's.The GeneticAlgorithm(GA),based on the biological mechanism of natural selection&heredity,is an adaptive and stochastic search algorithm. It can be highly implemented inparallel. For the stability and generality of GA,GA is often used to solve complicated NPC Problem. In this paper Genetic Algorithm(GA)is applied to solve curriculum scheduling for a university. The main works as follows: Firstly,all kinds of potential factors,multi-restrictions of curriculum scheduling problem are discussed. What's more,the difficulties to resolve the curriculum scheduling problem and the optimal object are represented,and the mathematic model about curriculum scheduling problem is designed.Secondly,some classic GA coding schemes,as well as their applications in curriculum scheduling system are analyzed and compared in details. Based on the studies about all coding schemes'strengths and weaknesses,an improved genetic algorithm for solving curriculum scheduling problems is proposed. Compared to traditional coding schemes,this scheme is simpler,more effective and easier to understand.Meanwhile,the cross and mutation genetic operations are all comprehensibly redesigned to correspond to this code scheme.