Job Shop Scheduling With Genetic Algorithms

Job shop Scheduling problem is a Typical NP-hard Problem,and it is a important study issues in the CIMS areas. Its study has great practical significance,and has profound theoretical significance. In recent years, Using the genetic algorithms to resolve the Job shop Scheduling problem has become the focus of researchers,which along with the continuous development and mature of genetic algorithms.In this paper, the operation-based representation are used to resolve the Job shop Scheduling problem. Because the strict model theorem about the Binary representation only been proved in theory for genetic algorithms, the model theorem about the the operation-based representation is proved in this paper. The Scheduling's Legitimacy which made by this representation on this basis. After has the guarantee of representation, a immune operator based the current best solution is Design in this paper,and the immune-genetic algorithms is Successed used to resolve Job shop Scheduling problem.It is found that the immune-genetic algorithms Obviously superior to the basic genetic algorithms through comparison with it.But the immune-genetic algorithms is no better than the mixed genetic algorithms when come to the relatively large Job shop Scheduling problem.so a typical mixed genetic algorithms(simulated annealing algorithms combined with the genetic algorithms ) is improved at the last of this paper, which is add with immune operator this paper designed. It is shown that the improved mixed genetic algorithms Obviously superior to the Original mixed genetic algorithms, through which the immune operator this paper designed is meaningful, can been proved .