Improved Niche Genetic Algorithm For Job Shop Problem

Scheduling theory stems from the manufacturing workshop production planning and control research.Workshop production scheduling and control technology is the key to achieve high efficiency,high flexibility and high reliability.As the core of manufacturing system,the production management workshop production scheduling system is an important means to achieve the production plan and to maximize production efficiency.Meanwhile,job shop scheduling problem is bound to NP studies play an important role because of its discrete,dynamic,multi-machine,multi-variables and constraints and other typical NP-hard resistance.Genetic algorithm that draws on the principles of biological evolution in nature is a high parallelism,adaptable and stochastic global search algorithm.It is a multi-parameter and multi-group optimization parallel approach which applied biological evolution that includes natural selection,survival of the fittest ideas and random variation to the problem solving spatial search of the optimal solution.In recent years,Genetic algorithm has been widely used in many areas for its simplicity,good operability and good optimization ability,such as complex multi-objective programming,artificial life,neural network problems,machine learning problems and intelligent control problems,have proved the genetic algorithm is one of the most effective method.In this paper,author proposed a isolated niche genetic algorithm based on adaptive technology to solve shop scheduling shop scheduling problem(JSP)that is a typical issue in shop scheduling.Improved hybrid genetic algorithm generates an initial solution space firstly,and divides the solution space into different sub-space.One hand this method can keep better the traditional genetic algorithm parallelism,on the other hand it makes the algorithm more in line with the natural evolution of the law after the introduction of inferior species do not live principle and the same kind of repulsion principle,and improves the global search ability for increased competition among the sub-groups.Meanwhile,Improved hybrid genetic algorithm introduces the adaptive crossover and mutation probability in the process of each sub-group's evolution,so that ensures the diversity of solution space of the entire algorithm,and increases the ability to escape from local optimal solutions,and improves the efficiency of cross and mutation greatly,and enhances the convergence rate.Finally,based on the above work,author the designs a shop scheduling system that can dynamically observe the entire scheduling process,so that further proves the effectiveness of the improved genetic algorithm.