Improved Adaptive Genetic Algorithm In Job-Shop Scheduling Problem Research

At the age of the growing global economic and the exacerbating market competition, the manufacturers have been improving production operation concept and developing rational management system constantly to consolidate their commercial position. In the real manufacturing environment. the production planning and shop scheduling have always been the key factors of affecting production costs and efficiency. However, the effective scheduling algorithm can make the benefits of enterprise maximize. Therefore shop scheduling problem attracts people’s widespread attention. And the study of job shop scheduling problem possesses great theoretical significance and practical value. The so-called shop scheduling problem can be described as a problem that how to make the reasonable arrangements in order to make production time or costs optimization. The job shop scheduling problem (Job Shop Scheduling Problem, referred to as JSSP) is one of the typical NP-Hard problems and the most difficult combinatorial optimization problems.In the past a few decades. many scholars have been looking for new scheduling algorithms to improve production efficiency, reduce production costs and enhance the competitiveness of enterprises. At present, the genetic algorithm is one of important algorithms in bionics methods, and it is also one of the evolutionary computation methods that used widely. It possesses the adaptability, global optimization and implicit parallelism in solving all sorts of nonlinear optimization problems, so it has irreplaceable advantages in scheduling optimization research. In this paper, it proposes an improved algorithm that is more suitable for Job-Shop scheduling, which is based on the adaptive genetic algorithm and combines with the artificial fish swarm algorithm. Through reading a lot of literature, I know that some of the existing adaptive genetic algorithms are easy to fall into local convergence, which affects the performance of the algorithm; the artificial fish swarm algorithm has strong local searching ability. So it can form a new algorithm that combines the adaptive genetic algorithm and the artificial fish swarm algorithm; the new algorithm introduces the local convergence index to judge whether the evolutionary population appear to the trend of local optimum, if it appears, the population will perform the artificial fish swarm searching, otherwise it will continue to perform the adaptive genetic algorithm; in the later period of algorithm execution, it will improve the diversity of population by increasing the mutation probability, which will produce more excellent individuals. The improved algorithm can not only improve the performance of the algorithm. but also be more suitable of solving Job-Shop scheduling problems.Finally. the paper makes the simulation experiments for the classic job shop scheduling problems. and implements the simulation system of the job shop scheduling. The results show the improved adaptive genetic algorithm has better performance and significantly improves the global optimization and speediness of the searching optimization process.