Application research of improved genetic algorithm based on machine learning in production scheduling
Kai Guo,Mei Yang,Hai Zhu
DOI: https://doi.org/10.1007/s00521-019-04571-5
2019-11-01
Neural Computing and Applications
Abstract:Job shop scheduling problem is a well-known NP problem. It is limited by various conditions. As the scale of the problem increases, the difficulty of finding the optimal solution will increase. It is a difficult combination problem. Limited by the constraints of the actual production environment, how to effectively arrange the processing order of each part will directly affect the production efficiency, the appropriate production scheduling algorithm can correctly and effectively plan the enterprise resources and rationally arrange the processing order and processing time of the workpiece. Proper use of existing resources, by optimizing production scheduling instructions, to meet the basic requirements of production scheduling, in order to obtain the optimization of total production time, has important theoretical significance for the actual production of enterprises. In this paper, the mathematical model is abstracted on the basis of the production scheduling problem. According to the different parts of the same machine and the different processes of the same part, the corresponding processing time and waiting time are obtained. At the same time, the genetic algorithm is improved by genetic algorithm. A dynamic genetic operator based on the number of iterations is proposed, which further enhances the convergence performance and search ability of the genetic algorithm. Through the simulation of MATLAB simulation program, combined with the scheduling standard example, the performance analysis of different algorithms is carried out, the search efficiency of genetic algorithm is improved, the convergence performance of the algorithm is improved, and different optimization choices are obtained for different time weights. The operation results of the system meet the requirements of production scheduling, which proves the feasibility and practicability of the improved genetic algorithm.
computer science, artificial intelligence