关键词: 神经网络, 作业车间调度, 组合优化, 计算能量函数

Abstract: This paper improved existing permutation matrix of job-shop scheduling problems. A new computational energy function of Hopfield neural networks for the job-shop scheduling problems is given. Then, the modified method solving job-shop scheduling problem based on Hopfield neural network is proposed. For avoidng Hopfield neural network convergence to local minimum value, simulated annealing algorithm is avoiding to Hopfield neural network and the production scheduling method based on stochastic neural networks is proposed. Compared with existing method, modified method can keep the steady outputs of neural networks as feasible solution for job-shop scheduling problem.

Key words: neural networks, job-shop scheduling, combinatorial optimization, computational energy function