A neural-based approach to production scheduling

An effective neural-based approach to production scheduling is proposed, which is apt for solving complex job-shop scheduling problems with available time and due date constraints, called constrained job-shop scheduling. A constraint neural network (CNN) is introduced to ensure the production constraints satisfied. A gradient search algorithm is applied to optimize the outputs of the CNN. The experiments have shown that the solutions generated by the neural-based approach are optimal scheduling for minimizing the sum of total job´s completion times in current processing sequence