Analysing the solution of production-inventory optimal control systems by neural networks

In this paper, a general production‐inventory optimal control system is proposed which can be used in most cases that might arise in the theory of production‐inventory control. The proposed general form is considered and approximately solved using neural networks. Since the obtained solutions are achieved based on neural networks, they have several advantages in practice. One of the important advantages is that the solutions can be easily used for post optimality and sensitivity analyses. The solutions of this model are compared with those of other existing methods and some illustrating notes are presented.