An Adaptive Policy of Dynamic Scheduling in Knowledgeable Manufacturing Environment

To overcome deficiency in the global capacity of a single dispatching rule, it is very important to select a dispatching rule in real time in dynamic scheduling. Although some literature addresses the method of selecting dispatching rules, little literature requires no domain knowledge or accurate training examples, which is rather difficult to acquire for the real production system. In order to obtain the dynamic scheduling knowledge effectively, a B-Q learning algorithm is proposed in this paper, and one kind of adaptive scheduling control policy is presented based on this algorithm. According to the transient state current system stays, different dispatching rules are selected to schedule the jobs in machine buffer. A case study is presented to illustrate the validity of the scheduling control policy.