A Petri net solution to the problem of deadlocks in systems of processes with resources

In this paper we introduce a new control policy for a class of nets that appear in the modeling of flexible manufacturing systems. Both the model of the system and its control are established by using Petri nets. The aim of the imposed control as to ensure that no deadlock problem can occur in the controlled system: a deadlock implies that the processing of some parts, once started, cannot be finished. This is a bad property for automated production systems. In the paper we use the structure of the Petri net model in order to characterize and solve the deadlock situations. We also present an experimental comparison between the proposed control policy and others that solve the same problem.