Robust supervisory control policy for avoiding deadlock in automated manufacturing systems with unreliable resources.

Developing effective supervisory control policies that guarantee deadlock-free operations for automated manufacturing systems (AMSs) has been an active area of research for the past two decades or so. A great deal of work has been done for the systems with reliable resources; while only a few works are about unreliable resources. This paper addresses the robust deadlock supervisory control problem in AMSs with multiple unreliable resources. The objective is to develop a robust supervisory control policy for AMS under which the system can continue producing all part types not requiring any of the failed resources without manual intervention. Our policy is made up of a modified Banker’s Algorithm and a set of remaining resource capacity constraints. A state is feasible if and only if it satisfies both of them. By using the improved version of the existing Banker’s Algorithm, our control policy has more permissibility. An example is provided to illustrate that the policy gains advantage over the original one in permissiveness. [ABSTRACT FROM PUBLISHER]