A stochastic optimization approach for robot scheduling

In this paper we are concerned with a stochastic optimization approach for determining the optimal job-sequencing in a robot-handler production system, such that the best value of some performance indices which depend on control parameters are obtained. The idea is to reflect the possible control policies of the system in its Stochastic Petri Net model (SPN) and to select a suitable conflict resolution rule whenever the transitions representing the possible actions of the robot are enabled. This rule would depend on a vectorx isin Ropf n of control parameters, and the problem results in finding the values of those parameters which would be in some sense optimal for the system. The objective function is defined as a linear combination of several performance indices that are estimated simultaneously. We propose a combined simulation and optimization approach aimed at solving the conflict situations arising in the system due to simultaneous requests of the robot from jobs in different queues; then we establish a stochastic optimization approach for deriving control policies that govern the flow in the SPN model. The theoretical optimization criteria are presented along with a case study.