Mathematical programming models for joint simulation–optimization applied to closed queueing networks

The optimization of stochastic Discrete Event Systems (DESs) is a critical and difficult task. The search for the optimal system configuration (optimization problem) requires the assessment of the system performance (simulation problem), resulting in a simulation-optimization problem. In the past ten years, a noticeable research effort has been devoted to this area. Recently, mathematical programming has been proposed to integrate simulation and optimization for multi-stage open queueing networks. This paper proposes the application of this approach to closed queueing networks. In particular, the optimal pallet allocation problem is tackled through linear mathematical programming models for simulation-optimization.