Simulation for predictive control of a distribution center

In this paper we present the application of Simulation for Predictive Control (SimPC) as a decision making tool for improvement of non-automated distribution centers (DCs). SimPC is focused on determining the viability of a given truck-dock assignment schedule, including arrival times and dock assignments for inbound and outbound trucks. SimPC also serves to perform iterative procedures of system parameter adjustments while searching for a viable schedule. The proposed model utilizes real-time data from DC's warehouse management system (WMS) to obtain the current state of the DC, which serves as initial conditions for the simulation. The model emulates the decision rules imbedded in WMS, which include assigning tasks to system-guided resources, selecting storage locations for inbound operations and determining retrieval locations for outbound operations. The SimPC model provides insights for the identification of scheduling problems, guidance for operational and tactical solutions, and serves as a tool to verify these solutions.