Stochastic modeling of multiline orders in integrated <scp>storage?order</scp> picking system

Due to demanding service levels in e-commerce order fulfillment, modeling and analysis of integrated storage picking processes warehouses deserve special attention. The upstream system can have a significant impact on the performance downstream process. With particular focus multiline orders, we develop an analytical framework for (shuttle-based retrieval system) (pick networks. To capture consolidation delays fulfilling pick is modeled with closed queuing network that includes synchronization nodes. configuration station adapted model variety profiles handled at station. For network, propose decomposition-based solution methodology results good accuracy. resulting semi-open (SOQN) analyzed using matrix-geometric method (MGM). improve accuracy estimates measures, hybrid simulation/analytical framework, where measures complex subnetworks are obtained from simulation. We also detailed simulation physical validating measures. experiments indicate approach reduces error throughput time 3% 18% model. Then, investigate effect (such as number aisles) mixed vs. dedicated picking, CONWIP control batching) times. Our provides threshold maximum numbers allowable orders (CONWIP control) aisles beyond which improvement average marginal. Numerical high-order arrivals highlight result reduction comparison picking.