Analysing the Effects of Congestion on Hybrid Order Picking Systems Using a Discrete-Event Simulator

In hybrid order-picking systems (OPSs), human workers collaborate alongside autonomous guided vehicles (AGVs) to pick up and transport items in a warehouse. Congestion occurs when multiple humans and AGVs operate in an area simultaneously. Congestion decreases AGV navigation performance and may cause queuing delays at packing stations. In this paper, we study the impact of congestion on the performance of hybrid OPSs. We simulate a hybrid OPS using a discrete-event simulator and evaluate the throughput under different levels of congestion and the number of AGVs. Using 10 AGVs under no congestion, we observe a throughput increase of 105% compared to a manual OPS with zero AGVs. However, this improvement decreases as the effects of congestion become stronger. Under our heaviest congestion model, there was only a 3% throughput increase for 10 AGVs. We also analyse the economic impact of adding AGVs to a hybrid OPS. Under medium congestion, the optimal number of AGVs for maximising long-term profit is 20.