Route choice control of automated baggage handling systems ∗

State-of-the-art baggage handling systems transport luggage in an automated way using destination coded vehicles (DCVs). These vehicles transport the bags at high speeds on a “mini” railway network. Currently, the networks are simple, with only a few junctions, since otherwise bottlenecks would be created at the junctions. This makes the system inefficient. In the research we conduct, more complex networks are considered. In order to optimize the performance of the system we develop and compare centralized and decentralized control methods that can be used to route the DCVs through the track network. The proposed centralized control method is model predictive control (MPC). Due to the large computation effort centralized MPC requires, decentralized MPC and a fast decentralized heuristic approach are also proposed. When implementing the decentralized approaches, each junction has its own local controller for positioning the switch going into the junction and the switch going out of the junction. In order to assess the advantages and disadvantages of centralized MPC, decentralized MPC, and the decentralized heuristic approach, we also discuss a simple benchmark case study. The considered control methods are compared for several scenarios. Results indicate that centralized MPC becomes intractable when a large stream of bags has to be handled, while decentralized MPC can still be used to suboptimally solve the problem. Moreover, the decentralized heuristic approach usually gives worse results than those obtained when using decentralized MPC, but with very low computation time. Tarău, De Schutter, Hellendoorn 3