Centralized versus decentralized route choice control in DCV-based baggage handling systems

The process of handling baggage in an airport is time-critical. Currently, the fastest way to transport the luggage is to use destination coded vehicles (DCVs). These vehicles transport the bags at high speed on a ldquominirdquo railway network. The route of each DCV has to be computed in order to optimize the performance of the system. In this paper we determine an event-based model of a DCV-based baggage handling system and we compare centralized and decentralized control methods for routing the DCVs through the network. The proposed centralized control methods are optimal control and model predictive control. Due to the large computational effort required, we also analyze a fully decentralized control approach. In this case, each junction has its own local controller for positioning the switch into the junction or out of it. The considered control methods are compared for several scenarios. Results indicate that optimal control becomes intractable when a large stream of bags has to be handled, while MPC can still be used to suboptimally solve the problem. On the other hand, the decentralized control method usually gives worse results than those obtained when using MPC, but with very low computation time.