Safeguarding Schiphol Airports accessibility for freight transport

Automated, underground freight transport should enable sustainable economic growth in the Amsterdam area in the Netherlands. An innovative transport system, which guarantees reliable logistics and which avoids congestion problems, is currently being developed. This logistics system will be highly automated, using AGVs (Automatic Guided Vehicles) for transport and automated loading and unloading equipment. It is unique in its scale, covering a 15-25 km tube system, and in its complexity, using 200-300 small truck size AGVs in a multi-terminal setting. It requires considerable innovations in technology, especially in AGV control and AGV fleet controL Decisions about the implementation of such a system are characterized by a high uncertainty. Object-oriented simulations have been used to get insight in a multitude of design and management options. In this way, we could clarifY the relations between logistics system performance and resource capacities, such as the number of docks and parking spaces per terminal and the number of AGVs for various demand scenario's. Furthermore, we compared several possible layouts, both at the total system and at the terminal level, leading to design optimization. The control structure - e.g. intelligent empty AGV management and intelligent traffic rules - has a noticeable effect on the performance of the system. We showed that the investment costs could be reduced by ±20% by including two-way tube sections in the design, with only a slight, acceptable decrease in logistics performance. The strict use of generic object classes in the simulation provided the flexibility to address the changing user demands during the project. The simulations are advanced to the level that simulated and real objects can be mixed. Simulation control structures are currently used for testing prototype AGVs on a test-site. In this way, the risks of using the new technology can be reduced. Furthermore, the development time of the logistical control systems can be reduced considerably