The Maasvlakte port area in Rotterdam is located in the Netherlands. There are 14 container terminals and empty container depots located within this area, some are existing and some will be built in the near future. At container terminals containers are being transhipped between different modalities. At empty container depots, empty containers are being stored for a certain time. The container terminals are sharing the same customs facility and empty container depots. Transport between the container terminals and these facilities has to be organised. It is also possible that containers arrive at one container terminal, but have to depart from another container terminal. This transport between different terminals and depots is called Inter Terminal Transport (ITT). This ITT has to be organized in the future. The main question of this research is: Which transport system is most appropriate for Inter Terminal Transport between the terminals and depots at Maasvlakte 1 and 2 in 2030 and why? Previous studies were performed to determine the most appropriate transport system for ITT. This study includes a new simulation model to compare the ITT systems, in which the most up-to-date information can be used for the generation of new ITT container demand scenarios. The minimum percentage of on-time containers is set as a constraint, which was not done in previous researches. In this way a fair comparison can be made of the ITT systems and they can be evaluated using multiple criteria. In this research the AGV, ALV, and MTS will be compared. The AGV and ALV are automated vehicles, have a capacity of 2 TEU, and are expected to drive with a speed of 40 km/h in 2030. The ALV is able to load and unload containers by itself. The MTS is a manned vehicle, has a capacity of 10 TEU, and is expected to drive with a speed of 30 km/h in 2030. These ITT systems will be compared for three different economic growth scenarios (defined by the Port of Rotterdam): low growth, high oil prices, and European trend. An existing container generator (which is made in an earlier project) is used to create a table with containers. Each containers has the following properties: start time, origin (terminal/depot), delivery time, destination (terminal/depot), and size in TEU. The amount of generated containers for ITT differs per economic growth scenario: - Low growth scenario: 0.70 million containers/1.17 million TEU per year - High oil prices scenario: 0.78 million containers/1.29 million TEU per year - European trend scenario: 0.99 million containers/1.66 million TEU per year The list of containers from the generator is used as input for a simulation model. For this research a dynamic, discrete-event stochastic model has to be built. Multiple simulation software packages exist that are able to build such kind of model, but a choice has to be made. Based on the availability of the programs at the university, the costs, and the user friendliness, Simio is chosen. Containers arrive at the terminals using their start time and origin property. Containers are labelled as urgent when their latest departure time (delivery time minus the expected loading, unloading and transport time) is within 1 hour, otherwise the containers are labelled as non-urgent. The containers are sorted in different waiting areas; for each destination a separate waiting area is created. Those waiting areas are divided in two parts: the urgent part and the non-urgent part. The containers are virtually sorted by means of their latest departure time; the containers that must depart first are in front of the waiting area (FEFO). The containers are loaded on the ITT vehicles using the terminal equipment (except for the ALV). Each terminal has its own type and number of terminal equipment. Only (urgent and/or non-urgent) containers with the same destination are loaded on the same vehicle. The ITT vehicles depart when they are fully loaded or when a certain time limit is reached. A dedicated road for ITT vehicles will be constructed. 3-way crossings and crossings with rail or public road are located within the ITT network, where vehicles can be delayed. When vehicles arrive at their destination terminal, the terminal equipment unloads the vehicle (except for the ALV). For each terminal this is the same terminal equipment that is also used for loading the ITT vehicles. After unloading, the ITT vehicles are empty again and determine their next transport assignment. It is checked if there are containers waiting at the current terminal or another close terminal (using a predefined list of terminals), starting from the closest to the most far away terminal. If there are urgent containers waiting, the vehicle drives to that terminal (and takes non-urgent containers from the current terminal with that terminal as destination if possible). If there are no urgent containers waiting, all the terminals are checked to see if there are non-urgent containers waiting, starting from the closest terminal (the current terminal) to the most far away terminal. If non-urgent containers are found, the ITT vehicle drives to that terminal. If no containers are found, the ITT vehicle waits and repeats the process of searching containers. It is by far the most important task to deliver the containers on time. Therefore a minimum percentage of 95% on-time containers is used, to get a fair comparison between the ITT systems. The number of vehicles is changed as input, until the minimum number of needed vehicles is found to achieve 95% on-time containers. This minimum number of needed vehicles per ITT system per scenario is shown in Table 0 1. Table 0 1: Needed number of vehicles per ITT system per scenario to reach 95% on time containers AGV ALV MTS Low growth 34 AGVs 24 ALVs 14 MTS-trucks, 90 trailers High oil prices 39 AGVs 27 ALVs 16 MTS-trucks, 105 trailers European trend 53 AGVs 34 ALVs 20 MTS-trucks, 130 trailers The AGV needs clearly the highest number of vehicles, then the ALV, and the MTS needs the lowest number of vehicles. The difference between the AGV and ALV can be declared by the fact that the ALV has the ability to load and unload itself. The capacity of the MTS is 5 times as large as the capacity of the automated systems. This is a clear advantage, because it needs less vehicles than the automated systems, though, the MTS has a lower vehicle speed and has to wait longer during loading and unloading. The automated systems can reach percentages of on-time containers close to 100%, but the MTS can reach only approximately 97%, because of its long loading and unloading time due to its high capacity. The output of the simulation model is used for a Multi Criteria Analysis (MCA). The MCA is used to evaluate the different ITT systems, based on two criteria: total costs and sustainability. The total costs of the three ITT systems are very comparable. For the low growth and high oil prices scenario the total costs of the AGV and MTS are equal. The total costs of the ALV are 4% higher than for the AGV or MTS. For the European trend scenario the MTS has the lowest costs. For this scenario the costs of the AGV are 6% higher than the costs of the MTS, and the costs of the ALV are 7% higher than the costs of the MTS. The vehicle costs of the MTS are much lower than the automated systems, but the personnel costs are on the other hand much higher. The sustainability of the ITT systems is measured by the electricity usage. The electricity usage of the automated systems is 99% to 166% higher than the electricity usage of the MTS. The ALV uses approximately 25% more electricity than the AGV. Independent on the weighing factors of the MCA the MTS would have the highest score. The MTS has the lowest costs for each of the scenarios and is the most sustainable ITT system. The difference between the MTS and the automated systems is relatively small looking at the total costs but there is a significant difference looking at the sustainability. The difference between the score (of the MCA) of the MTS and the automated systems depends on how important sustainability is considered by the decision makers. The higher the importance of sustainability is considered, the higher the difference of the score between the MTS and the automated systems will be. Eventually, the MTS will receive in all cases the highest score. Therefore the MTS is considered as the most appropriate ITT system for each of the economic growth scenarios. The input data did not contain large flows from and to empty depots, but in reality these flows will be present. This causes that the amount of ITT flows will be larger, which probably causes that the MTS scores better on total costs, because the MTS scores better on total costs with the highest volume of ITT in the European trend scenario compared to the other two scenarios. Next to this, the MTS is probably a safe option to choose, because for the AGV and ALV large developments are assumed according to the vehicle speed in contrast to the MTS. It is recommended that future research will be done on different fields. It is interesting to see what happens if the ITT flows are divided differently over the ITT network, and if the large flows from and to the empty depots are taken into account. It is interesting to model multiple years instead only 2030, to find out at which moment it is best to implement a new system. It can be investigated if combinations of ITT systems are more appropriate. Another dispatching method can be used to see if a decrease in the number of vehicles can be reached. It is important to know if there are important decision criteria (and their corresponding weighing factors) missing in the MCA, according to the decision makers. This can have an influence on the results of a MCA. It is also interesting to see what the amount of extra costs is to achieve percentages of on time containers close to 100%. It can be considered if the extra on time delivery of the containers can outweigh the extra costs.