The goal of the project DONAUHANSE is to establish a network of cities, which provides sound solutions to similar problems with the greatest possible benefit for all participating cities and regions along the Danube.


The simulation tool includes elements that can be modelled on a macroscopic level as well as explicit micro simulation exercises that need a vehicle and process based decision process during each simulation run.

In terms of network development OpenStreetMap (OSM) can be used in order to have alternative data sources in addition to official sources such as municipality authorities where data needs to be requested explicitly. Open data is made increasingly available on an international level which is one of the benefits since the simulation system should be able to handle different hub environments on an international level. Bearing in mind different available software product VISUM/VISSIM was chosen for carrying out the modelling background.

VISUM can be used for the establishment of the basic network with nodes and lines and can provide input for the generation of the simulation environment in VISSIM using the interface for data exchange between both programs. OSM can be used to import road network elements independent of the current area of investigation thanks the unique coding of the OSM elements. Minor efforts are required to get a routing capable network as basis for investigations. The network nodes are defined automatically based on the start and end points of the imported network line elements.

For visualisation purposes digital background maps can be used such as aerial photography or OSM maps. During simulations 3D-views can be accessed with detailed models of vehicles such as trains, cranes, ships, trailers and cargo units that can be created individually or imported from e.g. the 3D Warehouse – Sketchup. Furthermore the state of these models can be changed in order to reproduce handling processes or container depot areas. Consequently detailed visualisations can be produced which are based on real-time running processes. This is a major benefit in comparison to state-of-the-art visualisations with other programs that are based on static elements.

Detailed vehicle actuated (traffic signal) definitions are one of decisive advantages of the chosen simulation tool since they can be implied with high flexibility in different ways without extensive programming skills. During the simulation concept development phase the vehicle actuated definitions were chosen as most promising option to represent some of the dependencies of different processes and vehicle based decisions. They can further be used to trigger other processes and decisions throughout the simulation runs.

Apart from the already identified advantages the main decisive element of the chosen simulation tool is the pre-defined interface opportunity with Excel and Excel VBA scripts - so called VISSIM-COM. This gathers for nearly unlimited interaction possibilities between data in Excel spreadsheets, algorithms in Excel scripts and Import/Export functions of VISSIM data of nearly each element of the program such as vehicle and simulation based data. This interface has been established in the past years by the provider PTV AG and has already proven high flexibility in terms of programing during the early project phase. Another advantage is the text-based program language of each VISSIM simulation file which allows the user to manipulate and copy individual network elements such as connectors, links, signal groups, detectors and static objects amongst others.

Within optihubs the interface described above is used to:

  • import specified data such as second-by-second interval based vehicle inputs (e.g. vehicle type, desired speed and start location, loading, target terminal, driving behaviour settings) from individually defined Excel spreadsheets that provide all the original data from real-world terminal operation systems (AVSIO-data);
  • define simulation settings based on Excel spreadsheet masks prior and during each simulation run that include but are not limited to time constraints (e.g. checking processes), number and types of cargo-handling machinery (e.g. number of reach stackers), priorities of work orders for staff;
  • calculate and initiate processes within the simulation run based on generated VISSIM data during the simulation run that is assessed through queries and reference tables;
  • export data such as vehicle specific timestamps at different locations as well as details of finished work schedules of each cargo handling machinery in order to derive key performance indicators.

Consequently the following simulation exercises can be:

  • reproduction of vehicle based traffic streams (e.g. including the assessment of traffic queuing developments, incidents and over-saturation situations at intersections and key points such as OCR gates)
  • flexible definition of specific areas (e.g. definition of access and exit points of the hubs and terminals, interaction of different structures and cargo handling areas)
  • reproduction of processes (e.g. analysis of performance, efficiency and effectiveness)
  • definition of different priorities of work orders such as priorities of handling incoming vehicles with delays
  • definition of different incoming and outgoing amounts of cargo units by type and time

A very interesting side effect during development of these preliminary algorithms was the fact that the decision processes could be based on extensive input data made available from the model itself that was not being available in real-world. Therefore the preliminary algorithms proved to be more efficient in some cases than the current real-world decision making processes in place (due to limited data availability).

In addition to the discussions with expert staff synergy effects could be taken into account from Briefer and Zimmermann (2011) who dealt with the analysis and replication of some specific processes of the container terminal that is to be simulated within optihubs. Consequently the items of the process overview could be analysed by the programmers prior to algorithm development.

Apart from subjective validation efforts also quantitative model validation can be carried out based on the available information of specific timestamps at several locations of each processed cargo unit from the real-world terminal management system.

Throughout the simulation tool development optihubs benefited from extensive amounts of terminal data as well as other data sources that are as follows:

  • Vehicle permits on a daily base from the in- and out gate of a terminal including the type of vehicle, desired direction and time of entry and exit
  • Layouts of proposed (alternative) in- and out gates including details and time intervals for checking periodFuture potentials in terms of cargo unit types and mode of transport
  • Briefer and Zimmermann (2011)  - refer to chapter 4.2 for details
  • Data from the work schedules of a terminal management system including details about each processed cargo unit and cargo units in stock such as encrypted number and name, length, height, type, condition, loading, weight, status, date and time of arrival and departure, incoming and outgoing mode of transport, incoming and outgoing transport unit, first and last position within the container depot areas, number of manipulations of the cargo unit between the first and last position
  • Timestamp from the OCR-Gate of the outgoing road transport units
  • Information regarding delays of incoming and outgoing trains
  • Information about the setup of the individual trains and corresponding location of each cargo unit
  • Map of current and planned terminal layouts
  • Container depot information (number and position of cargo units)
  • Technical parameters of different cargo handling machinery types
  • Traffic signal programs for the area of interest around the port of Vienna

In comparison to other research projects that have already been carried out by the corresponding author’s company the amount of input data made available for optihubs is extraordinary. Consequently the following inputs have already been derived for the simulation tool (additional input will be created along the remaining project time):

  • Setup of incoming and outgoing trains and corresponding location of each simulated cargo unit including timestamp of arrival and departure
  • Location of the individual loading areas as well as in- and out gates
  • Link of incoming and outgoing cargo units and transport units
  • Location of incoming road transport units prior to cargo handling
  • Background information for all processed cargo units and cargo units in stock to derive key performance indicators and triggering of processes according to different cargo unit types
  • Container depot situation (cargo units that won’t be accessed throughout the simulation period reduce the availability of container parking positions)·      

Simulation scenarios and key performance indicators

In cooperation with all project partners of OPTIHUBS several simulation scenarios have been defined. For these scenarios the simulation tool should preferable provide input in terms of potential optimization on different levels of available measures. These scenarios are:

  • Analysis of traffic queuing caused by heavy vehicles in conjunction with existing traffic signals at existing and planned in- and out gates of multimodal hubs
  • Assessment of the impact of heavy load cargo handling in combination with parallel running “normal” processes
  • Effects of different train track lengths on cargo handling and key performance indicators
  • Analysis of future layouts and changes in the amount of incoming and outgoing cargo
  • Effects of increased cargo transport along the inland waterways and its handling in addition to current handling processes at multimodal hubs
  • Effects of different in- and out gate configurations regarding traffic flow, dwell times, processing times, traffic queuing, available loading areas and checking procedures
  • Comparison of different cargo handling machinery usage in terms of key performance indicators
  • Manipulation of different existing and planned cargo unit types and their effects on container depot areas
  • Input for process optimization through increased data availability
  • Effects of different measures in terms of cargo unit arrivals

All of these scenarios will include simulations of different settings such as existing and planned configurations in order to be able to provide the requested input for processes optimization. Several key performance indicators are implemented in the model that can be used for assessing the effects of different settings and scenarios. The current setting of the key performance indicators is described in the following table.

Indicator (Average, Max, Min)



Cargo unit handling output

tons/hour, tons/day

Storage capacity (per terminal)

cargo units/day

Dwell time


Cargo unit handling time

minutes/cargo unit

Degree of capacity utilization of cargo handling machinery



Further performance indicators can also be made available or derived based on data exports from the simulation tool.