This repository provides a MATSim transport model for Munich, provided by the Associate Professorship of Travel Behavior of Technische Universität München. The Munich MATSim scenario includes a network generated from OpenStreetMap and a plan file, which is the result of the agent-based travel demand model MITO. This model simulates travel demand at the individual resolution, based on the results of the household travel survey Mobilität in Deutschland. For further details see our prefered citation:
- Moeckel, Rolf, Kuehnel, Nico, Llorca, Carlos, Moreno, Ana Tsui and Rayaprolu, Hema (2020) Agent-based Simulation to Improve Policy Sensitivity of Trip-Based Models. Journal of Advanced Transportation, 2020. https://doi.org/10.1155/2020/1902162
The scenario is ultimately coming from a SILO model that is maintaned elsewhere.
Handling of large files within git is not without problems (git lfs files are not included in the zip download; we have to pay; ...). In order to avoid this, large files, both on the input and on the output side, reside at https://svn.vsp.tu-berlin.de/repos/public-svn/matsim/scenarios/countries/de/munich/munich-v1/input/. These data are open and can be downloaded.
- Click on
Clone or download
and then onDownload ZIP
. - Unzip the repository.
- Go to "Download the data" below.
- Install git for the command line.
- Type
git clone https://github.com/matsim-vsp/matsim-munich.git
in the command line.
(Or use your IDE, e.g. Eclipse, IntelliJ, to clone the repository.)
This will result in a new matsim-munich
directory. Memorize where you have put it. You can move it, as a whole, to some other place.
- Go into the
matsim-munich
directory. - Type
git pull
(Or use your IDE, e.g. Eclipse, IntelliJ, to update the repository.)
This will update your repository to the newest version.
There is no need to download the data to run the base scenario provided by the team from TUM. The is downloaded directly when running matsim via their url. In any case, the input data can be downloaded from the svn server.
- Install svn for the command line.
svn checkout https://svn.vsp.tu-berlin.de/repos/public-svn/matsim/scenarios/countries/de/munich/munich-v1/input/
- Go into the matsim-munich data directory.
- Type
svn update
.
(Requires either cloning or downloading the repository.)
- Set up the project in your IDE.
- Make sure the project is configured as maven project.
- Run the JAVA class
src/main/java/org/matsim/gui/MATSimGUI.java
. A simple GUI should open. - In the GUI, click on the "Choose" button for configuration file. Navigate to the
scenario/tumTb/configBase.xml
to load the configuration file. - Increase memory in the GUI.
- Press the "Start MATSim" button. This should run MATSim.
- "Open" the output directory. You can drag files into VIA as explaned below.
- Edit the config file or adjust the run class. In particular, it makes sense to reduce the lastIteration number for trying around. Re-run MATSim.
To work outside the GUI, you need
- the code (from github) in
<someDirectoryRoot>/<something>/matsim-munich
- the data (from svn) in
<someDirectoryRoot>/shared-svn/projects/matsim-munich
More technically, the code searches, from the java root, for the data by ../../shared-svn/projects/matsim-munich
. This is a bit annoying, but everything else (e.g. git lsf, git fat) is worse.
So please move the code directory and the data directory accordingly.
- Get VIA from https://www.simunto.com/via/. (There is a free license for a small number of agents; that will probably work but only display a small number of vehicles/agents.)
- Go to http://svn.vsp.tu-berlin.de/repos/shared-svn/projects/matsim-munich/scenarios/v2/output/2019-03-26-1pct-1it/ .
- Download
*.output_network.xml.gz
and*.output_events.xml.gz
. Best make sure that they do not uncompress, e.g. by "Download linked file as ...". - Get these files into VIA. This can be achieved in various ways; one is to open VIA and then drag the files from a file browser into VIA.
- Run VIA and enjoy.
For more information about the Munich scenario, see here: https://www.mos.ed.tum.de/en/tb/
For more information about MATSim, see here: https://www.matsim.org/.
Internal documentation can be found here: https://docs.google.com/document/d/133CuXaMuWL0NcstyFHodk4y7J5yFLy9nrLtbBVpNWzM/edit?usp=drive_link