Purgatory/SettingSimulationUp

generated on 2019-08-27 00:07:38.283623 from the wiki page for Purgatory/SettingSimulationUp for SUMO 1.3.1

Setting up a Simulation for SUMO

Needed Data

At first, you need the network the traffic to simulate takes place on. As SUMO is meant to work with large networks, we mainly concentrated our work on importing networks and the computation of needed values that are missing within the imported data. Due to this, no graphical editor for networks is available, yet. Beside information about a network's roads, information about traffic lights is needed.

Further, you need information about the traffic demand. While many traffic simulations use a statistical distribution which is laid over the network, each vehicle within SUMO knows its route, where "route" is a list of edges to pass.

You basically have to perform the following steps in order to implement a simulation scenario:

1. Build your network

This can be either done by a) generating an abstract network using NETGENERATE, b) setting up an own description in XML and importing it using NETCONVERT or by c) importing an existing road network using NETCONVERT; see building the networks for further information

2. Build the demand

This can be either done by a) describing explicit vehicle routes, b) using flows and turning percentages only, c) generating random routes, d) importing OD-matrices, or e) importing existing routes; see building the demand for further information

3a. If needed, compute the dynamic user assignment

3b. Calibrate the simulation using given measures

4. Perform the simulation to get your desired output;

See performing the simulation and available simulation outputs

This process is also visualised within the next figure.

Flow complete small.png

Figure 1.3. Process of simulation with SUMO (rounded: definite data types; boxes: applications; octagons: abstract data types)

Best Practice

Choosing a proper scenario depends on the done investigation. Abstract, generated networks may be the best solution if one wants to keep the results easily evaluable. Abstract networks have for example been used for evaluating the vulnerability of C2C-networks by Clemens Honomichl.

If one wants to investigate real life traffic, he should try to reuse the data he has. If no data is available, one could try to use the free scenarios available within SUMO.