• Lane- or Edge-based Traffic Measures

Values within this output describe the situation within the network in terms of traffic science by giving macroscopic values such as the mean vehicle speed, the mean density, etc. This output can be generated either for lanes, edges or groups of edges.

In the following, it is described how both outputs are generated and which values they contain. Then, the meanings of the values are given as well as a description of intervals. At last, some additional possibilities to constrain the outputs are given.

Instantiating within the Simulation#

Meandata Definition#

An edge-based state dump is defined within an additional-file:

<additional>
    <edgeData id="<MEASUREMENT_ID>" file="<OUTPUT_FILE>" .../>
</additional>

For a lane based dump simply write:

<additional>
    <laneData id="<MEASUREMENT_ID>" file="<OUTPUT_FILE>" .../>
</additional>

The defined file can either be loaded with sumo option --additional-files FILENAME or declared in a .sumocfg file with <additional-files value="FILENAME"/>

For additional attributes see the table below.

Simplified definition#

If none of the additional attributes listed above are needed, a meandata definition can be declared without defining an additional file by setting one of the following options:

• --edgedata-output FILENAME
• --lanedata-output FILENAME

The generated definitions will be named DEFAULT_EDGEDATA and DEFAULT_LANEDATA respectively and will aggregate their values over the whole length of the simulation (edges without traffic will not be included in the output).

Generated Output#

Edge-Based Network States#

For edge-based state dumps, the output file will look like the following:

<meandata>
    <interval begin="<INTERVAL_BEGIN>" end="<INTERVAL_END>" id="<MEASUREMENT_ID>">
      <edge id="<EDGE_ID>" sampledSeconds="<COLLECTED_VEHICLE_SECONDS>" \
            traveltime="<MEAN_TRAVEL_TIME>" \
            density="<MEAN_DENSITY>" occupancy="<MEAN_OCCUPANCY>" \
            waitingTime="<TOTAL_WAITING_TIME>" speed="<MEAN_SPEED>" \
            departed="<EMITTED_VEH_NUMBER>" arrived="<ARRIVED_VEH_NUMBER>" \
            entered="<ENTERED_VEH_NUMBER>" left="<LEFT_VEH_NUMBER>" \
            laneChangedFrom="NUMBER_OF_LANE_LEAVERS" laneChangedTO="NUMBER_OF_LANE_ENTERER"/>

      ... more edges ...

    </interval>

    ... further intervals ...

</meandata>

Please note, that in contrary to the example above, for each edge, all values are reported in one line.

Lane-Based Network States#

The generated output looks like the following:

<meandata>
    <interval begin="<INTERVAL_BEGIN>" end="<INTERVAL_END>" id="<MEASUREMENT_ID>">
      <edge id="<EDGE_ID>">
          <lane id="<LANE_ID>" sampledSeconds="<COLLECTED_VEHICLE_SECONDS>" \
                traveltime="<MEAN_TRAVEL_TIME>" \
                density="<MEAN_DENSITY>" occupancy="<MEAN_OCCUPANCY>" \
                waitingTime="<TOTAL_WAITING_TIME>" speed="<MEAN_SPEED>" \
                departed="<EMITTED_VEH_NUMBER>" arrived="<ARRIVED_VEH_NUMBER>" \
                entered="<ENTERED_VEH_NUMBER>" left="<LEFT_VEH_NUMBER>" \
                laneChangedFrom="<NUMBER_OF_LANE_LEAVERS>" laneChangedTo="<NUMBER_OF_LANE_ENTERER>" \
                speedRelative="<MEAN_SPEED_RELATIVE>" />

          ... more lanes...

      </edge>

      ... more edges ...

    </interval>

    ... further intervals ...

</meandata>

Please note, that in contrary to the example above, for each lane, all values are reported in one line.

Value Descriptions#

Both the edge-dump and the lane-dump are computing the values the same way: every vehicle move - even those with v=0 - is recorded and saved during the interval. After the interval has passed, these values are written into the file after being normalized.

With regard to edgeData density, the values are normalized by the number of the collected vehicle movements and the length of the lane. For laneDensity they are also normalized by the number of lanes of the edge.

The meanings of the written values are given in the following table.

Please note that in the case of edge meandata both laneChanged entries are equal to the total number of lane changes on the edge. Furthermore each lane change is counted separately, that means if one vehicle changes from lane 1 to lane 3 it will generate two lane change counts.

How to aggregate / derive further values#

In order to aggregate the output for the same edge / lane over larger time periods the values which count vehicles have to be simply summed, density and occupancy can be simply averaged (if the interval lengths are equal) and speed should be a weighted average with sampledSeconds as weight. To aggregate the traveltime further, calculate the mean speed first and then divide edge/lane length by the speed.

Spatial aggregation (over multiple edges or lanes) is not as easy, especially if they are consecutive, because each vehicle generates data on each lane it is on, even if it is on multiple lanes. In the case of many short and fast moving vehicles the error is negligible but if you have long and slow vehicles you cannot aggregate as above.

The following measurements can be derived from the values given (period denotes the length of the aggregation interval):

• Average number of vehicles on the edge (#) = sampledSeconds / period
• Average traffic volume (#/h) = speed * 3.6 * density
• Traffic volume at the begin of the lane / edge (#/h) = 3600 * entered / period
• Traffic volume at the end of the lane / edge (#/h) = 3600 * left / period
• Total distance travelled (m) = speed * sampledSeconds
• Edge length = sampledSeconds / period * 1000 / density

Sometimes one wants to know how many vehicles were on an edge. The exact definition of this value depends on how departed, arrived and lane-changing vehicles are counted. Each vehicle is either in arrived, left or laneChangedFrom and each vehicle is also either in departed, entered or laneChangedTo. However, a vehicle may be counted multiple times in laneChangedFrom and laneChangedTo if it performed back-and-forth lane changing.

The following <edgeData> values are all reasonable counts in that they count each vehicle only once but they include/exclude some special cases.

• entered
• entered + departed
• left
• left + arrived

Notes#

• Per default, all edges are written, even those on which no vehicle drove. If the lane / edge collected no data the attributes speed, traveltime, density, occupancy and waitingTime are not written. One can disable writing unused edges/lanes by setting the excludeEmpty attribute to true.
• Even on lanes/edges which have sampledSeconds="0.00" since the real number of sampledSeconds may be 0.001 and was cut off in the output. To define a minimum number of samples before you consider your data valid, use minSamples in the meandata definition.
• Even though no data was collected (sampledSeconds="0") departed and laneChangedTo may be different from 0 because internally lane changes and departures happen at the end of a time step.
• The interval end is the interval begin + aggregation time, meaning that values were collected within these steps excluding the end time itself. If the simulation ends before the last interval is over, the interval will be pruned.
• The output file will be generated, does not have to exist earlier and will be overwritten if existing without any warning. The folder the output file shall be generated in must exist.
• If you need only information about the network states during certain time periods, you may constraint generation of the dumps by creating multiple dumps with the attributes begin="" and end="". All of them may have identical ids and write to the same output file.

Visualization / Plotting#

• sumo-gui can load edgeData files and color network edges/lanes according to any of its attributes
• sumo-gui can color edges/lanes by any of the attributes being collected while the simulation is running
• netedit can be used to load/view/edit edgeData files
• The mpl_dump_onNet.py script can display values of this output as a colored net (and further visualization tools exist).
• plotXMLAttributes can extract and plot values such as the Fundamental Diagram below

See Also#

• edge/lane-based vehicular pollutant emission output and edge/lane-based vehicular noise emission output which have similar formats
• You can generate mean data definitions automatically. See output tools for more information.