Speed assumptions cycling-electric and cycling-regular in QGIS

Simon.delrue · March 15, 2020, 10:38am

Hi all

For my master thesis I am researching the potential of e-bikes (up to 25 km/h) in combination with train travel in the province of Antwerp (Belgium) . Therefore, I would like to calculate catchment areas (isochrones) for train stations in Antwerp, for both regular cycling and electric cycling. However, I will need to review (and change if necessary) the speed assumptions for these modes.

I understand the ORS tool which I am using in QGIS is opensource, so I think it should be possible to do this. However, I am new to coding and I am not finding sufficient information online which could help me.

Is there anyone who could explain me how to find these assumptions and change them?

You would be a lifesaver!

Thanks!

nils · March 15, 2020, 11:50pm

Hi,

nope, sorry, that’s not possible that easily. What you’re using in QGIS is merely a client to request ORS, it doesn’t do much really.

Unfortunately, you can’t set the speed for bike dynamically in ORS. So you’d have to set up ORS yourself and change a few things. Easiest is to set up via Docker:

You can find the speed assumptions in the flag encoders, e.g.

github.com

GIScience/openrouteservice/blob/870d4dfc78efe27cccff934ab248140e4570748c/openrouteservice/src/main/java/org/heigit/ors/routing/graphhopper/extensions/flagencoders/bike/ElectroBikeFlagEncoder.java#L21


      
          *  You should have received a copy of the GNU Lesser General Public License along with this library; 
          *  if not, see <https://www.gnu.org/licenses/>.  
          */
          package org.heigit.ors.routing.graphhopper.extensions.flagencoders.bike;
          
          
import com.graphhopper.util.PMap;
          import org.heigit.ors.routing.graphhopper.extensions.flagencoders.FlagEncoderNames;
          
          
public class ElectroBikeFlagEncoder extends CommonBikeFlagEncoder
          {
             private static final int MEAN_SPEED = 20;
          
          
   public ElectroBikeFlagEncoder(PMap properties )
             {
                 this((int) properties.getLong("speed_bits", 4) + (properties.getBool("consider_elevation", false) ? 1 : 0),
                         properties.getLong("speed_factor", 2),
                         properties.getBool("turn_costs", false) ? 1 : 0, properties.getBool("consider_elevation", false));
                 this.properties = properties;
                 this.setBlockFords(properties.getBool("block_fords", true));
             }

However, speed depends on a lot things, like road class/surface etc You can set whatever speed you like and then recompile the project (i.e. build the Docker image) and run ORS locally.

I recommend you get some help from someone with a little more experience, this is fairly involved.

Simon.delrue · March 16, 2020, 10:10am

Hi Nils

Thanks for your answer! As I am new to coding, is there somewhere I can find a summary of the speed assumptions for the different modes/road class/surface etc.?
And do you by any chance know on what these assumptions were based?

Thank you for helping me!

amandus · March 16, 2020, 5:04pm

@Simon.delrue,

for assumptions check this doc page or the cycling flagencoders in the sourcecode as nils mentioned already.

The guys that came up with these are cycling enthusiasts themselves so they have a pretty good understanding on how fast you can go on what surface/waytype. But basically this is just roughly mapping logical values to the ways where no speed information is given.
By no means should you regard this as the exact speed with which you can travel on the different waytypes!
It is a classification to help returning reasonable times/distances.

Best regards