Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] weight update inteval
If your scenario is prone to jamming, a smaller adaptation interval allows vehicles to react more quickly to developing jams and route around them. This can avoid large-scale jams and thereby reduce the time spent by each vehicle in the simulation (and thus eliminate subsequent rerouting trials).
You cannot compare routing algorithm duration without regard to the number of times rerouting was triggered.
Furthermore, fastest routes tend to become longer in a congested network which also impacts the duration spent computing routes.

Am Mi., 10. Juni 2020 um 01:19 Uhr schrieb Raheleh Zarei <raheleh.zarei@xxxxxxxxx>:
That is also what I expected, but for a scenario--a city map with about 1700 nodes, 100 random trips, and CH as routing algorithm-- duration is less when the interval is set to 130 than when it is 210.
For Dijkstra and Astar also, it doesn't follow this rule. (smaller interval, smaller duration). Is there any specific reason for this? or duration is not a solid performance parameter to use for comparison?

Thanks

On Sun, Jun 7, 2020 at 3:19 AM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
Yes. If device.rerouting.adaptation-interval is small, then CHRouter will spend more time updating itself and this will slow down the simulation.

Am Sa., 6. Juni 2020 um 16:10 Uhr schrieb Raheleh Zarei <raheleh.zarei@xxxxxxxxx>:
So, Can I say when the interval is a small number, then the duration variable, in output, will be bigger, because of updating weights in each interval? Also, as a consequence, the real-time factor will be smaller (simulation time doesn't change that much, as I noticed)?

On Wed, Jun 3, 2020 at 2:05 AM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
At the given interval, the weights are updated and those are used automatically by Astar and Dijkstra. The reason for 'myValidUntil' is the unique need to by CHRouter to perform pre-processing once the weights change.

Am Di., 2. Juni 2020 um 17:58 Uhr schrieb Raheleh Zarei <raheleh.zarei@xxxxxxxxx>:
Thanks.
So, Could we say that "myValidUntil" parameter in CHRouter.h file, is set based on the value given to the option --device.rerouting.adaptation-interval?
How this is used in Astar and Dijkstra?

On Sat, May 16, 2020 at 2:46 PM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:

Am Sa., 16. Mai 2020 um 20:16 Uhr schrieb Raheleh Zarei <raheleh.zarei@xxxxxxxxx>:
Thanks.
May I know how this travel time is calculated?And when rerouting happens?

On Sat, May 16, 2020 at 1:28 PM Jakob Erdmann <namdre.sumo@xxxxxxxxx> wrote:
It works for all routing algorithms. You can visualize the current weights in SUMO-GUI (color by edges "by routing device assumed speed") and you can also save the values to xml using option --device.rerouting.output FILE



Am Sa., 16. Mai 2020 um 18:10 Uhr schrieb Raheleh Zarei <raheleh.zarei@xxxxxxxxx>:
Hello,

Is the option  --device.rerouting.adaptation-interval  works for all the routing algorithms or just CH? I can't track its effect in DijkstraRouter.h. and AstarRouter.h.

Thanks,
Raheleh
--
RAZ
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user


--
RAZ
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user


--
RAZ
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user


--
RAZ
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user


--
RAZ
_______________________________________________
sumo-user mailing list
sumo-user@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user

Back to the top