Hi all, I simulate heavily congested networks, and I found the results to be sensitive on the max-depart-delay parameter. This makes sense as in many cases the links are probably so congested that it cannot insert any additional vehicle, and they are backlogged for the max-depart-delay time.
Is my understanding correct: If I have a network with many taz, where N_edges>>1 per taz: - using OD2trips: a vehicle generation is random on any edge in the taz - using device.rerouting.with-taz is NOT RANDOM but the edges which minimize travel times between the two taz are selected (according to doc: https://sumo.dlr.de/docs/Definition_of_Vehicles,_Vehicle_Types,_and_Routes.html#traffic_assignement_zones_taz). Does that mean the insertion edge depends on the actual travel time derived from the last simulation step or is it just travel distance? I believe that in my case, this leads to only few links actually generating vehicles, instead of the whole TAZ, which creates unrealistic results. It would be great if device.rerouting.with-taz would also insert vehicles randomly inside the taz. I find it more convenient to run device.rerouting.with-taz and not to run OD2trips every time I'd like some variation in my ODs. Or is there something, I am missing about the advantages of a non-random trip insertion in device.rerouting.with-taz? As always, many thanks! Lukas
_______________________________________________ sumo-user mailing list sumo-user@eclipse.org To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
