Hello, in general, you cannot remove the dependency on traffic states (if there is an upstream jam on some approaches, vehicles will inevitably be delayed). Also, jtrrouter does not offer options for reducing randomness due to sampling from the turn-ratio distribution.
As an alternative, you could use turning count data with the routeSampler tool. This will give you exact counts within the specified time intervals. By varying the size of the time interval you can influence the maximum deviation from the desired ratios (i.e. 50 in 100 vehicles turning right or 5 in every 10 vehicles turning right). See https://sumo.dlr.de/docs/Tools/Turns.html#routesamplerpy regards, Jakob Am Di., 28. Apr. 2020 um 18:06 Uhr schrieb Tilg, Gabriel < [email protected]>: > Hello all, > > > > I am aiming to use the JTR router in order to route flows according to > turning ratios. I noticed that the actual turning ratios are dependent of > traffic states and include stochasticity. > > > > For example, I created a network (~24 nodes), created some origins (~10), > and defined the turning ratios. They are either 100, 50 or 25%. So, I was > expecting every second or fourth vehicle to turn, but apparently the > turning movements are not that deterministic. Additionally, there seems to > exist a relation to congestion. > > > > Is there a way to reduce both the dependency on traffic states, and the > stochasticity? > > > > Thank you very much in advance! > > > > Best regards, > > Gabriel > _______________________________________________ > sumo-user mailing list > [email protected] > To unsubscribe from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
