It's hard to tell without seeing the traffic patterns but there could be complex interactions due to intersections and whatever the rerouters are doing. Observing the scenario while it runs is usually the best way to figure these things out. It also helps to set vehicle coloring to "by speed".
Am Do., 22. Dez. 2022 um 09:51 Uhr schrieb 191 000 <[email protected] >: > I have attached a screenshot for your reference. > > On Thu, Dec 22, 2022 at 2:12 PM 191 000 <[email protected]> wrote: > >> I have rechecked all my edges, and all the units are in m/s. The >> simulation scenario is for 39600-time steps, where initially vehicles enter >> from one edge and get out at another, travelling with maximum allowable >> speed but only using some of the edges of the network. After about 3600 >> timesteps, a few more vehicles enter, but now they use all the edges and >> may slow down or stop based on traffic conditions. In this scenario, the >> average speed for initial 3600 timesteps, the average speed is around >> 3-4m/s, while after 3600, it goes up to 7-8m/s, which I find >> counterintuitive. Hence I tried matching it with edgeData file values. >> >> On Thu, Dec 22, 2022 at 2:04 PM Jakob Erdmann <[email protected]> >> wrote: >> >>> This sounds like a km/h vs m/s mismatch. If you define your edges with >>> 16km/h than the average speed would typically be a bit lower i.e. ~14 km/h >>> (due to random slow-downs and interactions of vehicles with different >>> desired speeds). This would then be presented to you by sumo as ~3.8 m/s . >>> >>> Am Do., 22. Dez. 2022 um 06:45 Uhr schrieb 191 000 < >>> [email protected]>: >>> >>>> Dear community, >>>> Regarding the previous discussion on average speeds in SUMO, I have >>>> encountered another setback. Earlier I was using SUMO-GUI to generate a >>>> graph of the average speed of the whole network (at 5-minute intervals) and >>>> save it as a CSV file. Then I thought that if I got aggregated average >>>> speed of all the edges in the network, it should be equal to the values >>>> obtained from SUMO-GUI, but this is not the case. Could you explain how the >>>> average speed for the whole network is calculated and how I can get the >>>> same values through the edgeData file? >>>> I wanted to compare these values as even though I had a max allowable >>>> speed of 16 and the vehicles were free-flowing, the average speed in >>>> SUMO-GUI would be around 3-4. >>>> Thanks and regards, >>>> Rohan >>>> >>>> On Tue, Dec 13, 2022 at 2:11 PM Jakob Erdmann <[email protected]> >>>> wrote: >>>> >>>>> The speed value is computed as travelledDistance / sampledSeconds. >>>>> You should be able to replicate the aggregated speed value by the >>>>> following procedure: >>>>> - for each edge compute travelled distances as speed * sampledSeconds >>>>> - sum up the distances and sampledSeconds >>>>> - take the quotient of the sums >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> Am Di., 13. Dez. 2022 um 09:23 Uhr schrieb 191 000 < >>>>> [email protected]>: >>>>> >>>>>> Dear community, >>>>>> >>>>>> I have been trying to understand how the average speed is >>>>>> calculated in edge-based traffic measures. I know it is space-mean speed >>>>>> but I am trying to aggregate data for multiple edges. >>>>>> I first generated average speed of two edges using edges="" and >>>>>> aggregate="True" method, then I generated average speed of the two edges >>>>>> separately. Now, I have tried arithmetic mean, harmonic mean, weighted >>>>>> average by numbers but cannot get the value equal to the aggregated value >>>>>> given by SUMO. >>>>>> Kindly suggest in what way can I manually aggregate the average speed >>>>>> data so as to get same value as when being done by aggregating directly >>>>>> through SUMO. >>>>>> >>>>>> Thanks and regards, >>>>>> Rohan >>>>>> _______________________________________________ >>>>>> 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 >>>>> >>>> _______________________________________________ >>>> 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 >>> >> _______________________________________________ > 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
