Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [sumo-user] Inquiry about Traffic Delay as a Performance Measure
The suggestion by Pedro Oliveira is spot on!
I would add, that unexpected results may also arrise from mismatching assumptions about scenario behavior. 
There is no substitute for looking at simulations and visually confirming that vehicles behave as expected.
You can use edgeData output to pinpoint edges that see more timeLoss than the base case (https://sumo.dlr.de/docs/Tools/Output.html#edgedatadiffpy may help).
However, you should also look at slowed or waiting cars in the simulation to check whether cars are using the new lanes as expected or queues form in unexpected places.
One reason why this might be happening is due to inappropriate lane-to-lane connections.

regards,
Jakob

Am Sa., 13. Juni 2026 um 16:48 Uhr schrieb Pedro Oliveira via sumo-user <sumo-user@xxxxxxxxxxx>:
Good morning!

The SUMO developers may have a more appropriate answer than I can provide.

One thing you can observe is whether, in the unmodified scenario, you had any virtual queues, meaning vehicles that couldn't enter the simulation. You can find this information in the output statistics under "waiting." This is important to verify if, when you modified the network, you reduced the virtual queue and consequently had more vehicles circulating on the network.

This SUMO page has a recommendation regarding the comparison of proposals, which suggests that the best approach is the total travel time between proposals.

https://sumo.dlr.de/docs/Simulation/Output/StatisticOutput.html#fair_traveltime_comparison_between_simulations

Sds,

Pedro Oliveira

Não contém vírus.www.avast.com

Em sex., 12 de jun. de 2026 às 12:13, Lorain Salufu via sumo-user <sumo-user@xxxxxxxxxxx> escreveu:
Good morning,

I am a student using SUMO to evaluate the operational impacts of implementing Business Access and Transit (BAT) lanes along an urban arterial corridor. I would appreciate some guidance regarding the interpretation and calculation of traffic delay in my study.

My current methodology uses the timeloss attribute from tripinfo.xml as a measure of vehicle delay. For each simulation seed, I calculate the average timeLoss for buses and general traffic separately, then average these values across seeds.

However, I am encountering a result that I am struggling to interpret.

The BAT lane scenario was expected to improve transit performance by providing a dedicated curb lane shared only by buses and right-turning vehicles. When comparing the Build and No-Build scenarios, I observe the following:

  • Bus travel times improve in some corridor segments, particularly within the BAT treatment area.
  • Bus progression plots show fewer stops in certain locations.
  • However, average trip-level timeLoss increases for both buses and general traffic in several sections of the corridor.
  • The increase is observed consistently across multiple simulation seeds.

My questions are:

  1. Is tripinfo.timeLoss generally considered an appropriate measure of traffic delay for corridor-level performance evaluation in SUMO?
  2. Would you recommend using another metric (e.g., waiting time, travel time relative to free-flow travel time, edge-based delay, queue measures, or another output) when assessing the operational impacts of transit-priority treatments?

Any guidance, references, or examples would be greatly appreciated.

Thank you for your time and assistance.

Best regards,

Lorain.
_______________________________________________
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

Back to the top