Hello,
this is not supported in version 1.12.0
However, in the latest development version you can define
<flow id="a" begin="0" end="3600" period="exp(0.5)" .../>
to achieve arrivals corresponding to a Poisson process (number of vehicles
follows poisson distribution).
You can download and updated development version starting *tomorrow* at
https://sumo.dlr.de/docs/Downloads.php#sumo_-_latest_development_version

regards,
Jakob

Am Di., 8. März 2022 um 11:47 Uhr schrieb Yajun She <[email protected]>:

> Dear Jakob,
>
> Is any way that I can define a flow that follows the Poisson distribution
> and at the same time control its traffic volume in each hour? Because I
> want to investigate the impacts of different demand ratios on intersection
> approaches.
>
> Thanks,
> Yajun
>
> On Mon, 7 Mar 2022 at 19:31, Jakob Erdmann <[email protected]> wrote:
>
>> When using a a flow with 'probability', insertion follows a binomial
>> distribution. Your example should be a good fit to a poisson distribution.
>> See here, for rules of thumb on approximation:
>> https://en.wikipedia.org/wiki/Binomial_distribution#Poisson_approximation
>>
>> Note that you cannot not set 'end' and 'number' at the same time when
>> using 'probability'. Either you set a fixed number of random trials ('end')
>> or you let the flow run until it has emitted a fixed number of vehicles
>> ('number').
>>
>> Am Mo., 7. März 2022 um 16:35 Uhr schrieb Yajun She <[email protected]
>> >:
>>
>>> Dear all,
>>>
>>> How to model a flow that a vehicle will be emitted following a Poisson
>>> distribution, and at the same time with a given traffic volume?
>>>
>>> For example, I tried to use attributes of "probability" and "number":
>>> <routes>
>>>   <vType id="normal_car" vClass="passenger" departspeed="13.89"
>>> maxSpeed="13.89" />
>>>
>>>   <flow id="upflow" type="normal_car" begin="0" end="5000"
>>> probability="0.01" number="1200" from="downin" to="upout" departPos="last"/>
>>> </routes>
>>>
>>> I am not sure the range of probability that will lead the vehicles to
>>> follow a Poisson distribution?
>>>
>>> Thanks,
>>> Yajun
>>>
>>>
>>>
>>> _______________________________________________
>>> 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

Reply via email to