Re: [sumo-user] Building a SUMO scenario

["Dominik S. Buse" <buse@xxxxxxxxxxxx>]

Hi Anders,

I can second what Lara said about building a scenario. I also built a
scenario of a city of around 150.000 people and the time investment was
about one year. Currently, I am preparing it for release which also
means getting it re-validated with current sumo versions.

While the conversion, adaption and validation of the road network may be
tedious (highly depending on the quality of the OpenStreetMap data for
your area), I found the generation and validation of the traffic demand
much more challenging. Think early on about what kind of traffic you
need and what input data you have for it. Lara's papers give a good
intro, if you plan to use the ACTIVITYGEN approach. But in the end it
really depends on the data traffic data you have about your city.

So if you like, you can put me on CC for your mails. Maybe I can
contribute some aspects, too.

Kind regards,
Dominik

On 28/02/2019 09.49, Lindman, Anders wrote:
> Hello Lara, 
>
> Thanks for your reply, I will send you an email shortly. 
> If anyone else has any experience with scenario building I would really appreciate if you would like to share it.
>
> Best regards, 
> Anders 
>
> -----Original Message-----
> From: Lara CODECA [mailto:lara.codeca@xxxxxxxxxx] 
> Sent: den 27 februari 2019 13:30
> To: Lindman, Anders <anders.lindman@xxxxxxxxxxxxx>
> Cc: Sumo project User discussions <sumo-user@xxxxxxxxxxx>
> Subject: Re: [sumo-user] Building a SUMO scenario
>
> Hello Anders,
>
> This scenario https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flcodeca%2FLuSTScenario&amp;data=02%7C01%7Canders.lindman%40volvocars.com%7C6aa1912de3634169d35d08d69caf5698%7C81fa766ea34948678bf4ab35e250a08f%7C0%7C0%7C636868674219412240&amp;sdata=hLLT38DL7bJ0Iy3ddo63%2Feq%2FFCrjIjBk%2FE%2FmRMYWUeM%3D&amp;reserved=0 was among the outputs of my PhD, it took me 2 years, I tried to maintain it for an additional 2 years, and now its obsolete compared with SUMO functionalities.
>
> This scenario https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Flcodeca%2FMoSTScenario&amp;data=02%7C01%7Canders.lindman%40volvocars.com%7C6aa1912de3634169d35d08d69caf5698%7C81fa766ea34948678bf4ab35e250a08f%7C0%7C0%7C636868674219412240&amp;sdata=Kn1p6oRVn8x7nzeimBjK4JKgcYnRhVpkJi3e6BL1fs8%3D&amp;reserved=0 took me 1 year and a half, it's up to date with SUMO features, and I'm going to try to maintain it for as long as I can. 
> This time I shared generation and mobility tools that are meant to be used in order to generate general purpose scenarios starting from OSM-Like data.
>
> Both are 1-(wo)man efforts, I have no idea about scenarios created by teams and how long it took to them.
>
> I'm not sure how much the community is interested in all the process, so I can just tell you: feel free to send and e-mail and I'm going to do my best to explain the journey.
>
> Cheers,
> Lara
>
> On Wed, 27 Feb 2019 11:52:12 +0000
> "Lindman, Anders" <anders.lindman@xxxxxxxxxxxxx> wrote:
>
>> Dear users,
>>
>> I am a fairly new to SUMO and traffic simulation in general and have spent a limited amount of time using the software but my impression so far is that a major effort is required in order to construct a realistic validated scenario for an urban area.
>> My intention in the long run is to create a scenario of a real city (a city with a population of at least 100k) but before I set out to do this it would be great to know roughly how much time that is required to do this, i.e., converting and modifying traffic networks from OSM, validating traffic demands, and so on.
>> Therefore, I would be really grateful if anyone of you who have done this would like to share how much time and effort you spent on building and maintaining a scenario, which parts of the implementation requires most time, if there were any unforeseen problems along the way, etc.
>>
>> Best regards,
>> Anders
>>
>
>
-- 
Dominik S. Buse M.Sc.
Distributed Embedded Systems (CCS Labs)
Heinz Nixdorf Institut, Paderborn University, Germany
http://www.ccs-labs.org/~buse/