[railML2] Do we need a solution to declare the states of rolling stock? [message #2582] |
Wed, 11 November 2020 10:57 |
Torben Brand
Messages: 162 Registered: March 2016
|
Senior Member |
|
|
Dear railML-community,
We have had a previous mishap where the wrong vehicle (and formation that is built on the vehicle) was used for runtime calculations. This as the same vehicle was in several versions in the same database/railML exchange file. The different versions were due to different states of the vehicle. In an early stage the vehicle was conceptual with "glossy marketing paper" values from the vendor. At a later stage the vehicle was planned and the values dropped/got more realistic. Finally, when the vehicle was in operation the values where investigated and found to be even lower.
To mitigate this source of potential error and to transmit more information in railML should we consider adding a construct like <states> under infrastructure (example: https://wiki2.railml.org/wiki/IS:state) in rolling stock?
Or do you suggest it's enough to demand from the user to transmit information outside railML of which correct vehicle to be used, to describe the vehicle further in the human-readable @description attribute or to purge the dataset transmitted on the desired state of vehicles (for instance only operational vehicles in data set)?
What does the community think? Any feedback is highly appreciated.
Kind regards
Torben Brand
Jernbanedirektoratet
|
|
|
|
Re: [railML2] Re: Do we need a solution to declare the states of rolling stock? [message #2811 is a reply to message #2810] |
Wed, 18 August 2021 16:07 |
Joerg von Lingen
Messages: 149 Registered: May 2011
|
Senior Member |
|
|
Dear all,
ticket #478 was created for the issue. A possible solution is included for
review in commits [1246] and [1247].
Basically the <state> element from infrastructure was re-used in rollingstock
for formation and vehicle. As a rule the information at formation shall
overwrite possible information in vehicle.
Best regards,
Joerg v. Lingen - Rollingstock Coordinator
On 18.08.2021 10:23, Carsten Weber wrote:
> Dear community,
> dear Torben,
>
> in my eyes your suggestion makes sense. Bound a data field like status to the
> vehicle so that you can check whether it is one current in use or it is only an
> idea or whatever. Maybe it should be checked before whether all entries of the
> infrastructure field could be taken one by one.
>
> Best regards.
>
> Carsten Weber
> Bahnberatung Carsten Weber
[Updated on: Wed, 18 August 2021 22:25] by Moderator Report message to a moderator
|
|
|