Home » railML newsgroups » railml.common » [Advanced example v14] Infrastructure managers
[Advanced example v14] Infrastructure managers [message #3895] Mon, 16 February 2026 23:24 Go to next message
Mathias Vanden Auweele is currently offline  Mathias Vanden Auweele
Messages: 143
Registered: February 2025
Location: Brussels
Senior Member
The advanced example defines 4 Infrastructure managers in the xml but only three are displayed in the pdf.

Was <organizationalUnit code="BN" id="im01"> intended? If no, then the asset set afi_01 needs to be reassigned.

There is some confusion to me which elements should have a @infrastructureManagerRef and what needs to be assumed when an element that can hold this attribute, doesn't have one. Do we infer the first IM?

For example, track with id="trc34" is missing the infrastructureManagerRef attribute. I assume this is an error in the example file?

But also the Operational points and the lines don't have the @infrastructureManagerRef ?


Mathias Vanden Auweele
Railway data freelancer
https://matdata.eu
Brussels, Belgium
Re: [Advanced example v14] Infrastructure managers [message #3899 is a reply to message #3895] Wed, 18 February 2026 11:28 Go to previous messageGo to next message
Larissa Zhuchyi is currently offline  Larissa Zhuchyi
Messages: 107
Registered: November 2022
Senior Member
Dear Mathias

Quote:
Do we infer the first IM?
railML3 Wiki guide was extended with missing clause https:// wiki3.railml.org/wiki/Dev:FileConventions#How_to_interpret_m issing_attributes_and_elements_in_the_file? . This approach is inherited from railML2. Thanks for pointing this out.

railML.org will reach you to answer the rest of you questions after an internal discussion.

Sincerely,


Larissa Zhuchyi – Ontology Researcher
railML.org (Registry of Associations: VR 5750)
Altplauen 19h; 01187 Dresden; Germany www.railML.org
Re: [Advanced example v14] Infrastructure managers [message #3902 is a reply to message #3899] Sun, 22 February 2026 11:48 Go to previous messageGo to next message
Mathias Vanden Auweele is currently offline  Mathias Vanden Auweele
Messages: 143
Registered: February 2025
Location: Brussels
Senior Member
Hello Larissa,

Thank you for your first answer and the explicit declaration of it in the wiki!

Although this raises another set of questions. So railML is applying the 'open world' principle where if something is not stated, it is assumed to be unknown.
-> So we can never say, "here's the complete list of Operational points of a network" by transmitting a railML file.
-> Or, in the case of the IM ref, an OP can have an infinite amount of IM's.

So how do we 'close' the lists than in railML? How do we state that something is complete?


Mathias Vanden Auweele
Railway data freelancer
https://matdata.eu
Brussels, Belgium
Re: [Advanced example v14] Infrastructure managers [message #3910 is a reply to message #3895] Mon, 23 February 2026 17:35 Go to previous messageGo to next message
Larissa Zhuchyi is currently offline  Larissa Zhuchyi
Messages: 107
Registered: November 2022
Senior Member
Dear Matthias

Thanks for your patience till the decision was made at internal discussion on modelling meeting 2026-02-23.

Following the rule "If an attribute or an element is missing in a railML file, it shall be read as "not known"" at https://wiki3.railml.org/wiki/Dev:FileConventions#How_to_int erpret_missing_attributes_and_elements_in_the_file? user shall explicitly state IM for line, operationalPoint and track if information is known.

Quote:
The advanced example defines ... track with id="trc34" ... missing the infrastructureManagerRef attribute ... (and) Operational points and the lines
infrastructureManagerRef is indeed missing for id="trc34", Operational points and lines in XML file. This will be corrected and published in fixed version.

Additionally railML.org filed an issue related to this problem https://development.railml.org/railml/version3/-/issues/707 .

Sincerely,


Larissa Zhuchyi – Ontology Researcher
railML.org (Registry of Associations: VR 5750)
Altplauen 19h; 01187 Dresden; Germany www.railML.org

[Updated on: Mon, 23 February 2026 17:35]

Report message to a moderator

Re: [Advanced example v14] Infrastructure managers [message #4032 is a reply to message #3910] Fri, 12 June 2026 10:53 Go to previous messageGo to next message
christian.rahmig is currently offline  christian.rahmig
Messages: 566
Registered: January 2016
Senior Member
Dear Mathias,
dear all,

from technical perspective, referencing an infrastructure manager is now possible from every infrastructure element using new child element <elementInfrastructureManagerRef> in any functional infrastructure element. If the information is not given, the general infrastructure manager specified in <infrastructure/infrastructureManagerRef> applies. The whole concept, which will implemented in railML 3.4, is described in Gitlab issue #707 [1].

[1] https://development.railml.org/railml/version3/-/issues/707

Best regards
Christian


Christian Rahmig – Infrastructure scheme coordinator
railML.org (Registry of Associations: VR 5750)
Altplauen 19h; 01187 Dresden; Germany www.railML.org
Re: [Advanced example v14] Infrastructure managers [message #4034 is a reply to message #4032] Mon, 15 June 2026 09:19 Go to previous message
Mathias Vanden Auweele is currently offline  Mathias Vanden Auweele
Messages: 143
Registered: February 2025
Location: Brussels
Senior Member
Thank you Christian. I see my report about confusion in the advanced example triggered a small remodelling of the way how infrastructure managers are defined. Thank you.

From a data consistency point of view, it would make sense to give an <infrastructureManager> element the possibility to define it's own <areaLocation> or <networkLocation>. Every infrastructure element that falls inside of this area is then attributed to the IM. <areaLocations> of IM's can overlap when required and since they can have validity ranges, the relationship with the infrastructure is also time based. If the IM applies to the whole file, the IM element can just use <networkLocation>. If this reasoning is followed, then the current modelling doesn't need to be deprecated as it can still be useful for a single element that overrides it's location based IM, when it is inside of the areaLocation of the IM to belong to another IM. Or for infrastructureElements without location.

RTM was not only designed to place <functionalInfrastructure> on the network but also for things like IM's.


Mathias Vanden Auweele
Railway data freelancer
https://matdata.eu
Brussels, Belgium
Previous Topic: Extending railML: multiple domain specific Extensions for the same Elemen
Next Topic: Designator for linearPositioningSystem
Goto Forum:
  


Current Time: Mon Aug 10 06:13:47 CEST 2026