Home » railML newsgroups » railml.timetable » Questions about "Category" and "Rosterings"
Questions about "Category" and "Rosterings" [message #1264] Thu, 04 September 2014 09:10 Go to next message
yutaka.manchu is currently offline  yutaka.manchu
Messages: 11
Registered: September 2014
Junior Member
Dear All,

Would you mind answering the following questions for me

1. Category
Regarding to the wiki, category means
'“train type” or “train category”. Depending on the system, these
terms are defined quite differently.'
( httpwiki.railml.orgindex.phptitle=TTcategory ).
And category can express “trainUsage”, “deadrun” and
“categoryPriority”.

But in TT_S-Bahn_ZH.xml, the category instance has only following data;

categories
category id=d2e125
categories

Could I get a sample file of category with “trainUsage”, “deadrun”
and “categoryPriority”


2. Rostering

2.1 blockPart - trainPartRef

According to the wiki,
trainPartRef This refers to the id attribute of the associated trainPart
element.
( httpwiki.railml.orgindex.phptitle=TTblockPart )

I'd like to know the relationship between JoinSplit operation and the
amount of trainPart. If there is a JoinSplit, could a blockPart have
multiple trainPart

If somebody has a example file that has both JoinSplit operation, please
send it to me.

2.2 block - blockGroupNumber

There's no precise description that explains blockGroupNumber on the wiki;
httpwiki.railml.orgindex.phptitle=TTblock

The role of “blockGroupNumber” is uncertain for me, even regarding
TT_Rostering_edit.xml as follows;

block id=d2e1628 blockGroupNumber=1
block

block id=d2e1672 blockGroupNumber=1
block

block id=d2e1711 blockGroupNumber=1
block

I’d like to know the meaning that “d2e1628”, “d2e1672” and
“d2e1711” belong to same “blockGroupNumber”.

1) What does blockGroupNumber mean
2) Is blockGroupNumber mandatory attribute

Best regards,
Utah (Yutaka Manchu)

--
----== posted via PHP Headliner ==----
Re: Questions about "Category" and "Rosterings" [message #1265 is a reply to message #1264] Thu, 11 September 2014 16:51 Go to previous messageGo to next message
Andreas Tanner is currently offline  Andreas Tanner
Messages: 51
Registered: March 2012
Member
Dear Utah,

welcome to the railML community. You are asking interesting questions!

Regarding category, the semantics is not completely defined within the
standard. Our customers use the train category mostly for distinguishing
local / regional / long distance etc. passenger trains and freight
trains. Transfer trips (purely operational trains that run outside
stations and do have an official train path) would use the "deadRun"
attribute.
Our system models different vehicle preparation time requirements and
staff rules depending on the train category.

We don't use categoryPriority. I would expect it to have integer value
as in the wiki example given, but it is defined as string.

Your questions on join/split operations is difficult. railML uses trains
http://wiki.railml.org/index.php?title=TT:train to express joins and
splits. One might criticize this approach for its implication that one
needs to construct trains to model vehicle shunting. Syntactically it is
possible to put multiple blockPartRefs into one blockPartSequence, but I
have never seen such data and it would be ambiguous if the number of
blockPartRefs changes from blockPartSequence to blockPartSequence.

The attribute blockGroupNumber is optional. Our system uses it for
defining the line in the roster matrix representation of the circulation.

I hope this helps. Best,

--Andreas.

Am 04.09.2014 09:10, schrieb Yutaka Manchu:
> Dear All,
>
> Would you mind answering the following questions for me
>
> 1. Category
> Regarding to the wiki, category means
> '“train typeâ€
Re: Questions about "Category" and "Rosterings" [message #1266 is a reply to message #1265] Fri, 12 September 2014 07:14 Go to previous messageGo to next message
yutaka.manchu is currently offline  yutaka.manchu
Messages: 11
Registered: September 2014
Junior Member
Dear Andreas,

Thank you very much for replying to my questions. :-)

I understood your answers.

About Split / Join, there're examples that don't have data under
/trainPartRef/s;
http://wiki.railml.org/index.php?title=TT:trainCouplingAndSh aring

May I confirm you if where the Split / Join occurs (i.e.
"Lille","SanAntonio" in above examples) are known from the /ocpRef/ of
<octTT>s of <trainPart> refered from the /trainPartRef/ or not?
If you have a "full railML timetable file" that contains Join / Split,
would you mind sending it for me?

Best regards,
Utah (Yutaka Manchu)

Andreas Tanner wrote:
>
> Dear Utah,
>
> welcome to the railML community. You are asking interesting questions!
>
> Regarding category, the semantics is not completely defined within the
> standard. Our customers use the train category mostly for distinguishing
> local / regional / long distance etc. passenger trains and freight
> trains. Transfer trips (purely operational trains that run outside
> stations and do have an official train path) would use the "deadRun"
> attribute.
> Our system models different vehicle preparation time requirements and
> staff rules depending on the train category.
>
> We don't use categoryPriority. I would expect it to have integer value
> as in the wiki example given, but it is defined as string.
>
> Your questions on join/split operations is difficult. railML uses trains
> http://wiki.railml.org/index.php?title=TT:train to express joins and
> splits. One might criticize this approach for its implication that one
> needs to construct trains to model vehicle shunting. Syntactically it is
> possible to put multiple blockPartRefs into one blockPartSequence, but I
> have never seen such data and it would be ambiguous if the number of
> blockPartRefs changes from blockPartSequence to blockPartSequence.
>
> The attribute blockGroupNumber is optional. Our system uses it for
> defining the line in the roster matrix representation of the circulation.
>
> I hope this helps. Best,
>
> --Andreas.
>
> Am 04.09.2014 09:10, schrieb Yutaka Manchu:
>> Dear All,
>>
>> Would you mind answering the following questions for me
>>
>> 1. Category
>> Regarding to the wiki, category means
>> '“train type” or “train category”. Depending on the system, these
>> terms are defined quite differently.'
>> ( httpwiki.railml.orgindex.phptitle=TTcategory ).
>> And category can express “trainUsage”, “deadrun” and
>> “categoryPriority”.
>>
>> But in TT_S-Bahn_ZH.xml, the category instance has only following data;
>>
>> categories
>> category id=d2e125
>> categories
>>
>> Could I get a sample file of category with “trainUsage”, “deadrun”
>> and “categoryPriority”
>>
>>
>> 2. Rostering
>>
>> 2.1 blockPart - trainPartRef
>>
>> According to the wiki,
>> trainPartRef This refers to the id attribute of the associated trainPart
>> element.
>> ( httpwiki.railml.orgindex.phptitle=TTblockPart )
>>
>> I'd like to know the relationship between JoinSplit operation and the
>> amount of trainPart. If there is a JoinSplit, could a blockPart have
>> multiple trainPart
>>
>> If somebody has a example file that has both JoinSplit operation, please
>> send it to me.
>>
>> 2.2 block - blockGroupNumber
>>
>> There's no precise description that explains blockGroupNumber on the
wiki;
>> httpwiki.railml.orgindex.phptitle=TTblock
>>
>> The role of “blockGroupNumber” is uncertain for me, even regarding
>> TT_Rostering_edit.xml as follows;
>>
>> block id=d2e1628 blockGroupNumber=1
>> block
>>
>> block id=d2e1672 blockGroupNumber=1
>> block
>>
>> block id=d2e1711 blockGroupNumber=1
>> block
>>
>> I’d like to know the meaning that “d2e1628”, “d2e1672” and
>> “d2e1711” belong to same “blockGroupNumber”.
>>
>> 1) What does blockGroupNumber mean
>> 2) Is blockGroupNumber mandatory attribute
>>
>> Best regards,
>> Utah (Yutaka Manchu)
>>
>
>
>



--
----== posted via PHP Headliner ==----
Re: Questions about "Category" and "Rosterings" [message #1267 is a reply to message #1266] Fri, 12 September 2014 17:05 Go to previous messageGo to next message
Andreas Tanner is currently offline  Andreas Tanner
Messages: 51
Registered: March 2012
Member
Am 12.09.2014 07:14, schrieb Yutaka Manchu:
>
> About Split / Join, there're examples that don't have data under
> /trainPartRef/s;
> http://wiki.railml.org/index.php?title=TT:trainCouplingAndSh aring
>
> May I confirm you if where the Split / Join occurs (i.e.
> "Lille","SanAntonio" in above examples) are known from the /ocpRef/ of
> <octTT>s of <trainPart> refered from the /trainPartRef/ or not?

yes, the location of the split/join is at the first / last ocpTT of the
referenced trainParts.
The following are assumed:
- the sequences of ocpTT of all trainParts within one trainPartSequence
coincide (at least when the operation periods are overlapping)
- the sequence of ocpTT within a train, that is, along the
trainPartSequences, form a valid train path. This means that there are
no location breaks and the times are ascending.

> If you have a "full railML timetable file" that contains Join / Split,
> would you mind sending it for me?
Sent per mail.

Best, Andreas.
Re: Questions about "Category" and "Rosterings" [message #1268 is a reply to message #1267] Tue, 16 September 2014 07:24 Go to previous message
yutaka.manchu is currently offline  yutaka.manchu
Messages: 11
Registered: September 2014
Junior Member
Dear Andreas,

Thank you for teaching me how to recognize the point of Join / Split.
Your valid condition description helps me very much. :-)

> Sent per mail.
Thank you, too.
There isn't any trainPartSequence that sequence isn't "1" in your sample
railML file, and every point of Join / Split in it seems to be either
start ocp or end ocp of a train.
If there is a trainPartSequence that sequence isn't "1", it means there
is either Join / Split even in the course of the train as
http://wiki.railml.org/index.php?title=TT:trainCouplingAndSh aring examples
and we can find out the point with corresponding ocpTTs. If you send me
another railML example file that has some Join / Split even in the course
of some trains, I appreciate it. ;-)

Best regards,
Utah (Yutaka Manchu)

Andreas Tanner wrote:
>
> Am 12.09.2014 07:14, schrieb Yutaka Manchu:
>>
>> About Split / Join, there're examples that don't have data under
>> /trainPartRef/s;
>> http://wiki.railml.org/index.php?title=TT:trainCouplingAndSh aring
>>
>> May I confirm you if where the Split / Join occurs (i.e.
>> "Lille","SanAntonio" in above examples) are known from the /ocpRef/ of
>> <octTT>s of <trainPart> refered from the /trainPartRef/ or not?
>
> yes, the location of the split/join is at the first / last ocpTT of the
> referenced trainParts.
> The following are assumed:
> - the sequences of ocpTT of all trainParts within one trainPartSequence
> coincide (at least when the operation periods are overlapping)
> - the sequence of ocpTT within a train, that is, along the
> trainPartSequences, form a valid train path. This means that there are
> no location breaks and the times are ascending.
>
>> If you have a "full railML timetable file" that contains Join / Split,
>> would you mind sending it for me?
> Sent per mail.
>
> Best, Andreas.
>
>



--
----== posted via PHP Headliner ==----
Previous Topic: Questions about "Category" and "Rosterings" of timetable
Next Topic: Aspects of timetable 3.0
Goto Forum:
  


Current Time: Thu May 25 18:06:16 CEST 2017