RE: [plcs-dex] Questions regarding Next_assembly_usage

[Peter Bergström <peter.bergstrom@eurostep.com>]

Thanks Dave,

Regarding IDs I got a similar response in private today, saying that the uniqueness can only be guaranteed in relations between the parent and the child, not even within an exchange file.

And in order to reference a next_assembly_usage, therefore, you need to specify the parent ID and child ID (and maybe also their versions, i.e. Part_version if you want to be specific), and furthermore the ID of the next_assembly_usage if there are more than one (which there might be due to occurrencies, e.g. left/right).

To me, your reply and this other one indicates that we can never rely on unique IDs for the Next_assembly_usage, and consequently there is no reason in demanding it. If we do, we might get generated numbers with no business meaning outside of the parent_child relationship.

So I'll stay with an optional ID assignment.


Regarding location_indicator:
If this is just a hangover, what in PLCS is replacing it? Assigning_location could be used, but that is really indicating where a product is located, not where it is planned to be located or where it is intended to be located. And it doesn't give us the option of providing a simple location_indicator, which is required for circuit board placement of components (e.g. 'C1'), or at least it makes it much more complicated - those locations are not general, they are often very specific to the product, and change between products, so a regional grid or something is to complicated IMHO.

Or did you think of some other means to provide an indication of location in an assembly?


Peter



-----Original Message-----
From: David Price [mailto:david.price@eurostep.com] 
Sent: den 30 mars 2007 13:54
To: plcs-dex@lists.oasis-open.org
Cc: Peter Bergström
Subject: Re: [plcs-dex] Questions regarding Next_assembly_usage

Hi Peter - two comments below.

BTW, great work in forcing some decisions in these and other areas. I like how 
thorough you're being with this stuff!

On Friday 30 March 2007 08:32, Peter Bergström wrote:
> yes for #1, because the assembly is often referred to as an assembly (not
> the constituents), and everybody will have to be able to read and write it
> anyhow. It is easier to have a rule that requires it to be there at all
> times, than to supply it on a needs basis. Furthermore, different
> organizations will need the id, and if it is supplied from the beginning
> much hassle is over.

But lots of internal PDM tools don't have identifiers for relationships so 
this id attribute will be just a generated number in many cases -- not 
particularly useful as there will be no guarantee of uniqueness across 
exchange files.

>
>
>
> No for #2, since many assemblies will be fine just using
> Neaxt_assembly_usage without subclassing it to e.g. BOM or Spare Parts
> List. It should be possible to assign reference data to it in order to be
> explicit, but not a requirement.
>
>
>
> Use the attribute, for #3. The only thing that might not work with this
> approach is if different organizations require different strings for the
> location indicator, but I think it is a long shot. Why would a circuit
> board be re-labeled? Also, for really complicated installations (ships,
> power plants, etc) the breakdown structure is more fit to be used than the
> assembly structure, and there you have the possibility to talk about the
> 'slot' or location as a breakdown_element, while the equipment fitted in
> the location is a Part or Product_as_individual.
>
> I think it is incorrect to use the asg_id for the location, as is done in
> template repr_promissary_usage. I don't think the location is always an ID
> ('left wing' for example), and certainly not an ID of the assembly
> (Next_assembly_usage).

I thought location_indicator was just a hangover from the old PDM schema and 
was only there for interoperability reasons. Perhaps the recommendation 
should be for PLCS translators to consume, but never produce, it?

>
>
>
>
>
> Peter
>
>
> This message contains information that may be privileged or confidential
> and is the property of Eurostep Group. It is intended only for the person
> to whom it is addressed. If you are not the intended recipient, you are not
> authorized to read, print, retain, copy, disseminate, distribute, or use
> this message or any part thereof. If you receive this message in error,
> please notify the sender immediately and delete all copies of this message.

-- 
Mobile +44 7788 561308
UK +44 2072217307
Skype +1 336 283 0606
http://www.eurostep.com