least imperfect part numbering scheme 1

[KENAT]

I started a thread over on GD&T but suspect I'd also get some good input here.

thread1103-160197

Basically my small department has been tasked to come up with a new part numbering scheme to standardize across our Division (essentially 3 sites which were originally several separate companies).

Currently we have 2 different active smart numbering schemes. We also have some legacy data from old companies which is a completely different scheme.

We think dumb (insignificant) numbering is the way to go, except maybe hardware (we have to assign a company pn to this for some reason, can't just ref the relevant spec).

Other departments, especially purchasing & manufacturing want a smart (significant) numbering scheme.

We're looking like being forced into a compromise, any input on the least bad way?

(please note I'm talking generic part numbering not specifically file naming or just drawing numbers plus we have an ERP system and are implementing a PDM system.)

 

[ddunn1950]

The best numbering system is a non-significant system. I have found that any significant numbering system will last only about 6 months before the significance is lost.

You can use a tab number (dash number) to indicate interchangeability but most ERP and PDM systems don't handle this very well. The look at part numbers without consideration of the controlling drawing number.

 

[mrMikee]

KENAT,

In my post to the following thread I mention a few of my favorites regarding engineering documentation.

thread781-136668

Concerning significant v. non-significant numbers the current wisdom is to use non-significant numbers. My opinion is that the answer is somewhere in-between, but I would have only a few categories... A for assemblies, P for piece parts, W for weldments, V for vendor parts, H for hardware.

I would never put the revision level in the part number however. Remeber, if the part is interchangable it keeps the same number but gets a rev, and if it isn't it gets a new number. It sounds simple but interchangability can be a difficult issue.

Regards,
-Mike

 

[ddunn1950]

A tab number does dot represent the revision. A tab is rolled to the next higher level when an item is revised such that it is only backward interchangeable.

 

[mrMikee]

ddunn1950,

It still sounds like the tab number is a rev. Do you have another system for indicating revisions?

-Mike

 

[ddunn1950]

The tab is not a revision. The document revision is listed in the traditional manner (A, B, C...). See rules of interchangeability below:

? Rules of Interchangeability
• Interchangeable – An item which possesses such functional and physical characteristics as to be equivalent in performance, reliability and maintainability to another item of similar or identical purpose and is capable of being exchanged for the other item without selection for fit or performance and without alteration of the items themselves or of adjoining items except for normal adjustment.
1. Non Interchangeable - An item which can not be exchanged with previous and subsequent versions of the item.
2. Backward Interchangeable - An item which can only be exchanged with all previous versions of the item.
3. Full Interchangeable – An item which can be exchanged with all previous and subsequent versions of the item.
• Change Class
1. Class I – Non Interchangeable – Requires and new part number
2. Class II – Backward Interchangeable – Requires a new revision and tab number
3. Class III- Full Interchangeable – Requires a new revision only

 

[mrMikee]

ddunn1950,

That's an interesting system. But it's not like anything I have seen before with my limited experience.

-Mike
Structural Engineer

 

[ddunn1950]

I have used this system in several companies. It helps with implementation, stocking and traceability.

The problem is with MRP and PLM system being part number focused rather than drawing focused. Adds more work on the CM people. But what the heck, we don't have anything to do anyhow.

 

[mrMikee]

I have taken two seminars on engineering documentation control and configuration management. One of the instructors is still teaching seminars and the link to current info is..

http://www.ecm5tools.com/sem5.html

I also recommend to anyone interested to visit his website at

http://www.ecm5tools.com

The book Frank wrote is a worthwhile investment with many practical hints. His coverage of revision control and interchangability of parts is very complete.

Regards,
-Mike

 

[KENAT]

I used to work defense in the UK and we had to do our own config control (I'm actually a desig engineer not CM) to govt standards so that side of things, when to rev when to give new number etc I'm OK on (not necessarily great but OK).

ddun your system looks pretty much like standard US govt numbering scheme which I've seen on UK/US joint projects and my other colleagues from US defence backgrounds are familiar with.

We looked at the 'tab' number (If it's the same thing I've always heard it called a 'dash' number) but our current systems won't handle it well. Interestingly though the other site does have something similar.

The real problem is we're being forced to go to a significant system by other departments with more influence amongst management, like it or not. My hope was to get some hints on how to minimize the problems.

We got it down to 10 'smart' categories amongst ourselves in an effort to please purchasing & manufacturing. It then got circulated and up to about 13-14 by our own site. Whith the other sites input we're now at 20.

On a side note do people usually put their documentation (workinstructions, specs, QA docs, policies, procedures etc) under 'part numbers' with the same ECO type process or do you normally have a separate numbering system for such docs with its own revision process?

 

[swertel]

Our work instructions have the same number as the drawing it references. Inside the work instructions are the specifics on how to manufacture each part number (dash number).

When the drawing is revised, a field on the drawing ECO flags whether or not the work instructions need to be modified. A new ECO is created to capture that change. Therefore, our work instructions actually have two revisions: 1) representing the rev of the drawing and 2) the actual revision of the work instructions - because typos or technical mistakes get revised on work instructions at a different time than the drawing.

Other specs, QA docs, policies, etc. are a part of the corporate structure and follow their own drummer.

--Scott

For some pleasure reading, try FAQ731-376

 

[ddunn1950]

Mike,
I am very familiar with Frank Watts’s book, Engineering Document Control Handbook. It is a good book for information on revisioning, interchangeability, Change Control and implementation. Frank does take a narrow view of CM as a bridge between Engineering and Production. I see CM more as the communication hub between all functions.

Kenat,
I know that sometimes you can be forced into a numbering system you do not agree with. All you can do is to point out some of the pit falls and try to control it best as possible. If you will let me know the 'smart' categories that are being suggested I'll try to provide some pros and cons.

On the side note, I have controlled work instructions, specs, QA docs, policies, procedures etc both ways. Under a separate numbering system and under the same numbering system. I do try to establish a separate Change Order process for WIs, SOPs, policies and other non-technical documents.

 

[mrMikee]

KENAT,

I've been through this process too. Many people want to invent complex numbers that code in lots of things that really aren't relavant in a part number. While you may end up with something you don't like, you have to do whatever you can to keep out the worst ideas. A few years ago I was involved in a restructuring project that went in a direction I told them not to go. After 6 months the new system failed and I was assigned the task of fixing the mess.

Part numbering, bills of materials, configuration management, etc., is very important to running an efficient organization. I recommend getting Frank Watts book, if not for you, for others involved in the project. I'm sure there are other books that could be recommended in this forum if anyone is interested.

-Mike

 

[oliebol]

Numbering conventions are FUN.

Try to gather information about a PBS (Product Breakdown Structure). On the internet there are several sites which give you info about this type of convention/strucure and also about the background of it (some MIL spec)

This convention can help to monitor CM processes during the complete lifecycle, is scalable and enables multi-level item groups.

Besideds a PBS it is also recommended to make a split-up between product numbering and project numbering.

This will allow modularity in your assemblies.

If you need more info about PBS don't hesistate to ask me.

Good Luck,


Martin


Unigraphics NX4,NX3,NX2,R17
Inventor 10,9,8,7
Solid Edge 10,9,8
Solid Works 2000,98
Mechanical Desktop 4,3
Autocad 2004DX,2000,R14, R12
Teamcenter 9

 

[KENAT]

Thanks for your replies.

We have just about reached our deadline for introducing the new scheme so once we have our proposal I'll try and post the basic categories.

Unfortunately I dont' think what we've come up with is that good, we had to put out a proposal before I could do enough research plus other departments had a big input but there you go.

Thanks again,

Ken