BOM Structures

[cloznuff]

I work for a smaller privately owned company. I am coming up on 20 years here. We are an OEM of industrial equipment including CNC milling machines. I am an Engineer and don't have a real strong understanding of business operating systems and MRP systems but I have a question. To date, we have used "Phantom" parts and "real" parts in our BOM structure. Pantom parts and assemblies have a bom structure but do not have routings/work orders tied to them. For a typical machine, we might have 10-20 "Real" assemblies, usually larger machining assemblies or sub assemblies that get completely assembled at a certain location. So for a typical machine going thru the shop that might have 2500 components, there are 10-20 work orders with accompanied cut lists, routing lists, etc.

We have a new "team member" who has come from a different background and is amazed we can make anything work. He is making a strong push to make every part and assembly "real" with associated work orders/routings. The advantage, supposedly, is being able to track actual time required to make any and every part, grouping of like components at the saw station, and determining economic order quantities. Great ideas, in theory.

I've been around long enough to see many changes in our manufacturing environment. In the last 10 years we have made many changes similar to this and have even hired consultants, twice, who were polar opposites in their manufacturing methodology. We (our management staff) drank the Kool Aid and it was a disaster both times. Engineering was forced to totally change our bom structures, once making assemblies of parts by manufacturing cell, totally ignoring the fact that drawings are required to show how to actually assemble something. Obviously, the world isn't perfect and no matter what you do, problems will arise. We have been very successful to date, despite all this and I am all for making things better but I'm having a hard time understanding how we will make this work by making everything "real". It seems the shop will get overwhelmed with work orders and spend more time punching in and out of jobs that they will getting any work done.

There is an immediate push for Engineers to make a change in how we structure our boms. Right now, if we have a weldment, the bom will call out a raw material part number, cut to a length. There is a push to create a new make part number and drawing for any part cut on the saw. This is another level to our boms and on a typical machine will require about 120 additional boms and drawings. In addition, the manufacturing engineers will have that many more parts to route, etc.

I don't see who will benefit from this, but I will be the first to admit I might be stuck in my ways. I'm trying to give the new guy the benefit of the doubt and am all for making things better. I just have the funny feeling in my gut this will be another push for change without any benefit.

Does anyone have insight into this topic of phantom vs real parts and creating drawings for every part?

Thanks

 

[KENAT]

Might not be the best forum, or maybe it is because you'll get both drawing purists and their nemesis folks with other back grounds reporting.

What the new guy is saying more closely follows the ASME Y14.100 series typical best practice by the sounds of it. At the moment sounds like you might have a lot of multi detail drawings which Y14.24-1999 section 3.2.2 cautions against. On the other hand weldments can be considered inseparable assembly's where it can be entirely appropriate to have a single drawing with multiple parts - especially if it's literally just 'cutting stock to length'.

If you use a lot of standard lengths from same stock, I can see there may be benefit in each different part having it's own part number (not necessarily it's own drawing). This might allow one work order to cut a whole bunch of stock to the various required lengths and be be put in a stock bin or similar.

For each common stock 'section'/ material you could have one tabulated drawing (Y14.24 section 1.8 figure 5) that basically gives all the standard cut lengths made from that material.

The other thing is this hang up on number of work orders etc. Always seems to me (and I get this argument too) that the ERP or what ever you call it system and supporting processes should allow for making multiple parts or even multiple steps of assembly on a single work order/routing or similar way to avoid unnecessary overhead but I must be missing something.

Not sure there's a one size fits all answer here, sorry.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[rb1957]

you currently have a system that works ... don't mess with it !

if the reason for the new way is to be able to track each and every part, then ask ...
"why?", "we don't do that now, why would we want to?", "we can get that information another way", "we don't need each and every part tracked", "what will we get having this data ? cheaper parts ? more profit ?", "our system works; if you don't like it, leave"

I've experienced all the mgmt trends d'jour ... CQI, 6sigma, MBO. We went through a phase of "one part, one drwg" ... disaster (well, many more drwgs than we typically needed). we've now drifted into a "skill" based system ... all sheet metal details on one drwg, all machinings on another, and installations and s/assembly drwgs ... now we have many drawings with the useless title like "sheet metal details".

bottom line ... if it works, don't mess with it !

another day in paradise, or is paradise one day closer ?

 

[MintJulep]

It's a business decision. Does the change result in a net gain to the bottom line?

 

[3DDave]

It's always a problem to back-drive engineering design from manufacturing process management. There is no good answer except "Yes, that's what we'll do!" if you like the idea of year 21.

What this is a symptom of is an inflexible and stupid segment of the company that has no idea how to do their own job and is keen on having some other segment of the company do it for them. My own irritation is when project/program management wants to know what % I have completed, but cannot supply a list of tasks to be done or means of evaluation to determine that %.

In the OP case it sounds like manufacturing management is incapable of creating the routing documentation they need and are looking to just rubber stamp the engineering docs as the manufacturing plan, once the engineers have created that plan for them.

Areas to look into -
Bar-coded floor docs, so that it takes 1-2 seconds for floor personnel to start/stop jobs.
Floating your resume. If this is the third shot in 10 years, it sounds like the company is struggling and management is making poor evaluations.

 

[verymadmac]

Is the new guy from a large company?

The question is it likely that there extra savings that would of set the cost of the extra work associated with the increased documents.

The first point I would make is how onto it are your shop floor guys / production people. Cheap labor general requires more comprehensive documentation, while the really on to it guys just have little note books full of work arounds. So asking the foreman etc, how well your system is working.

How many of the sub-assemblys are used in multiple assemblies, if there is a lot, then the use of Phantom parts and assemblies could possibly be a bottle neck for the production planners.

Creating a drawing every time raw stock is cut to length seems like stupid waste of design office resource (maybe not so much in large companies but in this case). A spec drawing that defines the stock material where the dash number of the stocks cut length works much better (or a simple company standard that defines the Part number structure for all raw stock).

Can't say I have ever encountered a design office where the system couldn't be improved, but one really needs to be pragmatic and consider all the interfaces with the rest of company.

 

[cloznuff]

Thanks for the responses everyone. We ran a small test assembly, creating make part numbers for every cut piece of steel and the reports didn't come out so well. It seems we have customized the reports for the way we do things now and if we switch our bom structures, we may have to modify the custom work orders and saw cut lists.

We had some lengthy discussions regarding the part number structure and, of course, there were many different opinions. Nothing was decided for immediate action. We'll see where it goes.

Thanks for the helpful comments.

 

[swertel]

Sounds like a simple case of Engineering BOM (EBOM) vs. Manufacturing BOM (MBOM).

EBOMs should not have phantom numbers. It is not up to engineering to define HOW a product gets manufactured unless there is some specific requirement in the process. The MBOM can have phantom numbers to aid manufacturing orders (travelers) and inventory locations and tracking.

For a small company, the distinction between engineering and manufacturing is often blurred due to the wearing of many hats, so differentiating an engineering process from a manufacturing process is more difficult. As long as someone is aware of the difference and keeps them separate, I think your current process maintains decent configuration management. You may want to ask this question again in the Configuration Management forum, phrased differently of course of avoid a repost flag. forum781

--Scott
www.wertel.pro