General Engineering Question 2

[debodine]

While this question is not strictly limited to Aircraft Engineering, since that is what I have been doing for years I wanted to asks my peers.

Some times when I review an engineering drawing to determine if it meets applicable standards (FAA, company, industry), I will find that an engineer has made a notation for some process of "Manipulate per Spec XYZ, Method A or Method B. Method B is preferred". The same will sometimes occur with material as in "Make from Material A or Material B. Material A is preferred."

I require engineers who need my review of their data to remove preference requirements. I do this because:

1. Engineering's job as I see it, at least in our industry to create FAA approved engineering data, is to determine what meets design intent.

If I put a note on a drawing similar to "Manipulate per Spec XYZ, Method A or Method B.", that means I have determined that both Method A and Method B meet the design intent and will result in a product that is in FAR compliance. If I put a note on a drawing similar to "Make from Material A or Material B", that means I have determined that both Material A and Material B meet the design intent and will result in a product that is in FAR compliance. I believe that is where engineering should stop.

2. When the manager of the department tasked with completing this task must choose between approved options, it is their opportunity to look at the current industry, company and market conditions to determine which of the two approved methods best fits their need for cost control, schedule adherance, inventory and manpower utilization, and any other factors that might need to be considered in choosing the method.

I believe that for engineering to specify a preference results in one of two typical outcomes:

A. The department manager who must choose between options will default to the engineering preference, possibly resulting in wasted resources (buying preferred material when in stock will do, ordering items to complete the preferred method when items are already on hand to complete the non-preferred method, etc.) I doubt this happens too often, but it is possible so I listed it.

B. The department manager will ignore the engineering preference callout because they have a whole host of conditions to consider when making the choice. Some or all of those conditions may have changed radically since the engineer decided to establish a "preference".

To my mind we have already in place supply chain procedures for establishing what is most cost effective for the project from any given set of options. And if scenario B is the common scenario (and I think it must be or the department manager charged with making the choice is not doing their job) then why put any preference on the engineering drawing?

I am open to your comments, including correcting me if I am missing anything truly important that would drive adding an engineering preference to any aircraft engineering drawing.

 

[KENAT]

I'm with you - per ASME Y14.5M-1994 at least this 'preference' type information doesn't really belong on the drawing. Define what is acceptable, not how to achieve it (except where the process explicitly affects end function etc. obviously).

Had a vaguely related thread over in the GD&T forum recently, regarding if unequal bilateral or unilateral tolerance imply a preference for the nominal value. I made similar arguments to you that what matters is the upper and lower limit of the tolerance range.

Posting guidelines faq731-376

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

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

 

[rb1957]

i also agree with everything you've posted ... Engineering's job is to create acceptable solutions. however, i don't mind noting that Engineering prefer one solution over another ... if they're looking at two equivalent solutions and need to pick, they might be guided by that preference (even if it is to pick the other one ! ... which leads to an infinite chain of double thinking one another).

 

[MiketheEngineer]

Over in the Structural/civil area we often spec a part, bolt, beam, etc and end with "OR EQUAL"

Reason being that is the mfg's manual I have handy and in some parts of the country a perfectly acceptable substitution can be used because its available and maybe cheaper.

Just my thoughts.

 

[debodine]

Thanks all for the feedback so far. Each of you has valid points, so let me share my thoughts one at a time.

KENAT:

You and I are definitely on the same page. In theory there is no question. In practice I sometimes junk my own theory, I have to admit.

rb1957:

That is a good point...even if I disagree with putting an engineering preference on a drawing, do I really need to be concerned? If two processes and/or materials are truly equal in all aspects (cost of procurement, tooling, current inventory, manpower requirements, etc) do I really care? I should be able to trust the supply chain to override our preference if it does not make sense.

MiketheEngineer:

You bring up a valid point even though I consider your scenario to be significantly different in one aspect. In my scenarios the supply chain manager had to choose between multiple "approved" options, meaning all available options had been reviewed and determined to meet design intent by engineering.

In your scenario it is possible for a non-engineering person to determine that something meets the design intent that has not been reviewed and approved by engineering.

As I confessed to KENAT sometimes I junk my own theory and go with allowing a preference even though I disagree with that, sometimes I will also allow OR EQUAL (our typical term is OR EQUIVALENT...just semantics) to exist on the drawing as well.

As noted, I struggle with the theoretical "engineering should determine acceptable alternatives with no prejudice to choose one or the other" since engineering rarely possesses all the current knowledge to determine which option is more reasonable at any given time.

In the same way I struggle with your point about OR EQUAL. To be very blunt, we have manufacturing people who through force of years of experience can better specify methods and materials than some of our engineers. I have 38 years in aviation but I still go the floor and confer with the "old hands" often because I don't do what they do every day.

I would consider that to be similar to what you face. Your drawings are probably utilized by a number of "old hands" who are capable of selecting an appropriate bolt, part or beam.

To also illustrate the difference between what little I know about your industry and to compare it to aviation, according to the regulations we have to follow for FAA approval, any changes to engineering drawings concerning materials, parts, processes, standards, tooling, etc must be approved by the FAA before being considered as approved data for use on an aircraft. This includes all changes to the Instructions for Continued Airworthiness which are the data used to return an aircraft to service after modification, repair or overhaul.

So far I have not seen a strong reason to change what I am doing, but I do sometimes wonder if I am being too harsh. Sometimes the FAA regulations seem to pre-empt engineering judgment, but I have usually found there is a good reason for it when I get all the facts and the backstory.

Again thanks all for discussing this with me. I am still open to learn more but right now don't plan to change much on how I determine what is acceptable on a drawing for these conditions.

debodine

 

[KENAT]

One thing, in highly regulated worlds like aerospace, defense or medical you have to be pretty careful with using 'or equivalent' due to concerns over who decides what is 'equivalent' etc.

I suppose one implicit way to suggest a preference is to detail the preferred way then have an 'Alternate method...'.

However, this still comes back to the same issue as that drawing tolerance example I gave. The drawing is still saying either is perfectly fine so isn't really giving a preference.

Plus honestly, if functionally either is adequate why should engineering have a preference?

Posting guidelines faq731-376

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

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

 

[debodine]

KENAT:

That is really my basic point. I have had persons from all departments including engineering mgt tell me it is ok to have an engineering preference on a drawing and I keep going back to why? I see no value in establishing an engineering preference because there is no such thing. Any options approved by engineering must meet design intent.

I am not saying there are not valid reasons for preferences in the final decision as there most certainly are valid reasons to consider in making the choice. I am just saying that engineering is not the department to make that determination and engineering drawings are not the place to document that determination of preference.

My bottom line is I appreciate my peers being willing to share their thoughts on the subject. I am always both encouraged and enlightened (even when, and maybe especially when, I am shown to be in error) by the responses of the people on this forum.

 

[KENAT]

I can't help but agree.

If there is a really big cost difference then you might choose to only detail that one option rather than giving it as a preference.

However, even here one day you may get a rush contract, and the production capacity to do the cheap method is full so they want to do the more expensive process to meet schedule, but if the drawing doesn't show it they end up needing a waiver or something - or may not even think to do it that way so lose out on the contract or something.

So, within reason, I prefer to give operations options and let them decide which is most appropriate for any given production run.

The other common aspect on preferences is 'aesthetics'. Arguably not a hardcore engineering requirement so perhaps shouldn't impact the engineering drawing. However, if not on the drawing, then where? Plus even then, if aesthetics is important enough that you have a preference, should you even allow the less aesthetic option?

Lots of things to consider but I keep coming back to thinking that 99.9% of the time there's no place for 'preference' on the engineering drawing.

Posting guidelines faq731-376

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

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

 

[dik]

If both methods are acceptable, but there is a (personal)preference for one (for whatever reason), then it's OK to stipulate both and note a preference. I don't see a problem with that.

If you have a good reason to stipulate one in preference of the other, then I would delete the other.

Dik

 

[IRstuff]

"Any options approved by engineering must meet design intent"

That doesn't preclude preferences. Let's say there's an MTBF requirement of 100 hrs. Option A is 100 hr, Option B is 1000; they both meet spec, but wouldn't you prefer Option B? All else being equal, Option B ostensibly has a lower lifecycle cost.

TTFN
faq731-376

 

[dik]

Sorry IRS... but if Option A is acceptable to the engineer and the lifetime of the object that this is the component of is only 75 hours then either Option A or B is acceptable with the preference being option B. The 1000 hour lifecycle has no merit... If it is an issue, then there shouldn't be a choice and the engineer should stipulate option B and delete option A.

Dik

 

[KENAT]

IRstuff was talking MTBF, so let's assume the life of the product is longer than the 100 hours for the alternate component.

If there is a big cost or lead time difference etc then it may well make sense to have both as options.

From a total product lifetime cost point of view the longer life one may be a better choice even if slightly more expensive.

I'm still not 100% sure this is a hardcore engineering function issue so arguably doesn't belong on the drawing. However, IRstuffs example has got me thinking.

Posting guidelines faq731-376

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

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

 

[IRstuff]

Yes, I was. Both were stipulated to meet requirements, but B has 10x better performance, so I would prefer that, all else being equal.

Clearly, in this case, if there are no downsides, then B is obviously more preferable, and what that would mean is that one would desire to purchase B, and if B cannot be obtained, one would settle for A.

TTFN,
Eden

TTFN
faq731-376

 

[debodine]

I certainly do appreciate all the inputs from fellow professionals to make me think. I confess I am a fairly hard-headed person and I find it difficult to change my mind as I tend to believe I am already correct. Sorry, I admit it is a character flaw and not backed up by reality. )

But the thoughts presented here do give me a lot to think about and may help me consider alternatives I would normally rule out.

I have to admit so far I think we all agree on a couple of basic ideas:

1. SOMEONE has to make a choice between acceptable options, and a preference on an engineering drawing could provide guidance if all other factors are reasonably close to equal.

2. Even if I conclude there isn't value in adding a preference to an engineering drawing (for our industry/market anyway), there does not appear to be any harm in doing so.

I think IRStuff created an excellent example for consideration about the two options, one with 100 hour MTBF and one with 1000 hour MTBF. Let's run with that...

Both Option A and Option B meet the 100 hour MTBF requirement, but Option B also exceeds that requirement by ten times. ALL OTHER FACTORS BEING EQUAL, IRStuff is absolutely correct the best option is Option B.

I would suggest that all other factors are never truly equal. To be ten times the life, something about Option B must be different from Option A. Suppose Option B weighs more and/or costs more and/or takes six weeks longer to obtain to get the extra life (a realistic possibility). Whether or not Option A or Option B should be the choice for this particular customer is now dependent upon characteristics that were not part of the decision for the engineer to add a note preferring Option B.

My point with the above is that non-engineering factors will ALWAYS control the choice between options for any given project or customer, and that is the way it should be. Thus I tend to reject putting preferences on engineering drawings as the choice will be made further down the line by program managers and supply chain.

But I readily admit each of you has made me think, and I will continue to consider this information.

Many thanks to all. Even though I already admitted I typically believe I am already correct, I do realize I AM NOT THAT GOOD and thus I may need to be schooled. I recognize most of the posters here as I have viewed this site for a number of years now. I like the fact that people I have learned to respect (in absentia, granted, as I have never met any of you as far as I know) are willing to converse with me about engineering.

 

[IRstuff]

"non-engineering factors will ALWAYS control the choice"

The fact that there is no requirement driving the choice of B automatically makes it a non-engineering choice, and given the option, there's no reason to reject the engineer's choice in favor of some random person's.

Moreover, a "preferred" label does not make it a mandatory, or even automatic, choice. "Preferred" simply says to someone else, "I like this one better," but I'm not going to force to choose it, but I have reasons for preferring it." If it were a forced choice, the engineer could simply not include A in the selection and only put B on the drawing.

One must presume, barring bribery or conflict of interest that the engineer has sound and rational non-technical reasons for his preference. They could be:
> price
> delivery
> service
> quality
> responsiveness
> previous performance/experience

Since the engineer is the subject matter expert, he should have the highest weighting. Forcing the choice to be unbiased when there is a preference expressed by a subject matter expert is tantamount to saying to him, "your opinion is irrelevant."

Bear in mind that in many government acquisitions, the bidder is often required to provide a Relevant Experience, and is sometimes required to solicit a customer evaluation for previous work. These are specifically intended to provide a way to objectively choose between two bidders that are otherwise fully compliant and equal in the technical evaluation. So, rather than flipping a coin, one can look at the non-technical factors presented by subject matter experts (previous customers).


TTFN
faq731-376

 

[KENAT]

One aspect though is that:

> price
> delivery
> service
> quality
> responsiveness
> previous performance/experience

Are all relatively variable, and even having a full technical documentation of options A & B does not prevent that.

Say when you design it option A & B are comparable on the above traits, and so you 'prefer' B with it's longer life.

10 years later on a follow up contract with the original engineer long gone etc. let's say that through the multiple mergers/divestitures/political instability... that most of these factors have changed. Option B is now much more expensive, longer lead time...

However, there is this nebulous 'preference' for B on the drawing. Can it be guaranteed the planner/manufacturing engineer or whoever spends the time to weigh the issues, or will they blindly follow the preference thinking 'the Engineer must have known something I don't' or similar?

If it was a simple either/or then they'd most likely make their decision based on the 6 above factors and pick A but that nebulous unexplained 'preference' is niggling in the back of their mind.

Obviously once you start second guessing folks ability to do their job properly where do you stop, but I'm just trying to think through this issue.

I'm tempted to say that 'preference' belongs outside of the actual drawings in a format where justification for the preference can be given. Be it in the quality plan, or notes in the ERP system or some other way.

Posting guidelines faq731-376

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

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

 

[KirbyWan]

Good discussion,

My feeling is that if you have options on the engineering drawing without a preference you would need to evaluate which option provides better value. This may happen everytime someone new picks up the drawing. If someone such as the engineer does this once and encodes the best option as a preference it doesn't need to be reevaluated every time. I wouldn't put it down as a preference though. I would say 'make it this way' and then offer optional methods or materials.

I know that there are times where the situation defines the prefered material or method. To extend the MTBF arguments before lets say there are to products C that needs to last for 500 hours and D that is a one time use that only needs to last for 10 hours. You have a component of this product that could be option A with 100 hour MTBF or option B with a 1000 hour MTBF, but at 5 times the price. You could use either component with either product, but you see a natural preference. I'm asuming that a failure of component A on product C is not critical and is replaceable. At this time I would say we're looking at different components, and it's time for a different part number.

So the optional materials or methods callouts should be simple and obvious, e.g. use 2024-T3 (Optional 2024-T42 or 7075-T6)

-Kirby

Kirby Wilkerson

Remember, first define the problem, then solve it.

 

[KENAT]

However, you come back to if 2024-T42 or 7075-T6 are acceptable from a functional point of view, then from an engineering point of view why is 2024-T3 preferable?

If it's due to cost, availability... these are all fairly variable factors that probably should be evaluated each time an order is placed.

Posting guidelines faq731-376

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

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

 

[WKTaylor]

Folks... my 2-cent worth, this issue.

I work for Company XX. Company listed parts, processes and materials are "by definition" preferred for internal use. In many cases, these specs are fairly restrictive, relative to vendors and processors that MUST be qualified [by the company] to do the work or make the part or raw material. All is well with the world if these are available and within cost/time budgets [these usually cost a lot more than "standard-whatevers" just for the warm-fuzzy QA feel they give].

HOWEVER, I also work in the real world, ultimately providing my service to the USAF where AMS, AS, MIL, MS, NAS, ASTM, AWS, etc are expected/demanded, whereever possible.

So, I attempt to serve 2 masters at one time; while allowing broadest range of options possible/rational. This puts pressure on everyone reviewing the requirements to ensure that the "least capable aspect" is still adequate to meet the task.

NOTE. For company XX hardliners, this "give'em options stance" cuts across their soul and many take a folded arms/sit-down/"hell-no" attitude, until a manager kicks the stool out from underneath them reminding them the USAF demands it.

Here is how I do it (typical wording)...

Fusion weld per xxxxx (preferred) or per AWS D17.1 (optional) or USAF T.O. 00-25-252 (optional), class **.

7050-T7451 plate per xxx-xxx class * or AMS4050 [ultrasonic inspect AMS material per ASTM B594 Class A] (xxx-xxx mandatory for use by XX and it's subcontractors).

Shear Bolt. (Preferred) xxxxx-4-24 or NAS6204-24 (Optional) NAS1104-24 or NAS464P4-24

Hope this makes sense.

Regards, Wil Taylor

Trust - But Verify!

We believe to be true what we prefer to be true.

For those who believe, no proof is required; for those who cannot believe, no proof is possible.

 

[bigTomHanks]

I think that if you are too restrictive on your drawing and there is in fact a viable alternative that would be more beneficial to the company then you will most likely be asked to change your design. The engineers who give options realize that engineering is only one piece of the design and fabrication of a product. Most people who touch the design from start to finish know enough about a product and it's design to know when it is appropriate to request a design change from engineering. This is very common where I work and the typical response is positive. Most understand that if a suggestion is valid and they still stick to their restrictive design then management will be coming down on their boss for slowing down production when a viable alternative is available. It doesn't take many incidents like that before they become more receptive to reasonable suggestions.