process of reducing engineering errors & ecn'n 2

[duk748]

hello - our company has recently evaluated our ecn or eco process & found that even w/ the use of 3d cad our shop drawings used to manufacture our equipment still have too many errors - we do no checking leanning on the person who created the drawing to be responsible for his work - has anyone a plan that seems to work well to eliminate the rework, mistakes & ultimate ecn's that come w/ them -
our errors are costing us many $$ - also anyone wish to share a plan or procedure that works to eliminate the number of ecn's in general within their company - any info or feed back would be graetly apprecaited - thank you

 

[Walterke]

Are the errors a result of bad drawings or are they a result of bad comprehension of correct drawings?

If the first, educate your machine operators to learn how to properly read the drawings

If the seconds, hire better drawers/invest in training them.

NX 7.5
Teamcenter 8

 

[fcsuper]

The drawing is a contract that explicitly details the requirements of the vendor; even when that vendor is internal. What this means is that drafters/designers/engineers have the responsibility of laywers without the same pay.

Walterke is correct. What is the source of the issue? I would even question who is determining that the drawings have errors, as it may be they do not understand drawings themselves. I used to have a boss that would come to me with claims of "mistakes" on the drawings (by me and others) which turned out to be someone else's evaluation who was never trained in any related trades and didn't understand drawings.

That said, in my experience, relying on engineers to make good drawings without a checking process is a questionable practice. Some engineers are good at making drawings, but many just throw together "something" they think is enough. A checking process (even a peer review) helps reduce this effect, though sometimes the checker ends up being blamed for "delays" if management doesn't understand that drawings are contracts on par with written contracts.

Matt Lorono, CSWP
Product Definition Specialist, DS SolidWorks Corp
Personal sites:
Lorono's SolidWorks Resources & SolidWorks Legion

 

[duk748]

hello again - i usually get the job of checking when i have time & have been a designer/chief draftsmans since 1978 - i understand our product & try to follow a standard practice set up by my former employer since there is no such standard here - most of the time it is a problem because of getting things out the door fast enough to start manufacturing & things get by - in my own opinion we have a number of young employees that just have no knowledge of drafting practice or do no self check at all - they leave it up to the designer or project engineer who in turn has no time either - the entire process is flawed & set up for failure - thank you again & any nore suggestions or help would be appreciated - thank you

 

[TheTick]

I think you identified your own problem: checking. Take checking seriously and budget enough time to be thorough.

Your employer has taken the time to quantify the waste from not checking. Shouldn't be too hard to justify taking more time.

 

[ewh]

Checking and training; and keep in mind that 3D is just a tool and doesn't solve everything ("... even w/ the use of 3d cad...").

Technically, the glass is always full.

 

[ptruitt]

Proper design skills, training for GD&T, and other elements are certainly required, but...

This may seem off-topic, but I think there is probably much, much more going on that needs attention. I would suggest that you and at least one other person in the Opreations group read "Engineering Documentation Control Handbook" by Frank Watts. Read every page until it is almost memorized. This is a business book as much as it is a documentation book. You seem to need a goal-oriented business system first of all and you need to fill in the gaps (GD&T skills?) after that.

Peter Truitt
Minnesota

 

[KENAT]

So, making the assumptions that all of the ecn/eco are actually because of drawing/drafting errors not folks knowing how to read a drawing... and so concentrating on the drafting aspect:

1. Set the expectations of the drafters - if nothing else invoke relevant industry standards (iso or asme etc.) and make it clear that drawings are expected to comply with them and show a good level of completeness etc. Make sure copies of the standard are available to staff for reference.

2. If appropriate some education/training for the drafters but I'll be honest with you, I think this element generally gets over rated/over used because it's 'easy' in so much as you get them some online training software or a weeks gd&t course and say 'they're trained now - problem fixed - my goal for the year complete...

3. Drawing check. However, don't just use drawing check as an inspection process to verify the drawing is complete, complies with industry standards and doesn't have obvious errors. Also use it as part of a manufacturability review and also most importantly use it as an ongoing education/training tool. Make sure the person making the mistakes is the one fixing them - the intent being they learn from having to fix problems. You need to strike a balance between the checker/check process being unpleasant enough that it insentivizes folks to do it right first time, but no so horrible they avoid it at all costs or even quit etc. You need the right person to be doing the checking highly skilled, highly experienced and with the right temperament. It doesn't have to be a 100% dedicated checker but the person/people doing it need the experience and discipline etc. to do it well. Peer review is better than nothing but is only as good as the 'peer' doing the reviewing and the time they have to spend on it - so in the OP situation I doubt it would achieve much.


(Why I think training is over rated as a solution - well I learned to draft & GD&T etc. with little to no formal training, not even some self paced on line classes. However it was made clear to me that I was expected to make drawings and they needed to be good. So I looked at other 'good drawings', asked questions, read standards, got my drawings bled on by experienced checkers/peer reviewers and for an all too brief time period worked with an excellent checker who I learnt a lot from. On the other hand, just before I joined my current employer they had a week's worth of GD&T training, and yet most of them are still useless as they haven't taken the initiative to run with it.)

Posting guidelines faq731-376

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

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

 

[KENAT]

Sorry forgot, the checker should be involved in tolerancing/tol analysis too. Be it spot checks or reviewing submitted calcs or however you do it.

Posting guidelines faq731-376

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

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

 

[drawoh]

duk748,

Do you guys have an on-going discussion on how to do mechanical design and drafting?

I am on a mailing list for a local Linux user's group, and I have shown up at the meetings. There is quite a discussion going on about how to write software. There is lots of disagreement about stuff, but the expertise is shared. I have seen nothing like this on mechanical design.

The alternatives are to have each mechanical designer figure stuff out on their own, or to have the CAD manager, whether they be a mechanical engineer or an MCSE, work out the process.

JHG

 

[DeanD3W]

There have already been some good suggestions made. I have a couple more I hope are helpful.

It's best to have drafters doing the drawings and the engineers/designers checking the drafters' work (which may already be the approach there). If engineers/designers are doing their own drawings then there will be more errors... It's really hard to see the errors in your own work, no matter how good and experienced you are.

It's best to have design review meetings which include design, drafting, procurement/manufacturing, inspection, and a GD&T consultant. In smaller organizations these meetings may be just a few people, since some may wear several hats at once. I say "design review" to emphasize that the meetings should be as early in the design process as possible. If the design is "done" before the drawing is started, that would tend to create more rework later in the process, since all the manufacturing and GD&T and inspection considerations were not brought up early in the process. I've commonly seen drawings thought of as something to do after the design is pretty much complete... Even when various design concepts are still being considered, and very likely once a concept has been settle upon and initial models are created, it may be worthwhile to have a meeting to discuss manufacturing, inspection and GD&T. I realize that I'm stating the obvious when I say that it's better to get all the input and as many questions as possible answered early in the design process, but I think a lot of organizations don't operate that way... The 3D is considered complete and sent to the shop and then the afterthought drawing follows. Then once someone with GD&T in mind looks things over and says "why not change the way these two parts mate" it's too late to make changes.

Drawing checkers can help, but it's best if they understand the function of the parts and are also just about good enough with GD&T that they could teach a class. If the engineer/designer or GD&T consultant are checking the drafters' work I think that is likely a better situation (any Checkers out there please don't throw stones... You may know your stuff very well and I may say you should just have a different title).

There is one rule about drawings though... There is always an error somewhere and you will find the next one just after you print a hard copy .


Dean

http://www.d3w-engineering.com

 

[ewh]

Dean makes excellent points about design reviews.

KENAT,
The type of training I was referring to is the kind you alluded to; multiple, repeatedly returned bloody drawings (from a competent checker). That's the best way I've seen to really get a handle on drafting!

About Dean's drawing rule... it is true, and a good checker will always find something to complain about. I've seen easy, intentional, fairly obvious mistakes made to draw the checkers attention, so that once corrected the drawing can get released. Even good checkers don't like admitting that they missed something.

Technically, the glass is always full.

 

[MadMango]

It is sometimes very hard to see the forest for the trees. Whoever is making the drawings cannot check the drawings themselves. Time has to be made for drawing check, plain and simple.

If your work is project based, it may be a good idea to try to have a design review prior to parts being made. Reviewing the models as a group will go far in catching mistakes and getting everyone at least partially familiar with the design (and design intent).

Cutting corners in order to "get the job done" hardly saves any money in the long run. A method of preventing shortcuts is to have clear and concise standards and procedures in place, especially if you are dealing with a young and impressionable team. Once everyone is doing everything the same, errors will be easier to catch. You might want to start reading up on phase-gate processes for your engineering team.

http://en.wikipedia.org/wiki/Stage–gate_model

"Art without engineering is dreaming; Engineering without art is calculating."


Have you read faq731-376 to make the best use of these Forums?

 

[HDS]

You say the move to 3D has not eliminated the errors. That implies that you are talking about dimensions missing ect not parts that are designed wrong. Are you sharing the 3D models with the guys making the parts or are they programing from the drawing by hand?

 

[duk748]

hello again & all i can say is wow - thank you so much - we have no design review meetings - the company i was w/ before taught me how to work in an engineering office - as follows & i have tried to make my points known but it consistantly falls on deaf ears -
how i was taught:
design review meeting to kick off the design
design engineers & deisgners working togather to make initial layouts - drafters then take layouts to make details - details checked & corrected - assembly drawings made - re checked & final design review - sign off by project engineer - off to manufacturing - our project engineers are so shut up in their offices that they only get involved after the job is released or at the initial release - i am in constant contact w/ the design engineer i work for - i have adapted him to the above system & it works well to a point - he has me check if time permits what others have done on the project - he keeps telling me to teach the others & i try to when i can but i have work to do too -
my reccomendations on a full time checker have failed by upper management only to be told that it costs too much to have 1 person
only checking - my reply was how much it was costing us in eco' & warrenty work could have paid for more then one full time checker - i have been asked to come up w/ some solutions for a dept meeting but i am afraid that my notes will be thrown under the rug as before - it is nice to hear that some companies have a plan in place - thank you again to all & have a great weekend

 

[KENAT]

Sadly duk, we eliminated the checking role a couple of years back, and for a year before that the guy doing it wasn't really qualified (me) however, for a couple years before that we had a great guy and it made a difference.

Sadly pointing out peoples errors doesn't necessarily make friends and he'd found too many errors on the wrong folks drawings and rather than seeing it as they needed to improve, they took at as he needed to be eliminated, so he was one of the first to go when the RIF's came.

I also don't have sympathy for engineers that do their own drafting and do a crappy job. Either find a job elsewhere where you don't have to do your own drawings, or learn to do it properly like I did.

Posting guidelines faq731-376

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

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

 

[israelkk]

Most of the time the managers that makes the decisions and control the budget has no engineering background or even if they had in the past they make a switch in their heads as soon as they become managers thereby "forgetting" all they knew to please their bosses. Therefore there is never the time and money to do thorough job. But there is always the justification, time and money for reworks and corrections that cost much more money.

 

[ajack1]

It is fairly easy to try and come up with the perfect answer but over look the important thing that companies need to make money and work within strict time and budget constraints.

I would doubt that anyone would disagree that if you gave the design back to the person who designed it a day later and allotted an amount of time to go through their own design, some of the mistakes would get picked up. If you gave this reviewed design to an experienced checker with a fresh set of eyes more errors may be found. If you then gave this to another checker they may find even more errors, so on and so on.

All of this however adds time and money and needs to be balanced against cost savings in rework, recalled products, the companies reputation and even possible law suits that will change dramatically depending on what the product is used for.

To me the most logical route to go down is some kind of continuous improvement program or whatever the latest buzz word is, basically don’t make the same mistake twice. If you document what went wrong, why and how much it cost to put right you usually start to see a pattern and this will give you a good basis to work out how to solve that problem and what procedures need putting in place.

 

[ak762]

It's best to have drafters doing the drawings and the engineers/designers checking the drafters' work (which may already be the approach there). If engineers/designers are doing their own drawings then there will be more errors...

This practice have pros&cons depend from company scale.
From my point of view if designer can not express design intend in drawing by himself then it is eliminate personal responsibility for such work.
Other way it will lead to non rational time consumption.

 

[ptruitt]

Formalizing a system that sets goals, identifies problems and addresses them with measurable corrective action is a deep subject that can be helped by reading thick books or by hiring staff that have had previous experience properly identifying and correcting design development problems. Having checkers, or not, is a minor issue compared to the importance of having a well organized approach to product development. Look for improvements such as refined customer requirements as a basis for formal interchangeability rules, holding the line on changes that are not mandated by customer requirements, disciplined releases of bundled design changes to correspond to the business cycle and market changes... Design changes can be dangerous and it may be years until a seemingly benign change comes back to haunt.

Peter Truitt
Minnesota