Design Review Process 1

[Remydon]

I've recently been tasked with creating a formal design review process for our mechanical engineering team, and standardizing our drawings. I think I've got a good handle on the latter, but I'm a bit at a loss on the former. It seems ilke such a huge, open task to start from scratch; anyone have any good links or references to get started? My first instinct is to start a checklist type deal for going over calculations and drawings, but I'm sure there is far more to do than just check hole alignment and tolerances...

Suggestions?

 

[IRstuff]

MIL-STD-1521 is the military standard for what is to be covered in all the design reviews. As a general rule, there should be a minimum of 2 and ideally, at least 3, design reviews:

System Requirements Review (SRR) -- This is before you start ANY design. You have understand the requirements and have agreement with the customer with what each requirement means to either of you. While this was always a formal review, it tended to get deleted in contracts of the past, but experience has demonstrated that skipping SRR can often lead to errors and omissions.

Preliminary Design Review (PDR) -- This is when specifications to lower level assemblies have been defined and released, based on a preliminary functional decomposition of the system. PDR is the gate for detailed design.

Critical Design Review (CDR) -- This the review of the detailed design, wherein the customer can convinced that the design is now complete (>95%) and you're ready to cut metal and fabricate.

Both NASA and INCOSE have systems engineering manuals that describe the design process in detail.

The NASA document is MSFC-HDBK-1912A DECEMBER 6, 1994 -- SYSTEM ENGINEERING HANDBOOK
which was published by the Marshall Space Flight Center

INCOSE requires membership to get their SE handbook

If you do a search for systems engineering process, you'll find:

http://www.dot.nd.gov/divisions/maintenance/docs/OverviewOfSEA.pdf

http://www.fhwa.dot.gov/cadiv/segb/views/process/index.htm

and gobs more

TTFN
faq731-376

 

[dhengr]

Tasked with, begs the question... Do you have any experience with? Are you designing paper weights or gas fired turbines? There may be a difference in the size, complexity and type of list. I think the form of the question reveals some serious lack of understanding and experience with the whole design process.

You might start by listing all the important design steps, and what is critical about each step, and then have some review process by an experienced engineer on each of these steps. You might do different types of checking at different stages of the design and drawing development. What works for your group and product? I think there are some ME, ASME, ISO, etc. procedures for setting some of this type process up. We try very hard to codify everything, so no common sense or engineering judgement and experience must be employed any long. In good part because there is less and less of these important commodities out there.

You might want to involve the experienced engineers in your design group. Hopefully you still have some good ones. They will be able to help list the areas which have been the biggest problems over the years. Ask everyone in the Dept. to list ten problems they have had on their part of the total design and drafting process and combine these lists. A bunch of the problems will have common denominators and some of them will be on every list. Go from there, how can you prevent these problems from reoccurring in the process. This is an ongoing process in any design group and with every product, some with bigger problems than others. And, this task and the fruits of it will never replace real knowledge, background, experience and judgement.

 

[Remydon]

Thanks for the replies!

As you said, my inexperience shows. The problem I'm having is there are VERY few experienced mechanical engineers in my department. I work for the DoD, and we are one of the few groups on this base that actually designs anything. It seems like most of the other engineers only oversee contractors, and never get this far down in the weeds.

Our projects are mostly electronics packaging, and I'm the most senior ME in our group (with 2 years under my belt ). I'd love to learn from a more experienced ME, but the ones who are interested in actually doing techinical work all seem to have fled to private industry...

 

[hydtools]

Then IRstuff's recommendations are for you.
Your position sounds like my early, early experience in the USAF. Mil hndbks are a good place to start.

Ted

 

[ornerynorsk]

We use a 5 phase "stage/gate" process at our workplace (medical equipment). It covers concept feasibility, planning & input, prototype and design development, pre-production verification and validation, and production release including post-release follow-up. Each phase has numerous individual tasks and requirements, with design reviews being held multiple times per stage during development and pre-production.

There are a lot of systems out there, you just need to apply what is appropriate to your product, industry, regulatory requirements, etc.

Just my 2 cents. Good luck.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[MintJulep]

Separate "drawing review" and "design review".

No point in having good drawings of a bad design, or bad drawings of a good design.

 

[Twoballcane]

We have 10 Gate Reviews with PDR as gate 5 and CDR as gate 10 and even more gate reviews after that when we are off to manufacturing. Now gates 3 and 4 are internal for preparation of gate 5 and these are usually done by others who are not on the design team. Our company is big enough to have other subject matter experts (SME) come in from other parts of the company to evaluate; it’s basically the same thing leading up to gate 10. The takeaway from this is that whoever is doing the internal review has to be independent from the design team to tone down influence and skilled enough to get pass the design’s team’s spin on things.



Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”

Perception is reality: Your reality is how others perceive you, not how you think of yourself.

 

[Hoaokapohaku]

Congratulations for being "in the weeds!" That's the place to be! You'll learn far more and gain far more competency designing and implementing yourself than you ever would any other way. I love doing stuff when I only half know what I'm doing.

 

[GregLocock]

Cheer up, IRstuff is letting you off lightly with 3, for a car there are more than a dozen (admittedly many of those are manufacturing related rather than pure design).

I strongly suggest you try and understand what each review is supposed to be about rather than just running through checklists. I also suggest you schedule formal follow up reviews to catch all the 'we'll fix that next time round' issues.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[Cockroach]

ISO 9001 c/w Design Control.

The best system ther ever was, the best system there ever will be!

I will put my reputation on the line. Simply the best. You can down load it online, it resembles some of present day systems, but they are all bastards thereof.

Kenny

Regards,
Cockroach

 

[geesamand]

Our company policy is that everything is reviewed by at least one set of eyes, initialed, and dated. In more sensitive situations perhaps two checks is appropriate. If that fails us we apply training to those involved as a corrective action.

ISO type audits will encourage checklists, procedures, etc. But Engineering is an open-ended topic and tends to defy any attempt at comprehensive checksheets. If you use a checksheet I would keep it basic for things that are easy to forget when you're focused on the design: perhaps things like having adequate dimensions, complete material description, dimension lines not overlapping the part edges, surface finish marks on all toleranced features, fully datum'd GD&T, etc. I would leave design functionality off the checklist because a checklist will draw attention away from the real thoughtful considerations. It's that whole side of the job that will never be taught in school or written in a book, much less captured in a checksheet. If anything it would help to leave a comments section on the checklist and have the checker note their ideas/concerns/etc, so that later on if there is an issue with the part you can see if it was identified at that stage.

David

 

[IRstuff]

Design review is quite different than drawing review.

TTFN
faq731-376

 

[KENAT]

OK, while it may or may not quite confirm to DOD definitions let's make sure we're distinguishing between different types of reviews etc.

From my experience there's a big difference between a 'program review', a 'design review' and detail design/drawing checking.

Program review is primarily a project management exercise.

Drawing check is something done by a suitable skilled and experienced individual in 'slow time' - not a group effort around a conference room table in a short duration.

Even once you've excluded those 2 activities a design review can range in scope.

At a minimum it is a review of the actual fundamental design itself, so primarily a technical meeting focused on performance (however factors impacting cost & schedule don't necessarily get to be ignored). It's a good time to get 'fresh sets of eyes' involved. Assuming you have a detailed requirement then it may be that you go through the primary requirements and verify that they are met.

If you don't have a detailed requirement and/or want more general list then for an electronics packaging exercise it might be things like 'does everything fit in the enclosure with adequate clearance'; 'Will it survive stress/vibration/temperature... environments it's likely to see in service' (may need to give summary of analysis/testing to verify); 'regulatory/safety compliance'; internal thermal management... and so on and so on.

You may want to extend it to include things like DFX (manufacturability, assemblability, service, reliability...) , FMEA etc. or these may be better as separate efforts with only a summary of the results and impact on the design being brought up at the design review.

As alluded to above, you'll want more than one design review. As a minimum for a program of any significant size I'd say one when you have your basic concept fleshed out, another when you've finished detail design/before you make parts for prototype and a 3rd when you have built your prototype, done your testing and incorporated necessary changes.

Posting guidelines faq731-376

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

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