Drawing completion time

[ebreeze]

I am a drafting supervisor at a growing company that is going lean and becoming six sigma based. In doing so, my boss, the Engineering manager, would like me to keep metrics on the drafters. I am expected to have completion times for drawings which I am finding to be near impossible for the fact that depending on the drawing type (detail, assembly, etc.) these times vary greatly. Does any one out there have to adhere to any similar requests and if so, any advise on how to handle this situation?

 

[ptruitt]

Try to glean the data semi-automatically from a database that tracks ECOs. Perhaps the ECOs are entered into a database. The documents on the ECO might be able to be tallied using a database query. Divide the man-hours in the period of time reporting by the number of documents ECOed. Rev. A drawing releases might be handled differently than >A revisions and, using a factor, estimate new drawing hours and drawing revision hours.

 

[MikeHalloran]

In the days of spreading lead, a drafting supervisor friend once said that he estimated 2 drawings per day, per head, regardless of size, complexity, or skill. He didn't attempt any finer resolution than that. It seemed like a decent estimate to me.

I don't think CAD has improved that number, at all.





Mike Halloran
Pembroke Pines, FL, USA

 

[SteveMartin]

I can understand the desire to have a measurement of this type of thing for estimating available capacity or having a solid basis of comparison between drafters, but I agree with you that it's not practical for many organisations. If the drafters know the metric is kept it also gives some emphasis to speed over quality, especially in the absence of any other metric.

If your projects are similar enough overall, and one project has only one drafter, you could try measuring completion time by drawing package, but in most cases that is as unsatisfactory as individual drawings.

 

[ctopher]

These days drawings can be created in minutes because a lot of people doing them don't understand tolerances or GD&T.
I once did a package of 40 D-size CAD drawings in one week. Another person may take a month.
The times vary among people, software, technique, training, education, etc.

Chris
SolidWorks 10 SP4.0
ctopher's home
SolidWorks Legion

 

[MadMango]

I concur with Mike's estimate of 2 per head per day. I've always used 4hrs per drawing which comes out as the same estimate. This accounts for the simple and complex drawings as a whole.

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

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

 

[ptruitt]

Try to present the information without adding steps. I gave my boss access to this information without adding any more work for staff. She could access it anytime she wanted it with a few mouse clicks. The request is reasonable. A good implementation of MRP, PDM, or (in my case) a custom MS Access Doc-Control database, will serve many purposes, including the one identified here. If you currently have a computer program managing documents and you can't glean the information needed from it, I would add that requirement to your 'Computer Systems Needs List' so it shows up when it is time to review new computer programs or program upgrades.

 

[TheTick]

While you are at it, take note of which drafters can produce with the fewest number of redline-n-redos.

 

[MechEng2005]

Determine the variables and start catagorizing.

A) Detail, assembly, ect
B) Type of project (i.e. Perhaps an automanufacturer would seperate based on model or car versus truck)
C) Drafter A, Drafter B, ect

As stated, be sure to further divide up time for different processes: initial drafting, checking, corrections, ect.

Then, once you have all the raw data, you can start determining the effects of variables and start to come up with some generalities.

Be sure to record the time it took for your analysis and include that in your "report" on how long a drawing takes.

The other option of course is to try to explain to the person requesting this information all the variables and difficulty in giving a straight min/drawing answer, but if the requester has their degree in business it will probably just be quicker to do a study.

-- MechEng2005

 

[MikeHalloran]

I don't think tracking redline/re-do time is necessarily a good measure.

My drafting supervisor friend and I differed dramatically in our evaluations of one experienced designer in particular.
DS liked him because he hardly ever spent any time erasing; if you asked for some new/forgotten feature, he'd 'just' add a bracket for it, changing existing parts only to accommodate the bracket. As a consequence, his designs comprised layered brackets, with horrible tolerance stackups, many unnecessary fasteners, and continuous production problems. There was never any eraser dust in the tray on his drafting board. He was really, really fast at producing terrible designs.

By contrast, the best designers, in my estimation, worked hard at removing parts and fasteners and combining functions, and always had a ton of eraser dust on their boards.

I never found a metric for CAD systems that was as useful as eraser dust.



Mike Halloran
Pembroke Pines, FL, USA

 

[ebreeze]

Well I appreciate all of the input.
Some packets that we recieve will have 20 drawings and others will have 1.
Some drawings are in autocad which then need to be converted to inventor, some are already in Inventor.
Some assemblies need all of the components modeled in Inventor when some do not because it is a newer model which is already done in Inventor, and so-on-and-so-forth.

I am trying to avoid going through and breaking out detail drawings, assembly drawings, outline drawings, schematics, etc. because at that point in time they will be paying me just to keep metrics for the department. I was hoping that someone had a generic measurement system that they use that seems to work but it sounds as if you all have the same feeling as I do....it is almost impossible. Wish bossman would accept that answer but instead it's "well you need to improve you system then"

 

[KENAT]

I know some places used to break it out based on 'sheet size' but even that doesn't really hold for CAD.

There really are a lot of variables to take into account.

Are the folks literally just drafting, or actually designing & tolerancing &...

Posting guidelines faq731-376

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

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

 

[ebreeze]

Literally just drafting. Making the print with the proper internal drafting standards, proper dimensioning, GD&T if necessary, etc. Anytime they do "design work" with the engineering department we do not count that towards the metrics. As of right now I have them completing 1 drawing per hour. That is fine if they are all detail drawings. I am having a hard time calculating how long it takes to do an assembly drawing. Granted I can do a series of thirty assemblies, take the average and call it a day. The fact of the matter is that some "assemblies" will only take a half hour because they are simply adding a note to the drawing. Other assemblies can take a few hours for the fact that they need to be assembled from scratch.

 

[vigildesign]

For Autocad, use the command "time". That will help you with that package. I am not sure Inventor has a similar command.

 

[Kwan]

I heard a story about a company that decided the bean counter's idea of measuring key strokes was a good metric. This eventually leaked out and an engineer wrote a simple batch that would insert thousands of key strokes with a simple ctrl/etc.

Imagine the reaction of the bean counters. WOW, our productivity just went thru the roof! Woohooo!

Measuring engineers is a very difficult task. If you figure it out you can easily retire, sail, and drink beer.

 

[ebreeze]

I could not agree with you more Kwan, although I am liking that scenario so I'll keep working on the answer

 

[ajack1]

Well a good place to start might be to look at how the job is quoted; I would assume there is a budget and timing plan. I am not sure this would reflect very well on the drafting supervisor admitting that whoever quotes the jobs has a better idea of how long the drawing will take but it is a start.

From there log how long previous jobs have actually taken and either try and compare similar jobs or just take a larger cross section over a period of time.