Tuesday, September 14, 2010

Considering the Cost vs. Benefit of CAD

As a manager or leader of product development, have you ever wondered what the real value is of all that CAD data that is being created? Are you getting a positive return on your CAD investment? Probably a strange question to ask since everyone seems to concur that CAD is one of the basic necessities of product development. I’m actually not questioning if CAD is adding value, what I’m questioning is just how much of CAD is adding value. Our CAD tools are incredibly powerful – whether you need it or not.

When I first started my career I worked as a machinist in a prototype and model shop at Hewlett Packard. One of my specialties was in designing and building soft tools (molds) for plastic parts. At that time most of the product designers and engineers were well on their way of moving from 2D CAD to 3D CAD. I could easily tell a difference in the drawings as they moved to 3D CAD. With 2D CAD I would get drawings with a single line that represented the side of a part or feature, with a note designating plus draft or minus draft – very easy to deal with in 2D CAM. With 3D CAD I started getting drawings with two lines representing the side – with no notes. At that point in time, the cost of the drafted 3D model was actually higher than the value – cost being related to the effort to create the draft features AND to use them. We actually stopped putting draft on models for a while. Of course as both CAD and CAM improved, and our use of the 3D model improved, the value of the drafted model soon surpassed the cost.

That’s a long story to make a simple point. Are you getting a positive return on your CAD efforts and resulting data? I have no question that you are getting positive value from a complete and accurate 3D model (even though in some situations we are still removing features that someone spent time creating). My question is more related to all the other stuff we are doing with CAD. Is it adding value beyond effort? We have some amazingly powerful capabilities in our modern day CAD tools. Being a guy that really enjoys CAD, I want to master and use them all, whether I need to or not.

I find it fascinating to see how much time a CAD jockey can waste on coming up with some elaborate associative parametric solution to a problem that really didn’t need to be solved in such a way. It can certainly be challenging and at times enjoyable, kind of like solving a Rubix Cube. Of course there are many times when this capability yields very high value, and that is why this power exists in the first place. But is it required in the design of all of your parts/products? What about simply getting parts out the door? Sometimes I think we get sidetracked from our real objective; and that is to design products, not model them. Just to be clear, I am not trying to make a case for or against history-based or history-free modeling. In CoCreate (the product I represent) I can fully parametrically control models and assemblies, in some very elaborate and intriguing ways. “Over modeling” can be done in most any CAD tool. (That’s my new term for the day: “over modeling”.)

It would be interesting to evaluate the effort that goes into all of the additional intelligence and associativity that we are building into our models versus the actual value that we are getting from it. Often times this intelligence is considered insurance against mistakes, although it is not too uncommon to apply this intelligence incorrectly. There are other, and sometimes cheaper, ways to ensure correctness. Perhaps we consider this extra layer of intelligence to be necessary to convey design intent to other team members and downstream functions. This is certainly valid in many situations, but there are less costly and simpler ways to convey design intent. Sometimes we are adding this extra layer of intelligence to the model simply because the CAD tool forces us to; it’s intrinsic to the modeling process.

What about the life-cycle of this intellectual property we call "model intelligence"? Will it be easy to “consume” and extract value from throughout the life-cycle of the resulting product? Can it be leveraged in other products we are designing? Is it built on standards or specific to a particular tool? What is the possible value? Does it yield a positive return?

Often times the problem is simply that we may not realize that there is an alternative to what we are doing. Have you considered the alternatives? If you do not recognize alternatives, you will never recognize opportunity.



R. paul Waddington said...

You raise some very good point in this post Paul.

"Sometimes I think we get sidetracked from our real objective; and that is to design products, not model them."

A complaint from a major project manager last year, to a customer of mine, was along the lines, CAD has saved us nothing, indeed "document cost and production time (in the drawing office) has inrceased because the guys spend too much time putting in too much detail."

CAD is a great tool but it must be managed 'sensibly' to get the 'best' from what it offers.

Mike said...

Makes me think of the scene in Spinal Tap when they are looking at the Amp that goes to 11. Well why not just make 10 louder? But this one goes to 11.