Case History: Good Old Human Know-How Beats Technology

Collaborative design is often billed as a tranformative technology that can overcome the traditional limitations of doing business globally. That’s what my friend, the CIO of a shoe company, thought. But he hadn’t yet met a person I call John Henry Hwang.

The original John Henry was a low-wage laborer who worked on railroad tunnels in the South after the Civil War. Those big strong men used pile drivers and hand drills to make holes for dynamite. According to the legend of John Henry, some clever salesperson (some say it was a manager) tried to put him out of a job by bringing in a steam-powered drill. Henry challenged the drill to a race. In the shale of those mountains, a strong man like Henry could, and did, outperform the drill, though Henry died in the attempt.

John Henry Hwang (not his real name) is a modern day Chinese version of the legendary John Henry, who has so far outperformed a stereolithography (STL) machine. Have you ever seen one of these? It can take a virtual 3-D image from a software package such as Catia and put out a colored 3-D model made of plastic. The STL machine divides the 3-D image into thin layers and makes the model by laser printing layer after layer of plastic.

Slick. And apparently useful at my friend’s shoe company. Here’s why. Right now, the shoe designers produce hundreds of designs for each season. The designs are sent to China, and the Chinese manufacturer’s John Henry Hwang makes a 3-D physical model from a design. Six weeks later, the models come back, the muckety-mucks get together and the final products are chosen.

What if we could just use the STL machine? We could save six weeks and who knows how much modeling cost, my CIO friend thought.

The Challenge

The CIO called a company that leases STL machines, and, serendipitously, it had a 3-D model of a shoe sitting around its office. The CIO got the model?and in one of those rare moments, about which you brag to your spouse?walked into the president’s office and plunked it on his desk.

The president’s reaction was the kind you like to come home and brag to your spouse about. If you’re thinking, “Wow. That’s great. Get on it right away,” you’ve got the gist of it. (The CIO’s wife almost certainly got a somewhat longer version.)

When the CIO walked the model around to the various division managers, their reaction was also positive.

To understand what happened next, we have to backtrack a little. Several years ago, the shoe company had bought design software that was built specifically for this kind of apparel. The implementation had run into difficulties, but the CIO was new to the company and figured that a few people were using the package and could give him a 3-D file that could drive the STL machine.

Nope. Nobody used the 3-D package anymore.

OK. What about a 2-D model? AutoCad and its myriad competitors were surely being used on all those Macs over in design. The CIO then contacted Catia to see whether it could take a 2-D image and convert it into a 3-D image that could drive the STL machine. Not a problem. You can feed 2-D images into Catia and get an exact rendering without too much manual tweaking.

Round 1: Flexibility Beats Precision

In the first division the CIO went to, the first designer he found didn’t actually use a computer. She would sketch a design and fax the sketch to China. A design engineer at that end would sit down with John Henry Hwang and talk the design through with him.

There’s a good reason for this. Not every fanciful whim of every designer can be turned into a real, wearable item. Materials can only be cut to such and such a thickness or sewn in such and such a way. (John Henry Hwang actually makes the protoype using the same machines that create the final product.) Many small adjustments are made as the model is built, and the result is not really a point-to-point replication of the sketch. It’s more of a design taken to the second generation, a prototype.

The effect of this is that John Henry Hwang has relieved the designers of any requirement to make their designs precise. He’s going to correct any mistakes, fix any proportions, anyway. Using a precision CAD tool?regardless of whether it’s 2-D or 3-D?is not only unnecessary, it’s extra work for the designers.

Round 2: Rapid Prototype Loses Ground

The CIO wasn’t about to throw in the towel yet. Hovering at his back, after all, were the words of his CEO, “Get on it.” There were still six weeks to save and the costs of creating the model.

My friend got me involved at that point, and we talked it over with the Catia people. They suggested the following idea. What if you created the design with a 3-D CAD package, then did a conference call with John Henry Hwang? You could each tweak the CAD design to take manufacturability into account, then use STL to create the prototype. A six-week process would drop down to one or two days. There’d be overhead?John Henry Hwang and the designers would need CAD seats?but you’d get a lot more control over the design, and the time to market would be under a week.

This, of course, is exactly why people do collaborative design (and buy design software). But alas, collaborative design isn’t coming to my friend’s company. He went back to the division with the most fashion cycles and said, “How about it. Two to three days instead of six weeks.” (He didn’t mention the overhead and the required change in their way of working.) And they said, “Don’t bother.”

Round 3: Technical Knockout

You see, John Henry Hwang had trumped the machine again. The designers in this area were already unhappy about lead times, and they had asked John Henry Hwang what he could do about it. Now, with the help of airfreight, some creative scheduling, some all-night work by Chinese employees who get paid about what John Henry did 100 years ago, and I don’t know what incentives to customs officials, they get models back within a week. And the price of the models has gone down to zero. (It’s now buried in the cost of manufacturing.)

Well, we were all taught a lesson. Appropriate technology works best, and sometimes no technology is the most appropriate. My friend’s the CIO, and he likes cool things (I too think STL and global collaborative design are cool), and he can see that the business problem could be solved with cool technology. But he didn’t need all that cool technology. All he needed was John Henry?Hwang.