Jim Brown

What I Learned: New Insight on Physical Prototyping
August 18, 2008

What I learned this week ... came from a visit to the Scottish Maritime Museum, specifically a lesser known location called The Denny Tank in Dumbarton, Scotland. The visit was almost accidental while I was on vacation in the UK, but the lessons learned was a much better view into the cost of physical prototyping and testing - but also a greater respect for the need for it. I will try to share some of the insights I had while visiting the facility, and apply it to today's modern product development.

Quick Background and a Chance Encounter
The Denny Tank is what is left of a very large shipbuilding complex in Dumbarton. A scale model of the former facility makes this test facility look like a small shed (this was shipbuilding on a very large scale). But the tank itself is impressive (over 100 yards long, and 20 feet across as I remember them saying). It is well preserved in a way that you can really imagine what working there was like when the operation was in full swing. My kids even got a chance to work with some tools to shape the wax models they used for testing, which was fun for them while I was learning as much as I could from this fascinating place. The funny thing is that we ended up there because were looking for something to do and on the GPS it said both "Maritime Museum" and "Tank Museum" and my son is very interested in military equipment and has been learning about World War II. But as they told me when we arrived - "different kind of tank."

The First Insight from the Visit - The Cost of Prototyping
So other than showing you my vacation photos (sorry), why should we care as an industry? What really struck me looking at this facility is the time, effort, and energy - and of course cost - that went into the process, the people, and the equipment. Making the prototypes was very manual, using clay beds to make wax models, shaping them, and then outfitting them with the appropriate gear. Then, simulating different wave conditions and drawing the model through the tank while taking measurements on time, drag, and other factors. And what if the design changed? They could simply melt the wax down and create a new prototype from scratch!

I have written about the cost of physical prototypes in the past, but this was a real eye opening experience to me. To see this process directly (even if in a museum) really prompted me to value what computer simulations can provide. Most people reading this blog, I assume, don't have products as complex as an ocean liner. But many companies do develop physical prototypes to evaluate designs. I don't think that prototyping is ready to be relegated to museums, but the more companies can do to minimize it in their own businesses the faster and more efficiently they will get products to market. But of course, they can't sacrifice product quality and reliability.

The Second Insight - This is not just History
Don't get me wrong, I am serious when I say prototyping is not ready for the world of museums. This is not just a history lesson on the way they did things a long time ago. Today, they use fiberglass instead of clay , but the facility is still used on occasion. And other test facilities like the David Taylor Model Basin in the US are used regularly. Let's face it, not everything can be accurately simulated. As much advancement in simulation technology as we have made as an industry, there is still no substitute for understanding the impact in the real world. To demonstrate this, the tank was used to evaluate new hull designs like bulbous bows. The museum holds a series of fiberglass bow models that were tested at the facility, although I don't know the exact date. The reason is twofold:

• Some things are very difficult to simulate (including complex fluid dynamics)
• Simulations frequently need to be validated by physical tests

So as much as we can decry the cost of physical simulations, they are still valuable and necessary in many circumstances. The challenge then is how can we optimize the use of physical prototypes, and continuously improve their correlation with physical test results to make better simulations over time. I should also add that other efforts are in place in companies to reduce the cost of producing prototypes, which is another area that has seen some dramatic improvements but a topic for another time and place.

So that's what I learned from my chance encounter, I hope you found it interesting. Who knew? I didn't, if you did let us know about it. I am not an expert on simulation or ship design, but I hope the insight was useful and I welcome your comments. At least now you know what industry analysts do when they go on vacation...

Posted by Jim Brown on August 18, 2008 | Comments (0)