A long time ago I did a course in Operations Management and was lucky enough to be taught by a guy with a PhD in the subject who had followed one of its leading lights by working on manufacturing in Japan, as well as other parts of the world.

That leading light was a guy called W. Edwards Deming, who started off as a physicist, got interested in statistics and gradually moved into the world of manufacturing where he relentlessly made the case that it could be used to constantly improve the processes and thus the competitiveness of any company. Despite having worked with US manufacturers during WWII he didn’t get much interest from them after the war. But there was one nation that did listen:

Deming gave his Japanese students not only statistical theory, but also confidence. “I told [Japanese industrialists] Japanese quality could be the best in the world, instead of the worst,” he said. Still, many were skeptical. “I was the only man in Japan who believed that Japanese industry could do that.” Deming made his prophetic statement that the Japanese could capture world market within five years if they followed his advice. “They beat my prediction. I had said it would need five years. It took four.”

The Japanese were so impressed that they created the Deming Prize in 1951, awarded to people who have best enabled quality improvement in business production of good and services. Things like the famous Kanban system (also called Just-In-Time or JIT manufacturing) is just one aspect of what is generally called Continuous Process Improvement, which you can read about at the link to ASQ, an organisation dedicated to it.

I don’t know who in SpaceX is going to be awarded the prize or if anybody ever will be singled out, but the following developments in their Raptor rocket engines are a perfect demonstration of the method at work.

What’s not mentioned in that X comment is that between the R1 and R3 versions the power increased by 67% while the weight dropped by 50%.

Earlier this week the company flew Flight Test #6 of its Starship system and there was some disappointment that the booster was not caught by the launch tower again as happened in FT#5. But apparently that was due to the tower not the rocket, with more tower damage done on this launch. The booster was instead landed nicely in the sea with the usual vertical touchdown a few miles offshore. The same thing happened with the Starship, which also did a soft, vertical touchdown in the Indian Ocean, this time in daylight and right beside the buoy that SpaceX uses to capture video images of the landings.

But the whole thing generated new data that will be used to improve every aspect of the system, the Deming approach in action. In recognition of this the FAA has approved SpaceX to fly up to 25 of these flight tests next year, which is an extraordinary rate for such a new system.

Readers may also enjoy this 20 minute video from “Curious Droid” (Paul Shillito) explaining why SpaceX are using 33 rocket engines in the SHB rather than the 5 huge engines in the Saturn V. Given the failures of the Soviet N1 Moon Rocket in the late 1960’s/early 1970’s, with its similar cluster of rocket engines, I’d always wondered why SpaceX took such a risky design route, even if computer control systems had vastly improved.

Shillito goes into the comparison in some detail and shows how it makes perfect sense for a vehicle that is supposed to be re-usable.