Is Elon’s “Idiot Index” an over-simplification?

Question

Edit: this question was originally closed because it is "not related to space exploration". However, SpaceX's major achievement is reducing the launch cost for space exploration. Elon credits his cost reduction strategies (including the Idiot Index) for reducing the cost of Raptor engines 10-fold.

---

Elon is famous for (amongst other things) his “Idiot Index” mantra. This quantity is a component’s production cost as a multiple of its raw material cost.

https://www.linkedin.com/pulse/elon-musks-management-mantra-idiot-index-beyond-thushan-pathirana-fv0hc

This index evaluates the efficiency of a component by comparing its total production cost to the cost of its raw materials. For instance, if a component costs 1000 to produce but the raw aluminum in it is only worth $100, it would have a high idiot index

Fundamental to Musk's philosophy is the practice of challenging every requirement, a mindset that has underpinned his achievements at Tesla and SpaceX.

The "idiot index" was particularly highlighted during Musk's focus on the Raptor engine for SpaceX's Starship, which necessitated manufacturing hundreds of units at a reasonable cost.

But parts with a higher Idiot Index are not necessarily more idiotic. Reducing material or substituting a cheaper material (like stainless for carbon composite) will raise, not lower, the Idiot Index. This despite the fact the new design may reduce overall production cost.

Conversely, some parts get almost their entire value from processing, and raw material cost is negligible (computer chips vs vacuum tubes). Chips may have an astronomic Idiot Index, but it would be truly idiotic to replace them with tubes.

Since what matters is overall production cost, why not just look at that instead of inventing a misleading “Index”?

Answer 1

I will wholeheartedly agree with you that it's an oversimplification, but I will also respectfully disagree with the statement, "Since what matters is overall production cost..."

In rockets, which are governed by razor-thin engineering margins and the rather unforgiving rocket equation, the importance of minimizing the mass of the structure is unusually high. But if engineers remove too much mass, or reduce mass in the wrong places, then the rocket may become unreliable. This would affect the company's reputation and insurance costs. So it's just as easy for a part to be non-optimal by being under-engineered as it is for it to be non-optimal by being over-engineered.

The only way that Elon's idiot index would be effective is if the industry had already gone way overboard in the over-engineering direction.

Now it's quite possible that Elon learned of a parasite problem within the space industry, which is defined here, back in 2002, as

There is no question that the space program has what Larry Niven calls a "parasite control" problem, too many people whose services are not really required trying to use it as their meal ticket, and a political system that prizes their votes and responds to their lobbying.

In that case, he would have seen the space sector as a target-rich environment for an entrepreneur with low-cost manufacturing know-how. But, in rocketry, it's quite easy to apply a strategy like the "idiot index" and end up overshooting the mark. If that happens you could end up with a rocket that, while cheap, is either too heavy or too unreliable. Then you would have to course-correct by applying the "not-an-idiot-index" (Trademark @phil1008) to drive the design back towards optimal.

What matters is a complex function of payload, safety, reliability, cost, and other important factors. Anyone who has the skills to analyze the trades associated with juggling all of these elements is a master of the not-an-idiot-index. These are the kinds of people that I'd want to keep around.

Answer 2

The function of the index is to trigger the implementation of the algorithm

Since what matters is overall production cost, why not just look at that instead of inventing a misleading "index"?

Let's look at the five verbs in "The Algorithm":

1. Question
2. Delete
3. Simplify
4. Accelerate
5. Automate

For instance, if a component costs \$1,000 to produce but the raw aluminum in it is only worth \$100, it would have a high idiot index.

If one takes notice of a component and gathers the information necessary to calculate the index, one is now questioning the component whether one realizes that or not.

If the index is high, and one is even vaguely familiar with the algorithm, then one will immediately ask oneself "Yikes, do we even need this thing? Can we just delete it?"

The index is likely high because the conversion of raw materials to components has some complexity or requires an expensive process tool. One then asks oneself "Can I simplify the design or the manufacturing process to make this cheaper?

The Accelerate step is about finding ways to speed up your bureaucratic processes.

Finally, if the component can't be deleted or its manufacture simplified, one asks oneself "Can its manufacture at least be automated thereby reducing labor costs or moving that labor to other areas where needed?"

---

There is an Everyday Astronaut two-part interview of Musk where they walk around the "shipyard" and Musk discusses aspects of the algorithm, and gives a few examples. I can't find it right now but one example is the system that deploys/locks then retracts (folds down) the grid fins. I think that it was an example of 1 → 2. At that point, it was decided that it wasn't the end of the world if the ship just launched with the grid fins sticking out.

I'll try to look for that later today.