The Physical AI ROI question isn’t the cost of the robot - it’s the cost of the next update.

Physical AI in Production - Part I (Article 1 of 3 )

I’ve been getting into a lot of discussions on Physical AI deployments, and the discussion on ROI is almost always “the Capex is the bottleneck – the robots, the sensors, the compute…”

This has been the reasonable mental model of classic automation so far: We buy/install the thing, we validate/test it, run it for many years, cost is amortized, and savings are consistent over time.

But the more time I’ve been spending looking at Physical AI pilots and deployments across industries, be it automotive, manufacturing, or warehousing - Physical AI is not a one-off asset.

It is a system that keeps changing - and its economics are defined by the cost of changing it safely.

So I spent last weekend dusting off the cobwebs of my economics degree, and found 3 theories that resonated with me:

1. Good Old Transaction Cost Economics (The Nature of the Firm, Ronald Coase) which discusses the cost of coordinating and proving change – This theory is actually well written about in the context of AI and AI Agents (one I really enjoyed reading was this perspective on the impact of AI and Agentic AI on MROs - )

But there are a couple of other theories which I think we should also pay attention to:

2. O-Ring Theory (Kremer) - Butterfly effect of the weakest link, or one failure mode can wipe out lots of gains

3. The Power of Modularity (Baldwin & Clark) – can we structure the system so change doesn’t require proving everything again and again

So I’m spending some time on how these work with Physical AI deployments:

There’s a common pattern across industries -

the Steady state is fine , The messy days are expensive. Messy days happen more often than we want them to.

For example, warehouses have peak and congestion spirals…manufacturing cells will work great till a part tolerance shifts, and then line stops multiply…a driver assist system is great till a rare edge case triggers a really expensive intervention loop… and so on.

Bottom line - Physical AI has a great value proposition, because it targets the “messy middle” - fewer interventions, fewer stops, faster recovery, less babysitting the system and more consistency in changeable conditions -

However, I think a lot of us are using the wrong ROI framework to evaluate it - because we treat it like classical automation (capex + integration = stable savings).

Physical AI deployments want repeated updates – because the world changes, the system learns new patterns, we discover new edge cases, policies get updated, roll-outs happen in stages – and now we see Coase smile as Transaction Costs come into play

Coase’s point is that the actual cost isn’t just producing output - it’s the cost of coordinating, monitoring, and enforcing how things are done. In Physical AI, every update carries overhead. We need evidence that it improves something, proof that it hasn't broken something else, a controlled rollout, we need to monitor it once it hits the real world, and we need a way back if it goes wrong. Unlike a software feature update, some of those failures have physical consequences.

So we need a new line item in the business case – the cost of safe change (Marginal Cost of Change x per update x times the number of updates you expect)

Now let’s talk about the O-Ring theory – which says that value isn’t always compounding in a system – it can be unstable because an operation is tightly coupled and one rare failure mode can wipe out a lot of gains seen from implementing the system

If we only measure averages (“throughput rose 3%”), we miss the edge economics (“one terrible day wiped it out”).

Take automotive fleets and ADAS systems: We can have ROI conversations based on sensors and compute needs, but the actual day-to-day cost is in interventions on bad days (remote support, investigations, customer handling, safety driver actions)

Napkin calculation - If we have 10,000 vehicles, each driving 15,000 miles a year – we’re looking at 150M miles annually. If intervention needs are about 1, or even 0.5/1000 miles – that’s about 75,000 interventions a year.

Now we try to quantify this – £50 per intervention – that’s about £3.75M a year. A 10% decrease is about $375K/year – decent savings.

Now let’s O-Ring this – we do an update, and it increases a rare but high-severity event category – we can lose that £375K quickly – and it’s not just direct cost; there are costs to rollbacks, investigations, remediation...

So the economics is actually about

what it costs to prove an update is safe, how well can we detect problems and how quickly we can roll back once we see them.

Which leads to theory three - Modularity.

Take a robotic cell which does machine kitting (with human intervention when it gets confused). The real cost of this autonomous operation isn’t the robot's average speed – it’s the stops and recoveries.

If this cell loses 30 minutes a day because of minor stoppages – at about £200/minute of downtime, that’s £6000 a day or £150K a year

If we reduce this by 15% by putting guardrails for autonomy - that’s about 4.5 minutes a day at £200 = £900 a day, or about £225K of savings (about 250 working days) a year!

Sounds awesome - but the catch is that if each improvement (or expansion to a new site) needs a week of bespoke revalidation because we can’t figure out what changed and reprove it, Marginal Validation Cost will eat up that £225K

And here is where we need to work with modularity economics -

A system gets cheaper to improve when you can change one part without having to re-prove the entire system.

In the manufacturing example, this will mean defining and maintaining clear safety controls, defining what stays stable vs. what can be changed, and automatically logging the right evidence to replay and compare. Without this, every update is basically a mini re-commissioning project, and we don’t drop the marginal cost of validation over time.

Another example of the O-Ring Theory is in warehousing where there is a tightly coupled flow - a jam in one zone can create knock-on delays, a bad dispatch instruction can starve the belt, a bad decision can cause gridlocks…

Napkin Calculation - If we price this as 30 interventions needed per day at a warehouse site, each about 7 minutes – that’s 210 minutes or about 3.5 hours a day or about £39,375 direct intervention cost (at £45 an hour for 250 working days).

If this warehouse also has 10 major gridlock events a year, each causing a couple of hours of disruption (say £ 2,000/hour in overtime, missed deadlines, rescheduling) = 10 x 2 x 2,000 - £40,000 a year.

We put in place a change to reduce interventions by 20% - that’s a saving of 20% x £ 39,375 = £ 7,875 of savings

But this change increases gridlocks by 2/year – that is 2x2x£2000 = £8000/year, wiping out the gains – so a change that can improve an average can still be negative overall.

So is it not worth it?

Physical AI benefits are absolutely real (and I’m still on this thought train so stay with me a few more days) – we need to call out the right ROI model –

weak-links decide outcomes + repeatability is the driver of the business case. Make safe change cheaper over time, and stop changes from touching everything – this way we reduce the surprise spend category

The business case for Physical AI is basically - “Do we have a credible plan to make the next update safer and cheaper than the last?”

-

Pallavi Chari, Moving Parts | (Linkedin)

---

_{Next Read in the Series: Physical AI in Production - Part II (Article 2 of 3 )}