A
Transcript

Will Stone Replace Steel and Concrete? - by Brian Potter

A McDonald's cheeseburger costs $2.59 despite a global supply chain with thousands of process steps — the DoorDash delivery alone is more expensive than every step combined. Brian Potter argues stone

Brian Potter · construction-physics.com

Gist

1.

A McDonald's cheeseburger costs $2.59 despite a global supply chain with thousands of process steps — the DoorDash delivery alone is more expensive than every step combined. Brian Potter argues stone construction's simpler supply chain is a real advantage, but one that economies of scale have already made irrelevant, and that no amount of automation can fix.

Logic

2.

Stone's supply chain is genuinely simpler — and that matters

  • Concrete requires crushing rock, heating limestone in a kiln to make cement, then mixing with water, sand, and admixtures; steel requires beneficiation, blast furnaces, basic oxygen furnaces, and rolling mills
  • Stone is quarried, carved, and shipped — no heating, no melting, no chemical reactions, no multistage processing
  • Energy costs reflect this: 20–30% of concrete's cost and 20–40% of steel's goes to energy; stone has no heating steps and consumes far less

3.

But economies of scale make extra steps nearly free at volume

  • Semiconductors require more process steps and more expensive fabs every generation, yet unit costs fell for decades — shrinking dies and larger wafers offset added complexity
  • A T-shirt shipped from Bangladesh to the U.S. costs a few pennies in transport; the global supply chain for a $2.59 cheeseburger is cheaper than a single DoorDash trip of a few miles
  • Steel and concrete use continuous process machinery and bulk methods like blasting — the "extra" steps are amortized across enormous volumes, making their unit cost almost insensitive to complexity

4.

OSB vs. dimensional lumber proves the point with identical automation

  • Dimensional lumber is cut directly from tree trunks; OSB is chopped into tiny pieces, glued with resins, extruded into sheets — a far more complex process
  • Both use highly automated manufacturing, eliminating the "stone just hasn't been automated yet" objection
  • On a board-foot basis, OSB is cheaper than almost every type of dimensional lumber, sometimes dramatically — complexity wins when volume is high enough

5.

Stone's mechanical properties are a statistical minefield

  • Steel, concrete, and engineered lumber average out raw-material variation across many small pieces; natural stone retains every flaw and impurity
  • Determining structural stone's safe capacity requires testing many individual samples and performing complex statistical analysis — one IStructE article spends a third of its pages on which Weibull distribution to use
  • Design strength (what you can actually rely on) is far below theoretical maximum, reducing or eliminating stone's theoretical strength advantage over concrete

6.

Stone can't flow — and that's a structural death sentence

  • Concrete pours into forms and cures; steel is hot-rolled or cold-formed into efficient shapes with high moment of inertia — thin, continuous slabs are standard in high-rise construction
  • Stone cannot be handled in a flowable state; making efficient shapes requires extensive cutting, and installing connectors or post-tensioning cables requires drilling very long, straight holes
  • Standard thin slabs are impossible — Webb Yates proposes corner-supported stone squares, a geometry that trades structural efficiency for material purity

7.

Code, supply chain, and workforce adoption take decades — stone has none of them

  • Mass timber advocates have pushed for over a decade in the U.S., had European and Canadian movements to build off of, and it remains extremely niche — stone has nowhere near that advocacy
  • A novel building system needs code pathways, officials willing to sign off, a robust supply chain, designers who know how to use it, and workers trained to install it — stone has none of these
  • Even if stone were cheaper at maturity, the path to get there would be long and burdensome, and right now the will to push it forward doesn't exist

Counter-Argument

8.

The entire cost comparison is based on a material that doesn't exist yet

  • Potter's central claim — that stone's simpler supply chain is overwhelmed by economies of scale — compares today's automated concrete and steel with today's unautomated stone, then extrapolates from OSB and semiconductors to argue automation won't close the gap
  • But the pro-stone case is explicitly about future automated stone, and Potter concedes "finishing stone can be automated (it obviously can), or whether that will reduce costs (it obviously will)" — the only question is scale effects, and nobody has run a high-volume automated stone operation to test them
  • The OSB comparison is Potter's strongest evidence, but he buries a critical caveat in a parenthetical: his own 2024 estimator shows OSB coming out more expensive than dimensional lumber — the one data point that directly tests the "complexity wins at scale" thesis actually contradicts it

Steelman

9.

The real question isn't which material is cheaper — it's which material can be built with fewer people

  • Both Potter and Springut frame this as a cost competition, but construction's binding constraint isn't material cost — it's labor availability; the U.S. faces a chronic shortage of skilled trades, and every material that reduces on-site labor has an advantage that raw cost comparisons miss
  • Stone's simpler supply chain and elimination of cladding, drywall, and painting steps don't just lower theoretical cost — they reduce the number of skilled workers required on a jobsite, which is the single most expensive and scarce resource in construction
  • If automated stone can be installed by a smaller, less specialized crew, it doesn't need to beat concrete on a board-foot basis — it needs to beat the total project cost of a building that can't find enough masons, drywallers, and painters to finish on time

Original

Continue Reading