Innovation Prototype Development Simulation

Fabrication Method

Target

Spec tolerance 1.0 mm

In progress

Prototype State

Iteration

Dim. error

10.00 mm

Cumulative cost

Cumulative time

0.0 hrs

How it works

Real product teams converge on a working design through a design → build → test → learn loop, not by getting it right the first time. Each cycle here removes a fraction of the remaining dimensional error, following an exponential learning curve:

error(n+1) = error(n) · (1 − α) + noise(n)
noise(n) ~ Uniform(−β·error(n), +β·error(n))

α is the method's learning rate — how much each test round teaches you — and β is how often a build introduces a fresh surprise (warping, tolerance stack-up, a missed fillet):

3D Print (FDM): α ≈ 0.22, β ≈ 0.15 — cheap, fast turnaround, noisier convergence.
CNC Machining: α ≈ 0.35, β ≈ 0.08 — expensive per cycle, but each cut is close to the CAD file, so error drops fast and cleanly.
Hand Fabrication: α ≈ 0.12, β ≈ 0.25 — cheapest per cycle, but slow to converge and prone to setbacks.

The prototype is "done" when error(n) ≤ your spec tolerance. The red spikes on the part are a stand-in for out-of-spec surface deviations — they shrink and thin out as the loop converges, exactly as scrap and rework drop off a real prototyping program (this is the same logic behind the Wright brothers' wind-tunnel-then-glider iteration and every modern rapid-prototyping shop's build log).

Prototype body Out-of-spec deviation

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.