CostKit

Manufacturing Process Selection Guide

The same part can cost three times more or less depending on the process, and the deciding factor is usually not unit price but whether the tooling can be amortised. This page puts six mainstream processes side by side, each linked to a calculator you can run yourself.

Quick comparison

ProcessCost driverTooling investmentEconomic batchTypical tolerance
Injection molding Resin + machine time + tooling CNY 30k–500k tool 3,000+ parts ±0.05–0.2 mm
Die casting Alloy + cycle + tooling CNY 80k–800k die 5,000+ parts ±0.1–0.5 mm
Stamping Coil + strokes + tooling CNY 10k–300k die 10,000+ parts ±0.05–0.15 mm
Sheet metal Sheet + cut time + bends Almost none (universal tools) 1–5,000 parts ±0.1–0.3 mm
CNC machining Stock + machine time + tools Fixture CNY 200–2,000 1–2,000 parts ±0.01–0.05 mm
3D printing Filament + print time + post None 1–200 parts ±0.2–0.5 mm

Tolerances and batch sizes are common industry ranges, not promises. The real boundary is part geometry: a 0.3 mm wall in a deep pocket is routine for molding and painful for CNC, while a ±0.01 mm mating face rules molding out entirely.

Three-step decision

Step one: material

Plastic parts mean molding, blow molding or 3D printing. Metal parts open die casting, stamping, sheet metal and machining. Once the material is fixed you are down to two or three candidates. Composites and powder metallurgy sit outside this page.

Step two: volume

Volume decides whether tooling amortises. Under 1,000 parts, CNC and sheet metal almost always win because there is no mold to pay for. Between 1,000 and 10,000 is the transition band where the arithmetic matters. Above 10,000, molding and stamping unit costs dominate.

Step three: tolerance and cosmetics

Inside ±0.05 mm, or with mirror finishes and complex surfaces, CNC and post-processed 3D printing are more reliable. Cosmetic parts that need texture and batch-to-batch colour consistency suit molding and casting. Tightening a tolerance one grade raises both cycle time and reject rate.

Tooling amortisation is the watershed

Tooling is a fixed cost; amortisation = tool cost ÷ life in shots. The common error is treating life as volume: a mold rated for 500,000 shots does not mean your project ships 500,000 parts. Amortise over the actual project volume, then add the process unit cost.

unit cost  = process cost + tool cost ÷ project volume (capped by tool life)
break-even = tool cost ÷ (alternative unit cost − this unit cost)

Example: a molded part at 2.10 CNY with a 200,000 CNY tool, versus the same part machined at 45 CNY with no tool. Break-even needs about 200,000 ÷ (45 − 2.10) ≈ 4,660 parts — before DFM iterations and trial runs. Below that, machining is cheaper; above it, the tool starts paying back.

Hidden cost by process

Common misjudgements

First, comparing prices at different volumes: a 1,000-piece CNC quote and a 50,000-piece molding quote do not belong in the same column. Second, ignoring yield: at 85% first-pass yield the real molded unit cost is 18% above the theoretical one. Third, ignoring finishing: castings without deburring and sheet metal without coating are not deliverable states. Fourth, ignoring time: 25–45 days for a tool versus 3–7 days for machining can outweigh the price difference on a launch schedule.

Turning selection into arithmetic

Selection is fixed cost ÷ volume plus variable cost. Compute the variable cost for each candidate, then cross-check with the break-even formula above: small volumes point to machining and sheet metal, mid volumes to stamping and casting, large volumes to molding. The only genuinely hard call is the few thousand parts in the transition band.

Every calculator states its error range and assumptions — use the conservative end. If two processes stay close under conservative assumptions, cost is not the deciding variable and you should choose on lead time, quality and supply chain instead.

See also: glossary · mold cost estimator · assembly cost · all tools