NPI, DFM/DFA & Engineering Change calculator

Pre-Production Capacity Calculator

Pre-production capacity is the realistic good-unit output a new line or process can deliver during the NPI build phase, before it has reached steady-state efficiency. NPI engineers, DFM/DFA teams, and ramp managers use it to size pilot and pre-production runs, because a brand-new process rarely hits nameplate throughput — early uptime is lower and first-pass yield is still climbing. This calculator multiplies output per cycle by available cycles to get gross capacity, then discounts it by expected uptime and first-pass yield to land on good units, and it breaks out exactly how many units are lost to downtime versus quality. That split is what tells you whether to fix the machine or fix the process.

What this calculator does

  • Estimate pre-production capacity for npi, dfm/dfa and engineering change using production-ready inputs so teams can confirm whether capacity can cover demand before committing the schedule.
  • Use it when pre-production capacity in npi, dfm/dfa and engineering change is being asked to take on more work and you need to know if there is room.
  • It computes good pre-production output by multiplying gross capacity (output per cycle times available cycles) by expected uptime and first-pass yield, and isolates downtime loss and yield loss.

Formula used

  • Gross pre-production capacity = pre-production capacity output per cycle × available pre-production capacity cycles
  • Good pre-production capacity = gross capacity × expected pre-production capacity uptime × expected pre-production capacity first-pass yield

Inputs explained

  • Pre-production capacity output per cycle: Use the good units, parts, cavities, assemblies, tests, or batches completed each cycle.
  • Available pre-production capacity cycles: Enter the planned cycles from the shift schedule, takt plan, asset plan, or run calendar.
  • Expected pre-production capacity uptime: Use recent uptime or availability from production reports, maintenance logs, or OEE data.
  • Expected pre-production capacity first-pass yield: Use first-pass yield from inspection, test, quality, or production records for the same scope.

How to use the result

  • Use it when planning a pilot or pre-production build during NPI, before the process has matured to steady-state rates.
  • Uptime and first-pass yield during ramp are estimates that improve as the process stabilizes, so early runs of the calculator should use conservative, learning-curve-adjusted figures.

Common questions

  • How do you calculate pre-production capacity? Multiply output per cycle by available cycles for gross capacity, then multiply by expected uptime and first-pass yield. Here 4 units/cycle times 480 cycles is 1,920 gross, and after 90% uptime and 97% yield you get 1,676.16 good units.
  • Why is pre-production capacity lower than rated capacity? A new process has not climbed the learning curve. Uptime suffers from teething faults and yield is still being dialed in. In the example, 90% uptime costs 192 units and 97% yield costs another 51.84, so gross 1,920 falls to 1,676.16 good.
  • What is the difference between downtime loss and yield loss? Downtime loss is capacity you never produced because the line was stopped (192 units here). Yield loss is units produced but scrapped or reworked for quality (51.84 units). Downtime points at equipment and availability; yield points at the process itself.
  • What is a good first-pass yield during pre-production? Early NPI yields are often well below mature targets and improve build over build. The 97% used here is healthy for late pre-production; pilot runs may start far lower, which is why you plan extra builds to hit good-unit goals.
  • How is this different from steady-state capacity planning? Steady-state planning assumes a stable, learned process with high uptime and yield. Pre-production capacity deliberately uses lower, ramp-phase uptime and first-pass yield because the line is still being debugged, so it forecasts realistic early output rather than nameplate.

Last reviewed 2026-05-12.